You are on page 1of 500

FYI

From: Miller, Tony


To: Rogers, Margot
Sent: Tue Apr 27 10:03:02 2010
Subject: Fw: More on the Department of Education's Gainful employment )_(_
5
_ ) __ _,
Fyi
Sent using Black:Beny
From: Gordon, Robert M.
To: Miller, Tony; Shireman, Bob
Cc: Sunstei n, Cass R.
Sent: Tue Apr 27 09:59:26 2010
Subject: FW: More on the Department of Education's Gainful employment reg-,_l (b-)(-S) ____ _,
Tony and Bob,
I am sure you agree that this is astonishing and unacceptable. Has the Department' s leadership signaled to all that this is a
grave matter, and are you taking steps to figure out who is responsible?
Thanks,
Robert
APRa 26, 2010, 10:57 A.M. ET
Debt, Job Rule Uncertainty Hits Shares OfF or-Profit Colleges
NEW YORK (Dow Jones)-A government proposal to hold colleges accountable for graduating students with high debt
loads and low income levels came front and center, again, after Credit Suisse analysts reversed their assumption that the
Department ofEducation had softened its stance on the subject.
Monday morning, Credit Suisse issued a note saying the Department ofEducation has returned to a stiffer set of rules on
post-graduation debt and gainful employment levels that colleges must meet. The news hit share prices of for-profit
universities that had rallied on expected softer rules.
Credit Suisse downgraded IIT Educational Services Inc. (ESI) and DeVty Inc. (DV) to neutral from outperform and
slashed their price targets, citing the potential fallout from their new take on gainful employment as well as concerns of
countercyclicality. ITT recently was off2.9% to $108.53, while DeVry was down 6.5% to $64.96. Credit Suisse had
upgraded the schools two weeks ago.
Other for-profit schools, including Strayer Education Inc. (STRA), Career Education Corp. (CECO), Corinthian
Colleges Inc. (COCO) and Apollo Group Inc. (APOL), were also trading down.
The Department of Education originally said schools could face scrutiny that could put federal assistance--often the
primary revenue stream for for-profit colleges--at risk if they did not have at least a 70% graduation rate and 70% job
placement in field of study. Two weeks ago, Credit Suisse, and others, sent out notes saying they believe the Department
ofEducation had softened the graduation level to 50%, sending shares higher, with some hitting 52-week highs.
"Oddly, we suspect the big upward move the stocks had when the investment community found out about the 500/o
exemption may have led to pressure on the DOE to remove the 50% exemption," the firm wrote Monday.
Some education insiders have had mixed feelings on predicting the gainful-employment measure, arguing that it's
improper to say what the Department of Education will put forth and it's best to wait until the official proposal is released
for public comment. That will happen by mid-June.
"I'm not sure who thinks they have what access to what's in this draft," said one for-profit school official, but "I would
really not be jumping to any conclusions just yet." He said there's no way of knowing "whether it's been hardened,
softened, pureed, whatever."
Representatives from the Department of Education weren't immediately available for comment.
http://online.wsj .com/article/BT-C0-201 00426-71 0339.html?mod=rss _Hot_ Stocks
Sharon Mar
Policy Analyst I Information Policy
OMB I Office oflnfmmation and Regulatory Affairs
Tel : 202.395.64661 Fax: 202.395.51671 smar@omb.eop.gov
From: Finley Steve
To: Yum Georgia
CC:
Date: 5/4/2010 12:01: 10 PM
Subject: FW: New Bloomberg article today: Obama Plans New Rules as For-Profit Colleges Mobilize
(b)(5)
" The speech was "highly negative" and was "drawing inappropriate and unwarranted parallels between developments in
higher education and the causes of the recent fmancial crisis," Harris Miller, president of the Career College Association
wrote in an April 29 letter to Duncan."
-----Original Message-----
From: Woodward, Jennifer
Sent: Tuesday, May 04, 2010 11:55 AM
To: Yuan, Georgia; Marinucci, Fred; Siegel, Brian; Wolff, Russell; Finley, Steve; Sann, Ronald; Wanner, Sarah
Subject: New Bloomberg article today: Obama Plans New Rules as For-Profit Colleges Mobilize
(b)(5)
"[T]he new regulations would shut 300,000 students out of classes and eliminate 2,000 educational programs, according
to a study commissioned by the Washington-based Career College Association, which represents more than 1,400 for-
profit
cotleges. The proposal would reduce opportunities for women and racial minorities who want to go to college, the group
said.
*********************************************************************
Obama Plans New Rules as For-Profit Colleges Mobilize for Fight
2010-05-04 04:02:00.0 GMT
By John Hechinger, Daniel Golden and John Lauennan
May 4 (Bloomberg)-- The Obama Administration is gearing up
to produce tougher regulations that may reduce the amount of
federal financial aid flowing to for-profit colleges, cutting
the companies' annual revenue growth by as much as a third.
In response, the $29 billion industry and its supporters
including Republican Senators have enlisted top Washington
lobbyists and are courting black and Hispanic legislators to
fight the proposed rules scheduled to be released as early as
this month. The companies draw students from low-income and
minority communities.
Federal aid to for-profit colleges has become an issue as it
has jumped to $26.5 billion in 2009 from $4.6 billion in 2000,
according to the Education Department, prompting concern that
these students are taking on too much debt. Twelve higher-
education stocks fell an average of7.4 percent for the week
ended April 30, according to Bloomberg data, following an April
28 speech by an Education Department official critical of for-
profit colleges. In the same period, the Standard & Poor's 500
Index dropped 1.7 percent.
"There' s an attempt to manage" for-profit colleges by the
Obama administration, Robert Wetenhall, an analyst with BMO
Capital Markets in New York, said in a telephone interview. The
education companies' influence in Washington has "radically
changed," from the years of the Bush administration, he said.
The tougher rules, which are expected to be released for
public comment in the next several weeks, would require ITT
Educational Services Inc. , Career Education Corp. and Apollo
Group Inc.' s University of Phoenix to show that their graduates
earn enough money to pay off their student loans. If for-profit
colleges can' t meet the standard, they could lose federal
financial aid, which typically makes up three-quarters of their
revenue.
Tuition Increases
The proposed rules may disqualify for-profits from
receiving federal financial aid if their graduates must spend
more than 8 percent of their starting salaries on repaying
student loans. The regulations may slow or even halt tuition
increases at ITT, Education Management Corp., Lincoln
Educational Services, Universal Technical Institute, and Career
Education because many graduates take low-paying jobs in
criminal justice, cooking and medical office work, Trace Urdan,
an analyst at Signal Hill Capital Group in San Francisco, said
in an intetview.
Education companies have increased revenue by as much as 15
percent and enrollment by 8 to 10 percent on an annual basis,
while raising tuition about 4 to 6 percent a year, Urdan said.
The new rules may slow their revenue growth by one third by
limiting their ability to raise tuition.
Pricing Power
"The days of 4 to 6 percent annual tuition price increases
are over," Urdan said. "The new proposed rules will bring some
school's power to increase prices down to zero."
Apollo closed at $58.32, up 91 cents, or 1.6 percent, in
New York Stock Exchange composite trading yesterday. ITT closed
at $104.22, up $3.09, or 3.1 percent. Career Education rose $97
cents or 3.3 percent to $30.24.
The Education Department plans to issue the regulations
without Congressional approval, unlike student-loan legislation
which passed in March.
"Congress has not held a single hearing on these new
enforcement mechanisms," Alexa Marrero, spokeswoman for John
Kline, the ranking Republican on the House education committee,
said in an e-mail. "No research has been offered by the
department to justify the controversial proposal."
U.S. Senator Lamar AJexander, who chairs the Senate
Republican Conference, is trying to persuade U.S. Education
Secretary Arne Duncan to reconsider the reguJations, said a
Republican aide on the education committee. If that doesn' t work,
AJexander, who is on the education and appropriation committees,
would try to kill the regulations by cutting off funding, the
aide said.
Enrollment in for-profit colleges increased to 1.8 million
in 2008 from 673,000 in 2000. Industry revenue will rise to
$29.2 billion this year from $9 billion in 2000, said Jeffrey
Silber, an analyst for BMO Capital Markets in New York. The
industry has grown in part by marketing to low-income students,
including the homeless, who qualifY for federal grants and loans.
Regulations' ln1pact
The new regulations would shut 300,000 students out of
classes and eliminate 2,000 educational programs, according to a
study commissioned by the Washington-based Career College
Association, which represents more than 1,400 for-profit
colleges.
The proposal would reduce opportunities for women and
racial minorities who want to go to college, the group said.
For-profit colleges have proposed alternative regulations that
would require companies to disclose more information about
students' debt and job prospects.
The Career College Association has retained the Podesta
Group, a Washington lobbying finn headed by Anthony Podesta,
whose brother, John, was President Bill Clinton's chief of staff,
according to federal filings. Clinton will be a keynote speaker
at the association' s annual meeting in June. Podesta' s Paul
Brathwaite, fanner executive director of the Congressional Black
Caucus, is also lobbying on the association' s behalf, records
show.
Phoenix Scholarships
The University ofPhoenix, the largest for-profit college
in the U.S. by enrollment, awarded 25 full-tuition scholarships
worth $1.25 million in the fiscal year ended August 31 to the
Congressional Black Caucus Foundation, which selects the
recipients, Apollo spokeswoman Sara Jones said in an e-mail.
More minority students earn degrees from Phoenix than from any
other U.S. university, she said.
In March, several members of the Congressional Black Caucus
sent a letter to Duncan, saying the regulations would reduce
educational opportunity.
Regulators need more tools to oversee publicly-traded
education companies receiving increasing amounts of federal
money, Robert Shireman, deputy undersecretary of the education
department, said in a speech on April28.
"I don' t thi nk we have the fi repower that we need," he
said, according to a transcript of hi s remarks.
The speech was "highly negative" and was "drawing
inappropriate and unwarranted parallels between developments in
higher education and the causes of the recent financial
crisis," Harris Miller, president of the Career College
Association wrote in an Apri l 29 lettertoDuncan.
For Related News and Information:
Stories about education: NI EDU
U.S. colleges and universities: USUY
Education organizations: EDOR
Stori es about the Department of Education:
NIEDN
--Editors: Robin D. Schatz,Jonathan Kaufman
To contact the reporters on this story:
John Hechinger in Boston at + 1-617-210-4614 or
jhechinger@bloomberg.net;
Daniel Golden in Boston at +1-617-210-4610 or
dlgolden@bloomberg.net;
John Lauerman in Boston at + 1-617-210-4630 or
jlauerman@bloomberg.net.
To contact the editor responsible for this story:
Jonathan Kaufinan at +J-617-210-4638 or Jkaufman17@bloomberg.net.
(b)(5)
From: Finley, Steve
From: Finley Steve
To: Chesley, Susan
CC:
Date: 9/16/2010 10:53:32 AM
Subject: FW: new report on GE
Sent: Thursday, September 16, 2010 10:40 AM
To: Yuan, Georgia; K vaal , James; Bergeron, David; Kol otos, John; Sellers, Fred
Cc: Sann, Ronald
Subject: new report on GE
The report described below is attached.
Here is the link:
http://www. educationsector.org/sites/ default/files/publications/Gainful-Report_ RELEASE. pdf
Are You Gainfully Employed? Setting Standards for For-Profit Degrees
Education policy events in Washington, D.C., attract a familiar cast of characters: think tank representatives, members of
organizations with eight-letter acronyms, and the occasional retired college professor. But an event at the New America
Foundation this summer attracted a different crowd: organizations with names that ended in "markets," "capital," and
"fund." The questions from the audience weren' t about quality teaching or common curriculum standards; they were
about income vs. earnings, access to federal databases, and student debt thresholds. Someone hearing just the audio
feed might have easily mistaken the conference for the quarterly earnings call of a Fortune 500 corporation.
The subject of the event was for-profit higher education. Representatives from the financial sector had come to New
America, a nonpartisan public policy institute, to hear a high-level Obama administration official talk about a regulatory
controversy that could make them-or lose them-hundreds of millions of dollars. Just a few days before, the U.S.
Department of Education had released a new proposal that would make it more difficult for for-profits to access billions
of dollars in federal funds. At the center of the proposal is a rule called "gainful employment" that would penalize for-
profit colleges and other vocational training programs for saddling students with more debt than they can pay back.
For-profits have grown by leaps and bounds in recent years, largely free of federal regulation. That freedom would be
significantly curtailed if the gainful employment standard takes effect. Vocational training programs would be judged by
the ratio of the debt that graduates assume relative to their current earnings and the rate at which they are able to repay
it. If programs offered by for-profit colleges exceed certain thresholds on those measures, they risk losing eligibility for
federal student aid. Given that many for-profit colleges receive close to 90 percent of their revenue from federal grants
and loans, losing access to these dollars would be a death sentence.
With such high stakes, the proposed gainful employment standard has generated intense debate. While the owners of
for-profits see a $29 billion industry that produces some of the best earnings ratios in the stock market, a group of well-
funded short-sellers paints a picture of a fraudulent, over-leveraged industry that' s poised for a subprime mortgage-style
collapse. The institutions argue that they serve a class of students excluded from traditional higher education and that they
are crucial for meeting the Obama administration' s college completion goals. But many lawmakers worry that in fulfilling
that mission, for-profits have relied too heavily on federal aid, forced students to borrow too much money, and produced
degrees of questionable worth. Sen. Tom Harkin, the Iowa Democrat who chairs the Senate committee overseeing these
schools, has warned that "even good actors in this industry are lwed into the vortex of bad practices in order to compete
and meet investors' expectations."
Critics of the gainful employment standard, meanwhile, have claimed the proposal "will eliminate quality programs while
doing little or nothing to address the issue of excessive student debt." Some have even gone so far as to say it "will
attack our freedom and individual liberty to make decisions that have consequences."
Yet despite all the noise and controversy, important questions have been left unanswered: Which institutions are most
vulnerable to the proposed rules? What types of programs are most likely to be affected? This report tries to answer
these questions, using publicly available data to present, for the first time, a picture of what effect the gainful employment
proposal could have at more than 12,600 vocational programs at colleges and universities across the country. This
includes more than 2,350 bachelor' s degree programs ....
EDUCATI ONSECTOR REPORTS
September 201 0
ARE YOU GAINFULLY EMPLOYED?
Setting Standards for For-Profit Degrees
By Ben Miller
ABOUT THE AUTHOR
BEN MILLER is a policy analyst at Education Sector. He can be
reached at bmiller@educationsector.org
ABOUT EDUCATION SECTOR
Education Sector is an independent think tank that challenges
conventional thinking in education policy. We are a nonprofit,
nonpartisan organization committed to achieving measurable
impact in education, both by improving existing reform initiatives
and by developing new, innovative solutions to our nation's most
pressing education problems.
Copyright 2010 Education Sector
Education Sector encourages the free use, reproduction, and distribution
of our ideas, perspectives, and analyses. Our Creative Commons licens-
ing allows for the noncommercial use of all Education Sector authored
or commissioned materials. We require attribution for all use. For more
information and instructions on the commercial use of our materials,
please visit our website, www.educationsector.org.
1201 Connecticut Ave., N.W, Suite 850, Washington, D.C. 20036
202.552.2840 www.educationsector.org
Education policy events in Washington, D.C., attract a familiar cast
of characters: think tank representatives, members of organizations
with eight-letter acronyms, and the occasional retired college
professor. But an event at the New America Foundation this summer
attracted a different crowd: organizations with names that ended in
"markets," "capital," and "fund. " The questions from the audience
weren't about quality teaching or common curriculum standards;
they were about income vs. earnings, access to federal databases,
and student debt thresholds. Someone hearing just the audio feed
might have easily mistaken the conference for the quarterly earnings
call of a Fortune 500 corporation.
The subject of the event was for-profit higher
education. Representatives from the financial
sector had come to New America, a nonpartisan
public policy instit ute, to hear a high-level Obama
administration official talk about a regulatory
cont roversy that could make them- or lose t hem-
hundreds of millions of dollars. Just a few days
before, the U.S. Department of Education had
released a new proposal t hat would make it more
difficult for for-profits to access bil lions of dollars in
federal funds. At t he center of the proposal is a rule
called "gainful employment " that would penalize for-
profit colleges and other vocational training programs
for saddling students with more debt than they can
pay back.
For-profits have grown by leaps and bounds in
recent years, largely free of federal regulation. That
freedom would be significantly curtailed if the gainful
employment st andard takes effect. Vocational
t raining programs would be judged by t he ratio of the
debt t hat graduates assume rel at ive t o their current
earnings and t he rate at which t hey are able to repay
it. If programs offered by for-profit colleges exceed
certain thresholds on t hose measures, t hey risk losing
eligibility for federal st udent aid. Given t hat many
for-profit colleges receive close to 90 percent of their
revenue from federal grants and loans, losing access
to these dollars would be a death sentence.
www.educationsector.org
With such high stakes, the proposed gainful
employment standard has generated intense debate.
While the owners of for-profits see a $29 billion
industry that produces some of t he best earnings
rat ios in the st ock market , a group of well -f unded
short-sellers paints a picture of a f raudulent , over-
leveraged industry that's poised f or a subprime
mortgage-style collapse. The institutions argue
that they serve a class of students excluded from
t raditional higher education and t hat they are crucial
for meeting t he Obama administration's college
complet ion goals. But many lawmakers worry t hat in
fulfilling that mission, for-profits have relied too heavily
on federal aid, forced st udents t o borrow t oo much
money, and produced degrees of questionable worth.
Sen. Tom Harkin, t he Iowa Democrat who chairs
the Senate committee overseeing t hese schools,
has warned t hat "even good actors in t his industry
are lured int o t he vortex of bad practices in order t o
compete and meet investors' expect ations."
1
Critics of t he gainful employment standard,
meanwhile, have claimed t he proposal "will eliminate
quality programs while doing little or not hing to
address the issue of excessive st udent debt. "
2
Some
have even gone so far as t o say it "will attack our
freedom and individual liberty to make decisions t hat
have consequences."
3
EDUCATION SECTOR REPORTS: Are You Gainfully Employed? 1
Yet despite all the noise and controversy, important
questions have been left unanswered: Which
institutions are most vulnerable to the proposed
rules? What types of programs are most likely to be
affected? This report tries to answer these questions,
using publicly available data to present, for t he first
time, a picture of what effect the gainful employment
proposal could have at more than 12,600 vocational
programs at colleges and universities across the
country. This includes more than 2,350 bachelor's
degree programs.
Much of the focus in Congress and in the media
has been on institutions, particularly those that
are publicly traded. But this analysis suggests that
individual programs within those institutions may
vary widely in how they perform under the proposed
gainf ul employment standard. An institution could
very well offer both programs that are unaffected and
programs that become ineligible for federal student
Colleges are forbidden by law
to make false promises of jobs
or to inflate salary d a ~ a , sc hey
play on emotions, appealing to
students' desires to be valued in
their careers.
aid. This analysis also finds that t he type of programs
that could lose el igibility under the gainful employment
standard vary significantly. For instance, there are
a large number of ineligible programs for medical
assistants, but these programs only exceed the
proposed debt-to-income standard by a few thousand
dollars, meaning they could avoid penalties if they
slightly reduced their costs. Others programs, like
those in culinary arts, are less likely to be ineligible,
but those that miss the mark often miss by a wide
margin.
Out of more than 12,600 programs, about 4 percent,
or just over 500 programs, would lose eligibility
because of the new standard. This includes 8 percent
of bachelor's degree programs, 6 percent of associate
degree programs, and 1 percent of programs that are
generally certificate programs of two years or less.
4
2 EDUCATION SECTOR REPORTS: Are You Gainfully Employed?
Overall, the programs most likely to be affected are
those tied to high-tech fields, such as e-commerce
or graphic design, or those tied to jobs with low
expected starting salaries, such as medical assistant
or chef.
Although these programs cover a range of different
jobs, they employ similar marketing tactics. Colleges
are forbidden by law to make false promises of jobs
or to inflate salary data, so they play on emotions,
appealing to students' desires to be valued in their
careers.
5
TESST College of Technology in Beltsville,
Md., a school owned by Kaplan Higher Education,
tells would-be medical assistants that they are
joining a "growing field t hat allows [them] to assist
others in need."
6
Students looking at Le Cordon
Bleu Institute of Culinary Arts, owned by Career
Education Corporation, are encouraged to "follow
[their] passion" and "explore [their] creativity. "
7
Kaplan
advertises its inf ormation technology degrees as a
chance for students to "be the most valued person at
work."
8
And at the Art Institute of Pittsburgh's Online
Division, the ads for the interior design program tell
prospective students that they can have a "profound
impact on people's lives."
9
These marketing pitches also often tout t he benefits
of an entire industry, like health care, rather than the
realities of the specific job for which the program
is preparing students- jobs that are entry-level
with low pay and typically have little opportunity for
advancement. And as this analysis shows, many of
these programs are poor investments for students.
While these at-risk programs comprise only a small
minority of all programs at for-profi t colleges, it
would be wrong to conclude that most for-profit
programs will emerge from the new federal standards
unscathed. A much larger number-some 65
percent-are likely to fall in a middle ground between
full eligibility and total ineligibility called "eligible with
a debt warning," which requires colleges to, among
other things, post prominent cigarette pack- style
"debt warnings" alerting potential students to the
likelihood that enrolling could be hazardous to their
financial health.
The gainful employment standard would not lead to
a wholesale shutdown of the for-profit sector. But it
would probably force many for-profits to substantially
change their pricing and approach to student debt.
www.educationsector.org
More broadly, it would establish a new federal
perspect ive on higher educat ion, involving close
examinat ion of college prices relat ive to graduat es'
future earnings, an idea that was first contemplated
decades ago but is only now seeing the light of day.
An Undefined Standard
Gainful employment is not a new standard. When
Congress passed t he Higher Education Act in 1965, it
required that non-accredited public or private not-
for-profit programs provide gainful employment in
a recognized occupation in order to receive federal
money for student aid.
10
When for-profit colleges later
became eligible for these aid programs, Congress
required them to meet t he same criteria.
But Congress never fully defined gainful employment
or explained how colleges could meet the standard.
Even when for-profits came under intense scrutiny
as part of a Congressional investigation into
widespread industry fraud in the early 1990s, the
term remained unclear. Instead of clarifying the
Categories of Eligibility
Eligible
These programs have a repayment rate of at least 45
percent, and the annual loan payment is less than or
equal t o 8 percent of the average annual earnings or 20
percent of discretionary income.
Ineligible
These programs have a repayment rate below 35
percent and an annual loan payment that is both above
12 percent of average annual earni ngs and 30 percent
of discretionary income.
Eligible with a Debt Warning
These programs have either a repayment rate or
debt-to-income ratio that exceeds the threshold for
an eligible program, but not both. In other words, the
repayment rate must be at or above 45 percent, or the
debt-to-income ratio must be at or below 8 percent or
20 percent, but both cannot occur.
Restricted
These programs have a repayment rate below 45
percent and a debt-to-income ratio below 8 percent
and 20 percent. They also have either a repayment ratio
at or above 35 percent or a debt- to-income ratio at or
below 12 percent or 30 percent, or both.
www.educationsector.org
requirement, Congress and federal agencies turned
their attention to closing schools, tracking the rate at
which borrowers defaulted on their loans, and limiting
the percentage of revenue for-profit colleges could
receive from federal aid programs.
Today, for-profits are again in the spotlight. Enrollment
at these schools grew by 160 percent in the last
decade.
11
And despite enrolling only about 10
percent of all students, for-profi ts consume 25
percent of all Pell Grant dollars disbursed and 21
percent of all federal student loan dollars.
12
That's a
large investment for a sector in which the average
graduation rate is 20 percent for bachelor's degrees
and just over 60 percent for programs of two years or
less, and in which it's estimated that as many as 40
percent of student loans go into default.
13
Those numbers have prompted the federal
government to take a fresh look at the sector. The
U.S. Senate has held multiple hearings about the
quality, recruitment practices, and cost of these
institutions, questioning whether the more than $26
billion in federal aid given to these schools each year
is a good investment.
Likewise, the Obama administ ration is seeking greater
control of these schools through the regulatory
process. This June, the U.S. Department of Education
proposed language on 15 issues that, among other
things, would prevent for-profit institutions from
paying recruiters based on student enrollment and
provide more consumer protections against false
advertising. A mont h later, it released a proposal to
define gainful employment by relying upon measures
of student borrowing, expected earnings, and student
loan repayment rates. That regulation is currently
going through a public comment period, and a final
version will be released later in 2010. If enacted, it will
go into effect on July 1, 2011. It if does, the standard
would tie college quality to work-force outcomes
for the first time, substantially changing the way
vocational programs are judged in the process.
Calculating Gainful
Employment
To determine whether a program meets the proposed
gainful employment standard, the department will
consider how much students borrow to attend that
EDUCATION SECTOR REPORTS: Are You Gainfully Employed? 3
program, the annual income of program graduates,
and the rate at which program graduates repay their
loans. Using those first two pieces of information,
the department will create a ratio of annual earnings
to debt payments. Each individual program offered
by a for-profit institution and all non-degree training
programs at public or private, not-for-profit colleges
would have to meet certain thresholds on the debt-to-
earnings comparison and maintain a certain minimum
repayment rate in order to remain eligible for federal
student aid funds.
A program's repayment rate looks at the status of
federal student loans that entered repayment in the
last four years. Among those loans, it measures the
dollar amount of all loans being actively repaid divided
by the amount of loans that entered repayment during
that time frame. The regulation defines the numerator
of this equation as the original outstanding principal
balance of all loans that entered repayment in the
past four years and were repaid in full or had enough
payments to reduce the principal owed in the last
fiscal year.
14
Note that the numerator is based on the
original principal value of a loan that is being repaid,
Ineligible programs have a
repayment rate below 35
percent and an annual loan
payment that is both above
12 percent of average annual
earnings and 30 percent of
discretionary income.
not the amount repaid. In other words, if the principal
owed is reduced by at least $0.01, then the entire
balance of the loan is counted in the numerator. For
example, consider a school that had two students
borrow $1 ,000 each. One borrower makes no
payments and does not reduce the principal , the other
pays enough to reduce the principal owed to $999.
In this case, the repayment rate is $1 ,000 divided by
$2,000, or 50 percent.
The repayment rate can also be thought of as the
percentage of loans being repaid, weighted f or the
size of the loans.
4 EDUCATION SECTOR REPORTS: Are You Gainfully Employed?
The proposed gainful employment standard also
judges programs based upon two ratios of students'
annual debt payments to their earnings. They are
calculated using the following three figures:
Annual loan payment: The median borrowing
amount among graduates from the past three years
is used to calculate an annual loan payment based
on the assumption that the debt is paid over 1 0
years with an interest rate equal to the standard
unsubsidized Stafford loan rate.
Average annual earnings: The average annual
income earned by program graduates over a
three-year period.
15
This data will be collected by
a federal agency, most likely the Social Security
Administration, and reported as a single figure to
the department.
Discret ionary income: The average annual
earnings minus 150 percent of the poverty
threshold for a single individual living in the
continental United States (about $16,245 in 2009).
For each program, the department calculates the
annual loan payment and then compares it to bot h
the annual average earnings and discretionary income
numbers for that program.
The department's proposal establishes thresholds for
repayment rates and debt-to-income ratios. Based
upon their performance relative to these thresholds,
programs are placed into four different categories of
eligibility: "eligible," "ineligible," "eligible with a debt
warning," or "restricted."
These categories rest on three types of ranges. If a
program is entirely above the upper bound, then it is
eligible; if it is entirely below the lower bound, then it
is ineligible. All others that fall somewhere between
these two thresholds are either eligible with a debt
warning or restricted.
Eligible: These programs have a repayment rate of
at least 45 percent, and the annual loan payment is
less than or equal to 8 percent of the average annual
earnings or 20 percent of discretionary income.
Eligible programs are free of any restrictions.
Ineligible: These programs have a repayment rate
below 35 percent and an annual loan payment that
is both above 12 percent of average annual earnings
www.educationsector.org
Table 1. How to Determine If a Program Meets the Gainful Employment Standard
AND

>20%
>8%
> 8% 12% > 20% 30%
> 8% 12% >30%
>12% > 20% and 30%
>12% >30%
and 30 percent of discretionary income. These
programs will lose federal student aid eligibility.
Ineligible programs may not provide federal student
aid to any new students and must warn existing
enrollees of debt dangers, in addition to all the other
disclosures described below.
Eligible with a debt warning: These programs have
either a repayment rate or debt-to-income ratio that
exceeds the threshold for an eligible program, but
not both. In other words, the repayment rate must be
at or above 45 percent, or the debt-to-i ncome ratio
must be at or below 8 percent or 20 percent, but both
cannot occur.
16
These programs may still participate
in the federal student aid programs, but they must
include in all materials and online a prominent warning
saying that students may have difficulty repaying their
loans. The school must also disclose its recent loan
repayment and debt-to-income rates.
Restricted: These programs do not meet any of the
requirements for an eligible program, but exceed
some of the benchmarks for an ineligible program.
They occupy a middle ground. These programs have
a repayment rate below 45 percent and a debt-to-
income ratio below 8 percent and 20 percent. They
also have either a repayment ratio at or above 35
percent or a debt-to-i ncome ratio at or below 12
percent or 30 percent, or both. Programs marked as
restricted by the department will have their enrollment
capped at the average of the past three years, must
provide warnings to consumers about debt levels, and
must get statements from area employers about why
the program should continue.
www.educationsector.org
Repayment rate is ...
Eligible
Eligible
Debt warning
Debt warning
Debt warning
Debt warning
Debt warning
Debt warning
Restricted
Restricted
Restricted
Restricted
Debt warning
Debt warning
Restrjcted
Restricted
Restricted
INELIGIBLE
These last two categories are opposites. A program
that is eligible with a debt warning is good enough on
one measure to meet the eligible threshold, but falls
short in the other. A restricted program doesn't meet
any of the eligible standards, but has at least one
calculation that is above the ineligible threshold. (See
Table 1.)
Other Important Considerations
In addition to the new set of standards and
measurements, the proposed gainful employment rule
contains a few other provisions worth mentioning.
The bottom 5 percent: The gainful employment
standard will go into effect in the 2012-13 academic
year, but penalties will be administered differently
that first year. Rather than preventing all ineligible
programs from receiving federal student aid, those
with low repayment and debt-to-income rates will
be broken down by the type of degree or credential
(associate, bachelor's, certificate, etc.) awarded.
In each category, the programs will be sorted by
repayment rate. The programs with the lowest rates
will lose their eligibility for aid until the enrollment
of all the ineligible programs equals 5 percent of
the enrollment of all programs in that category. The
remaining programs will face the same penalties as
restricted programs.
New programs: Traditionally, new programs have
been able to gain access to federal student aid dollars
as long as they are offered at an accredited institution.
EDUCATION SECTOR REPORTS: Are You Gainfull y Employed? 5
Methodology
It is possible to estimate potential income levels for
individual programs because all institutions are required
to report an instructional program code for each of their
offerings.
1
Each code corresponds to specific professions
and their earnings data-information that is kept by the
U.S. Bureau of Labor Statistics.
2
Because one code can
be linked to multiple professions, the salary estimate
ref lects the average annual 25th percentile earnings for
each instructional code. For codes tied to multiple jobs,
the earnings are weighted by the number of people
employed in each job.
Multiplying the salary amount by 8 percent and 12 percent
produced income thresholds for the average annual
earnings. Calculating discretionary income required
subtracting 150 percent of the poverty threshold for
a single individual and then multiplying the remaining
amount by 20 percent and 30 percent. Under the
proposed gainful employment calculation, the income
thresholds will be compared to the annual payment on
debt owed by a program's students. Because actual
earnings data are not available, this analysis instead
calculated how much debt a student could take on
if his or her annual loan payments were equal to the
income thresholds determined earlier. Loan amounts
were calculated by dividing the average annual earnings
and the discretionary income levels by 12 to determine
a monthly payment. The resulting figure was then
used to calculate the original amount a student would
have borrowed if he was making that payment each
month tor 1 0 years and had a fixed interest rate of 6.8
percent.
3
These loan terms are identical to those on an
unsubsidized federal Stafford loan.
The resulting original Joan balance amounts represent
the maximum debt load a student could take on.
Students with higher debt burdens would be devoting
a larger percentage of their income to making annual
loan payments than what is allowed under the proposed
standard.
While the U.S. Department of Education did release
estimated federal student loan borrowing levels by
institution, these figures did not include average private
Joan borrowing by school-additional debt that is also
included in the proposed gainful employment calculations.
Private student loans can be a significant source of
additional debt. Moreover, the borrowing information was
not reported uniformly-some inst itutional data reflected
both graduate and undergraduate borrowers, while some
offered separate Hgures for the two. The Jack of private
student loan borrowing is especially problematic because
it is a significant source of additional debt.
4
Unfortunately,
there is no central repository of reliable information on
private student loan borrowing at the institutional level.
fi EDUCATION SECTOR REPORTS: Are You Gainfully Employed?
So instead of relying on reported borrowing information,
this analysis approximates students' debt levels by
looking at the cost of their program, minus federal grant
aid received. Where available, program costs were
calculated using actual pricing information for tuition,
fees, books, and supplies reported by institutions.
5
This information ls available for 1 ,890 schools in the
department's Integrated Postsecondary Education
Data System, or I PEDS, that also had repayment rate
information available. For programs that take more than
a year to complete, the total cost figure encompasses all
charges that may be spread out across several years. If
an institution did not specifically report costs by program,
its cost estimate per program is based on the school's
published figures for tuition and tees and books and
supplies. Bachelor's degree programs reflect the past
four years of cost information; associate degree programs
reflect the past two.
While students at proprietary colleges and universities do
take on large levels of debt, they also receive significant
amounts of federal grant aid. To account for this, each
program's cost estimate was reduced by the average
amount of federal grant aid received by students at
that institution, data that are also reported to I PEDS.
Bachelor's degree programs had their costs reduced by
the average federal grant aid received over the past four
years; associate degree programs had their costs reduced
by the average grant aid over the past two years.
Programs' eligibility under the proposed gainful
employment standard was then determined by taking the
cost estimates and subtracting from them the borrowing
thresholds established using the Bureau of Labor
Statistics data. If a program's cost minus the borrowing
thresholds yielded a positive number, a student was likely
to borrow more than the income estimates would allow.
If the subtraction produced a negative number, students
were likely to borrow less than the maximum allowable
amount. This information was paired with the institutional
repayment rates reported by the department to determine
whether a school would be affected by the proposed
standard. While there is no guarantee that such programs
would actually be sanctioned under the new standard,
this analysis does suggest which types of programs
and institutions may need to consider reining in their
borrowing over the next few years.
There are several limitations to this approach. Students
may borrow less than the full amount to cover program
costs, or they may borrow more than the full amount The
assumption that students will borrow the full cost of their
education, minus available federal grants, is supported by
a number of statistics on student debt and the revenue
structure of for-profit colleges. An analysis of data from
the National Postsecondary Student Aid Survey published
www.educationsector.org
Methodology, cont.
by Education Sector last year shows that over 92 percent
of students at for-profit colleges take out a loan to finance
their And the debt levels these students
assume are quite high. The department estimates that
the median debt level of for-profit students ranges from
$18,415 for an associate degree to over $31 ,000 for a
bachelor's degree.
7
Similarly, the College Board found
that over 53 percent of bachelor's degree recipients at
for-profit institutions graduated owing over $30,500.
8
And
while it is reasonable to expect that not every student
would borrow at such a high level, for-profit colleges and
their defenders note that they cannot control how much
students borrow and that some take out too much debt.
That over-borrowing could balance out statistics for
students who take out a lesser amount in loans.
The revenue structure of for-profit colleges also bolsters
the assumptions about student borrowing used in this
report. For-profit institutions may not take in more than
90 percent of their revenue from the federal student aid
programs, and many, especially the large publicly traded
companies are close to this threshold.
9
After subtracting
federal grant dollars, the remaining loan money represents
a very significant portion of a school's revenue. And there
is evidence that the revenue that does not come from
federal aid programs still comes in the form of loans. For
example, Corinthian Colleges Inc., a large publicly traded
company, reported that 89 percent of its revenue comes
from federal aid programs and only about 1 to 2 percent
comes from cash payments from students.
10
That leaves
private student loans as the most likely other source of
financing.
Students' actual earnings could also be higher than the
Bureau of Labor Statistics figures. Estimates of federal
grant aid may also be too generous, because they reflect
awards given to first-time, full-time students. These
amounts are going to be higher than what the large
number of part-time students would receive.
While this approach could incorrectly identify programs
as at-risk for violating the gainful employment standard
when they are not, it is also possible for the opposite to
occur. Some programs may produce graduates with even
lower earnings than the national figures, further lowering
the income threshold. This analysis is meant to provide
an overall estimate of how the proposed federal rules
will change the for-profit higher education sector as a
whole and how different program types are likely to be
affected. It is not a fool-proof predictor of which individual
institutions or programs will meet the standards.
www.educationsector.org
Notes
1. These institutions are known as program reporters and must
break down cost by program because their offerings start
at multiple times throughout the year and do not follow the
standard academic calendar.
2. Each student aid-eligible offering is assigned a six digit
Classification of Instructional Programs (CIP) code that
identifies the content area covered by a program. For
example, a program in cosmetology may be assigned
the code 12.0401, which corresponds to "Cosmetology/
Cosmetologist, General." Using information from the National
Crosswalk Service Center, institutions can then link a
program's CIP code to the corresponding Bureau of Labor
Statistics codes. To continue the example, a program with
a CIP code of 12.0401 has 'four corresponding codes in the
bureau's database: 39-5012 (hairdressers, hairstylists, and
cosmetologists), 39-5091 (makeup artists, theatrical and
performance), 39-5092 (manicurists and pedicurists), 39-5094
(skin care specialists). For each profession, the SOC database
provides information on the number of people employed in an
occupation and thei r earnings at the 1Oth, 25th, 50th, 75th,
and 90th percentiles.
3. Stan Brown, "Loan or Investment Formulas," Oak Road
Systems, February 19, 2010, http://oakroadsystems.com/
math/loan.htm#LoanAmount (accessed September 7, 2010).
4. Sandy Baum and Patricia Steele, "Who Borrows Most?
Bachelor's Degree Recipients with High Levels of Student
Debt," College Board, April 2010, http://advocacy.
cotlegeboard.org/sitesldefault/1iles!Trends-Who-8orrows--
MostBrief.pdf (accessed September 7, 201 0), 3.
5. Program reporters must provide the total cost for a program.
Thus, if a two-year program is listed, the expense figure that
goes with it covers both years of enrollment.
6. Kevin Carey and Erin Dillon, ''Drowning in Debt,"
(Washington, DC: Education Sector, July 9, 2009) http://
www.educationsector.org/analysis/analysis_show.htm?doc_
id=964333 (accessed September 7, 2010).
7. "Program Integrity: Gainful Employment (Notice of Proposed
Rulemaking)," Federal Register Page 43647.
8. Sandy Baum and Patricia Steele, "Who Borrows Most?
Bachelor's Degree Recipients with High Levels of Student
Debt," 1.
9. "Emerging Risk: An Overview of Growth, Spending, Student
Debt and Unanswered Questions in For-Profrt Higher
Education," U.S. Senate Health, Education, Labor, and
Pensions Committee, June 24, 2010, http://harkin.senate.gov/
documents/pdf/4c23515814dca.pdf (accessed September
10, 2010), 4.
10. "COCO - 02 201 0 Corinthian Colleges Earnings Conference
Call," Final Transcript, Thomson StreetEvents, February 2,
2010, http://phx.corporate-ir.net/External. File?item=UGFyZW
50SUQ9Mzc3NjM4fENoaWxkSUQ9Mzc2NTI4fFR5cGU9M0=
=&1=1, (accessed September 7, 2010), 10.
EDUCATION SECTOR REPORTS: Are You Gainfully Employed? 7
Under the proposed gainful employment standard,
new programs will also have to be approved by the
department. They will have to fi le an applicati on with
the department and include enrollment projections
along with comments from unaffiliated employers
about the need for such a program and the availability
of related jobs. The department will then determine
whether to grant the program eligibility.
Longer time frames: Programs that prepare
people for careers in which income levels increase
substantially after a few years (such as doctors or
social workers) can have their data calculated over the
The ineligible programs in
culinary arts have an average
repayment rate of 27 percent
and have costs more than
$29,000 above the borrowing
limits.
fourth, fifth, and sixth years after their students leave
school. To stay eligible, programs that use the longer
time frame must show that their students' annual loan
payment is no more than 20 percent of discretionary
income or no more than 8 percent of average annual
earnings.
Overall Results
Under the gainful employment proposal, the
department estimates that about 5 percent of
programs would be deemed ineligible and 8 percent
would be restricted.
17
Another 48 percent would be
eligible with a debt warning, and 39 percent would
be fully eligible.
18
Among for-profit institutions, the
department expects about 1 ,658 programs to be
declared ineligible, but it did not say how many for-
profit programs it considered overall.
19
It provided
no detai ls on which institutions or program types are
likely to fall into this category.
It is impossible to independently verify the
department's estimates because student income data
8 EDUCATION SECTOR REPORTS: Are You Gainfully Employed?
are protected by federal privacy laws. This analysis
estimates how many and what kind of programs are
at risk under the proposed standards by comparing
program costs to the potential income levels of
program graduates and by looking at the repayment
rate for the institution overall. In other words, if a
student borrowed to cover the entire cost of his or
her program, after receiving available federal grant
aid, would that amount, coupled with the institution's
repayment rate, put that program in danger of losing
eligibility? (See "Methodology" sidebar on page 6 for
a complete explanation of these estimates.)
This analysis examined 12,662 programs offered by
2,667 colleges or universities. It encompasses three
different types of programs, all of which are classified
by an instructional code known as a CIP. First, the
analysis looked at institutions that report the total
cost of specific programs.
20
This group includes 793
programs offered at public or private, not-for-profit
institutions, only two of which would be negatively
affected by the standard. These 6,140 programs at
1,890 institutions report students' exact cost to attend
that program. The analysis also included any program
offered at a for-profit institution that produced at
least one bachelor's or associate degree last year.
This includes 2,351 bachelor's degree programs at
431 schools and 4,171 associate degree programs
at 721 schools. In every case, only institutions with
repayment rate information were included.
Of this sample of more than 12,600 programs,
504-or about 4 percent- would be ineligible for
federal student aid funds based upon this analysis.
That percentage is a bit lower than the department's
estimates. Of the remaining programs, 16 percent
would be eligible, 65 percent would be eligible with a
debt warning, and 15 percent would be restricted.
21
(See Figure 1.)
Ineligible Programs
The 504 programs with high cost-to-income ratios
and low repayment rates are offered at 222 different
colleges or universities. Of those, 1 02 colleges had
more than one ineligible program. But 196 of the
institutions with an ineligible program had at least
one other program that would retain its eligibility.
This means that even if that program lost student aid
eligibility, the school could continue offering other
www.educationsector.org
Figure 1. Eligibility Status Under Gainful
Employment
Bachelor's Degree Associate Degree
Programs Programs
-Ineligible
Restricted
Debt
warning
Eligible
programs, so, it would not be put out of business
entirely.
These 504 programs represent 87 different instruc-
tional codes. The instructional type with the largest
number of ineligible programs is medical/clinical
assistant, which represented 7 4 of the 504 violations.
Other common types were programs for culinary arts/
chef trai ning (34), e-commerce (31), and accounting
technology/technician and bookkeeping (26).
Though medical/clinical assistant offerings were
among the most common program types to violate the
borrowing standards, the average amount by which
they exceeded these thresholds was much lower than
that for other program types-particularly those in the
food services. On average, medical/clinical assistant
exceeded the borrowing threshold for 12 percent of
average annual income by just over $7,900. Similarly,
other health-related programs like health information
and medical records, medical insurance coding, and
medical office assistant all had several programs in
violation, but these exceeded the 12 percent threshold
by an average of between $7,000 and $9,000. Since
many of these programs are offered at the associate
or bachelor's degree levels, that works out to only a
few thousand dollars over each year.
Other program types are nowhere near meeting the
gainful employment standard. The ineligible programs
www.educationsector.org
in culinary arts have an average repayment rate of 27
percent and have costs more than $29,000 above the
borrowing limits. The 12 ineligible programs in baking
and pastry arts also fare poorly with an average
repayment rate of 25 percent and costs more than
$22,000 above the limits.
Others have already raised concerns about student
debt at culinary institutions. In March, the New York
Times ran a front-page article on student debt at for-
profit colleges, featuring a picture of students in chefs'
uniforms. Inside, it discussed the story of Andrew
Newburg, who paid $41 ,000 for a program at Le
Cordon Bleu with the promise of a $38,000 line cook
job, only to find out classmates were taking $8-an-
hour dishwashing jobs.
22
Thirteen programs at Le
Cordon Bleu show up on the list of ineligible schools,
and 18 are on the list of restricted schools.
Restricted Programs
According to the analysis, 1 ,899 programs, or 15
percent, would be restricted under the proposed
gainful employment standard. These offerings all had
repayment rates that were too low or cost-to-income
ratios that were too high, but not both. The programs
are offered by 807 different institutions. Cosmetology
programs were the most common type of offering to
Table 2. Most Common Types of Ineligible
Programs
Instructional category
-
-
MedicaVCiinical Assistant 74
Culinary Arts/Chef Training 34
E-Commerce/Eiectronic Commerce 31
Accounting Technology/Technician and 26
Bookkeeping
Graphic Design 22
Health Information/Medical Records 21
Technology/Technician
Interior Design 20
Administrative Assistant and Secretarial 13
Science, General
Baking and Pastry Arts/Baker/f:>astry Chef 12
Design and Visual Communications, General 11
Medical Insurance Coding Specialist/Coder
Fashion Merchandising
EDUCATION SECTOR REPORTS: Are You Gainfully Employed? 9
end up in this category, with 210 programs restricted.
Other program types that show up in large numbers
include medical/clinical assistant (142 instances),
animation, interactive technology, video graphics,
and special effects (83), and electrical , electronic
and communications engineering technology (70).
Forty-five culinary arts programs are categorized as
restricted, while 15 baker/pastry chef programs fell
into this category.
Many of these restricted programs could be in further
trouble if their graduates' earnings end up being
lower than the estimates. Over 800 of the restricted
programs had a repayment rate below 35 percent,
and 163 of these had a repayment rate below 20
percent. If these programs end up violating the debt-
to-income ratio they will become ineligible.
Bachelor's and Associate
Degree Programs
One particular concern raised by critics of the gainful
employment standard is that it would "preclude
for-profit colleges from offering bachelor's degree
programs," and eliminate many associate degree
programs, all due to their high cost-2
3
To test these
assertions, the analysis separated out all programs
Table 3. Most Common Types of Restricted
Programs
Instructional category
Cosmetology/Cosmetologist, General 210
Medical/Clinical Assistant 142
Animation, Interactive Technology, Video 83
Graphics and Special Effects
Electrical, Electronic and Communications
Engineering Technology!fechnician
CAD/CADD Drafting and/or Design
Technology !Technician
Corrections and Criminal Justice, Other
Legal Assistant/Paralegal
Administrative Assistant and Secretarial
Science, General
Graphic Design
Interior Design
Culinary ArtS/Chef Training
70
68
66
55
53
47
45
1 Q EDUCATION SECTOR REPORTS: Are You Gainfully Employed?
offered by for-profit colleges in these two degree
types. This was done by using completion data from
I PEDS, which each year reports the type of program
and degree level for every credential conferred by
a college or university. A program's total cost was
estimated using the figures for tuition, fees, books,
and supplies for the total number of years it would
take to complete a program. This means summing the
cost over four years for bachelor's degrees and over
two years for associate degrees.
24
This methodology has a few additional limitations.
Unlike the programs in which the institutions report
specific costs, exact tuition charges are not available
for these bachelor's degrees. Instead, the analysis
assumes that the tuition is the same for each offering
at a school. Recent research indicates price variability
is less pronounced at four-year institutions than at
two-year colleges. According to a research paper
published in May 201 0 by a fellow at the Association
for Institutional Research and the National Center
for Education Statistics, only about 13.3 percent of
for-profit four-year inst itutions vary their t uition by
program, and 6.7 percent vary their fees by program.
25
Second, it is possible that a student may take longer
to complete a degree. In that case, the cost would be
even higher than the estimate.
Bachelor's Degree Results
The bachelor's degree subset includes 2,351
programs offered at 431 institutions. Out of all the
bachelor's degree programs considered, 62 percent
would be either eligible or eligible with a debt warning
under the proposed gainful employment standard.
An additional 29 percent would be restricted, and 8
percent- or 193 programs- would be ineligible. (See
Figure 1.)
Ineligible Programs
The 193 programs that would be ineligible are
offered at 78 colleges and universities. This includes
programs at branches of the Art Institutes, the
Int ernat ional Academy of Design and Technology, ITT
Technical Institute, and Westwood College. Of the
ineligible programs, 31 are in e-commerce--t he most
of any program type. Other program types with large
numbers of ineligible programs include interior design
(19) and graphic design (16). On average, all of these
www.educationsector.org
Table 4. Most Common Types of Ineligible
Programs, Bachelor's Degrees
Instructional category
E-Commerce/Eiectronic Commerce
Interior Design
Graphic Design
Fashion Merchandising
Animation, Interactive Technology, Video
Graphics and Special Effects
Web Page, Digital/Multimedia and
Information Resources Design
Accounting Technology/Technician and
Bookkeeping
Computer Graphics
Design and Visual Communications, General
Fashion/Apparel Design
Cinematography and FilmNideo Production
-
mtmz:all
31
19
16
9
8
7
7
7
7
programs are well below t he minimum repayment rate
of 35 percent and are more than $20,000 away f rom
meeting t he debt -t o-income st andard.
These results indicate t hat programs connected
to greater work-force needs are less likely to be
ineligible. Only a handful of programs in accounting
and one program each in business management , legal
assistant/paralegal, and nursing would be ineligible.
By contrast , most ineligible programs are in "dream
job" areas: t hey provide t raining in cutt ing-edge
fields like online businesses and graphic design, or
in luxury occupations like int erior design or fashion
merchandising. These areas are associated wit h
relat ively high borrowing levels, but do not offer large
numbers of jobs.
Restricted Programs
About 29 percent of bachelor's degree programs
would be rest ricted-meaning t hey would not be
able to offer federal financial aid to new students and
would have t o demonstrate a continued need f or t heir
program f rom the local business community. The
most common types of programs in this category are
in animation, interactive technology, video graphics
and special effects (79 programs}, followed by
electrical, electronic and communications engineering
technology (70}, correct ions and criminal justice (45},
and interior design (36}. Programs in legal assistant/
www.educationsector.org
Table 5. Most Common Types of Restricted
Programs, Bachelor's Degrees
Instructional category
Animation, Interactive Technology, Video
Graphics and Special Effects
Electrical, Electronic and Communications
Engineering Technology/Technician
Corrections and Criminal Justice, Other
Interior Design
Legal Assistant/Paralegal
Accounting
Web Page, Digital/Multimedia and
Information Resources Desrgn
Graphic Design
Psychology, General
Computer a.nd Information Systems
Security
79
70
45
36
32
31
24
23
21
21
paralegal, accounting, web page, digi tal/multimedia
and inf ormat ion resources design, and graphic design
also appeared numerous times.
Associate Degree Results
The associate degree subset includes 4, 171 different
programs offered by 721 instit utions.
26
Of t hese
programs, 6 percent, or 267 programs, would be
ineligible under the gainful employment st andard.
Anot her 75 percent of programs would be f ully eligible
or eligible with a debt warning, while the remaining 19
percent would be rest ricted. (See Figure 1.}
Ineligible Programs
The 267 ineligible programs are offered by 142
different colleges or universi ties. This includes
programs offered by well -known college chains like
the Art Institutes, Everest (college, university, and
institute}, ITT Technical Institute, and Kaplan (college
and university}.
Fifty-seven different t ypes of programs fall into
the ineligible category. Medical/clinical assistant
programs are the most common type of offering to be
ineligible, with 67 programs fall ing into this category.
Programs in culinary arts/chef t raining (22} and health
inf ormat ion/medical records technology (21) also
EDUCATION SECTOR REPORTS: Are You Gainfully Employed? 11
Table 6. Most Common Types of Ineligible
Programs, Associate Degrees
Program instructional category
..
.
Medical/Clinical Assistant
Culinary Arts/Chef Training
Health Information/Medical Records
Tectmology/Techoician
Accounting Technology/Technician and
Bookkeeping
Administrative Assistant and Secretarial
Science, General
Medical Office Management/Administration
Baking and Pastry Arts/Baker/Past ry Cllef
Medical Insurance Coding Specialist/Coder
Medical Office Assistant/Specialist
Cosmetology/Cosmetologist, General
Graphic Design
67
22
21
"18
13
10
10
10
9
appear frequently on the list of ineligible program
types. On average, none of t he most common
types of ineligible programs is close to meeting the
repayment rate st andard, but many are less t han
$10,000 above the borrowing t hreshold. These
programs would have to either significantly reduce
costs or raise their repayment rates t o avoid losing
eligibility.
One int eresting trend that emerges f rom the data for
associate programs is t he large number of medical-
related programs that would be ineligible under
the proposed st andard. In addition t o the medical-
relat ed categories already mentioned, medical
programs in insurance coding, office assistant,
office management, and secretary all appear in high
numbers on the list of ineligible programs. All of t hese
professions have low starting salaries. As a result,
many would not be able to justify t he price tag of as
much as $28,000 for some of t hese programs.
Restricted Programs
The 787 restricted programs are offered at 375
different colleges and universities. This includes
multiple branches of Argosy University, the Art
Inst itutes, Brown Mackie College, Bryant and St ratton
College, DeVry University, Everest (college, insti tute,
and university), ITT Technical Inst itute, and Le Cordon
Bleu. The most common program types on this list
12 EDUCATION SECTOR REPORTS: Are You Gainfully Employed?
Table 7. Most Common Types of Restricted
Programs, Associate Degrees
Program instructional category
MedicaVClinical Assistant
CAD/CADD Drafting and/or Design
Technology/Technician
Administrative Assistant and Secretarial
Science, General
Health Information/Medical Records
Technology/Technician
Medical Administrative/Executive Assistant
. and Medical Secretary
Medical Insurance Coding Specialist/Coder
Allied Health and Medical Assisting Services/
Other
Medical Office Management/Administration
Graphic Design
Account ing Technology/Technician and
Bookkeeping
Pharmacy Technician/Assistant
Culinary Arts/Chef Training
lltl!ft!B
~
108
68
53
39
34
27
27
26
23
21
21
21
varied somewhat from t hose on the list of ineligible
programs. Medical/clinical assistant programs again
showed up on t he top with 1 08 rest ricted programs.
They are followed by programs in computer aided
design and drafting (68), administ rat ive assistant
and secretarial science (53), and health informat ion/
medical records t echnology (39).
In aggregate, data on associate and bachelor's degree
programs suggest t hat a larger percentage of t hese
programs might be in danger of losing eligibility
than the overall sample that also included certificate
programs. For bachelor's degrees, the programs at
greatest risk appear to be in fields t hat are related to
high-t ech jobs or luxury professions t hat may not be
in high demand; for associate degrees, the concerns
arise around expensive programs associated with
low-salary professions. The data suggests t hat degree
programs that provide training for wel l-paying, high-
demand jobs should not be affected too severely.
Publicly Traded Companies
Several programs at publicly t raded companies
show up on the lists of either ineligible or rest ricted
programs. The Art Institutes, Everest University, ITT
www.educationsector.org
Technical Institute, and Westwood College would all
have both ineligible and restricted programs. While the
University of Phoenix, Strayer University, DeVry Inc.,
and American InterContinental University would not
have any ineligible programs, they all show up on the
list of restricted programs. Table 8 shows estimates
of the percentage of programs at colleges owned
by publicly traded companies that would be eligible,
eligible with a debt warning, restricted, or ineligible.
It includes programs of all levels from certificate to
bachelor's degree.
An Important Step Forward
The proposed gainful employment regulation is a
significant departure from existing laissez-faire policy
on federal student aid, policy that requires little
accountability for the use of funds or outcomes for
students. Formally acknowledging the link between
training and earnings is an important codification of
the promises about jobs and salaries that for-profit
institutions highlight in their marketing materials.
The standard also represents a first step in better
engaging employers in discussions about higher
education. It means that before an institution can offer
a new program it must provide assurance from local
companies that the curricula are aligned with needed
ski lls and that sufficient job demand exists. Restricted
programs must provide the same information. Local
companies hire area graduates and can recognize
a quality training program. Seeking their feedback
recognizes the valuable role they can play in helping
to ensure that students are entering a program that is
likely to produce jobs.
The gainful employment standard is also a gain
for students as consumers. Publishing repayment
rates, debt ratios, and cost warnings gives potential
enrollees information that, using a set formula, can
be compared across all institutions. This information
is more useful than job placement data, which can
be calculated in different ways and is not easily
verifiable. It is also more helpful t han graduat ion rate
information, which doesn't give an accurate picture
of student success at schools with large numbers of
non-traditional or part-time students. But perhaps the
greatest consumer benefit is providing t his information
on a program-by-program basis, rather than
aggregating it across an inst itut ion. For-profit colleges
offer a wide variety of training programs in completely
unrelated fields, so breaking apart t he information by
program ensures that a nursing student, for example,
can see how his or her program actually performed
without getting results conflated with business
programs that serve students seeking very different
careers.
Table 8. How Publicly Traded Companies Would Fare Under the Gainful Employment Standard
Company Eligible(%) Debt warning (%) Restricted(%) Ineligible(%) Programs
All publicly traded companies 4 64 21 6 3,991
American Public Education 85 15 0 0 52
Bridgepoint Education 43 57 0 0 30
Capella Education Co. 0 62 38 0 '13
Career Education Corp. 0 61 23 16 430
Corinthian Colleges Inc. 5 80 4 626
DeVry Inc.
Education Management Corp. 1 40 46 13
Grand Canyon Education Inc. 22 78 0 0 46
ITI Educational Services Inc. 2 66 25 7 825
Kaplan Higher Education 5 77 12 5 316
Lincoln Educational Services 7 66 18 9 139
Strayer Education Inc. 0 84 16 0 19
Universal Technical Institute Inc. 4 96 0 0 46
University of Phoenix 0 90 10 0 487
www.educationsector.org EDUCATION SECTOR REPORTS: Are You Gainfully Employed?
13
The gainful employment standard also explicitly
introduces value as one way of judging quality and
cost at the same time. Current college rankings,
especially those compiled by U.S. News & World
Report, already list "best value" colleges that offer
good quality for their cost. But the opposite can
also be true; at a certain price point, the quality of a
college becomes irrelevant. A student could get the
best clinical/medical assistant training in the world,
but if it costs $35,000 or more, and if the average
annual income for these positions is just $29,000, then
the student is going to have trouble paying off the
debt. The inclusion of debt-to-income ratios by the
gainful employment standard is a worthwhile attempt
to capture this idea with easily understandable data.
At a certain price point, the
quality of a college becomes
irrelevant. A student could
get the best clinical/medical
assistant training in the world,
but if it costs $35,000 or more,
and if the average annual
income for these positions is
just $29,000, then the student is
going to have trouble paying off
the debt.
But the proposed standard isn't perfect. It applies to
all for-profit programs (except those in the liberal arts),
but only to non-degree certificate programs at public
and private not-for-profit colleges. Thus, a for-profit
institution and a neighboring community college could
offer the exact same program, but only one of them
would have to meet the gainful employment standard.
If the offerings and instructional program codes are
the same, then the same standards should be applied
to all institutions. Programs in the liberal arts should
continue to be excluded because they do not carry
an implicit promise of a job in a specific profession.
But strictly vocational offerings should be subject to
14 EDUCATION SECTOR REPORTS: Are You Gainfully Employed?
the gainful employment standard regardless of the
college's tax status.
The calculation of the repayment rate also presents
some difficulties. Loans are counted as being repaid
if there has been any reduction in the principal
owed. But this standard fails to consider whether
borrowers are actually on track to pay off the debt on
time or whether they are just making a few minimum
installments. For example, a borrower with a $1 ,000
loan who makes enough payments to reduce the
principal by $1 a year for four years is considered to
be in repayment even though, at that rate, he or she
wouldn't be able to pay down the debt in 1 0 years.
Including both federal and alternative, or private,
student loans in the borrowing figures also makes the
debt-to-income ratio harder to determine. Much of
private borrowing is direct to the student. There is no
central repository of information on private student
borrowing information, and schools may have no
way of capturing all the private loans taken out by
their students. Also, assuming a low fixed rat e will
make the annual payment on private loans seem
cheaper than it actually is. The loan also may have
a 20- or 30-year repayment time frame, rather than
the assumed length of 10 years. In either case, exact
costs are misstated. Private student loans also don' t
fit the rationale of why gainful employment needs to
be defined in the first place. From a taxpayer's point
of view, poor usage of federal student aid dollars
is a waste of scarce resources. There is no similar
taxpayer investment with private loans.
Another concern with the proposed standard's debt-
to-income ratio is that it only considers program
completers. While this makes sense from the
standpoint of wanting to make sure graduates are
gainfully employed, it is also important to remember
the large numbers of borrowers who drop out.
Students who fail to graduate are frequently left
with significant amount s of debt but none of the
economic benefits associated with a college degree
or certificate. One 2005 study found that students
who borrowed and did not complete their program
were "twice as likely to be unemployed as borrowers
who received a degree, and more than 1 0 t imes as
likely to default on their loan. "
27
A program that fails to
graduate large numbers of its students should also be
seen as not providing gainful employment.
www.educationsector.org
The debt-to-income ratio also fails to recognize
that f ederal student loan regulations allow students
to borrow beyond the total cost of their program.
Students who take out far more in loans than
necessary can inflate borrowing totals, which can lead
to a school being penalized for something largely out
of its control.
28
One way around this problem would be
to compare the average borrowing amounts relative to
program costs. Those that seem fairly close should be
judged according to the debt figures, while programs
with a large discrepancy should be subject to further
investigation. If an investigation shows that increased
borrowing is solely due to student decisions, then
the school should be judged only on a portion of that
borrowing.
A common critique of the new gainful employment
proposal is that it holds institutions accountable
for other outcomes beyond their control. More
specifically, it judges programs based on their
graduates' earnings several years after they have left
school. This issue recalls the rhetoric around cohort
default rates, in which colleges argue that student
characteristics, not program quality, play the largest
role in assessing whether a student is likely to default.
But this argument contradicts the schools' marketing
claims that their programs can improve students' lives
and lead to better jobs. Either programs lead to better
jobs and higher earnings-in which case those results
should be measured using a standard like gainful
employment-or they don't, in which case their
institutions should not be making these claims.
Gaining From Gainful
Employment
The gainful employment standard is just one part of a
larger movement to regulate for-profit colleges. Other
regulations released in June propose to eliminate
exemptions that previously allowed schools to tie
part of recruiters' compensation to getting students
to enroll. The regulations would also crack down on
so-called ability-to-benefit t ests-exams given to
students without a high school degree to determine
their ability to handle college-level work. A U.S.
Government Accountability Office report in September
2009 found that colleges were coaching ill-prepared
students through these tests so they could receive
federal student aid.
29
www.educationsector.org
Establishing clear connections
between employers, jobs,
wages, and training-oriented
programs is a welcome new
way of thinking about education
not just in terms of quality, but
also in terms of value.
Congress could also take legislative action in this
area. The Senate's Health, Education, Labor, and
Pensions Committee has already held two hearings
on for-profit colleges and promises to convene more
in the fall. The hearings could prompt legislation to
further limit the percentage of revenue that proprietary
colleges can receive from federal aid programs. Such
legislation might also prevent these colleges from
using money from these aid programs to pay for
marketing.
America's higher education system is among the
most diverse in the world in terms of the types of
colleges and variety of programs it provides. But, to
date, federal student aid programs have done little
to formally acknowledge that diversity. The gainful
employment standard is an important first step in
addressing t hat flaw. Establishing clear connections
between employers, jobs, wages, and training-
oriented programs is a welcome new way of thinking
about education not just in terms of quality, but also
in terms of value. Most important, it does this at the
sub-institutional level , acknowledging that these types
of programs have little overlap with other offerings at
the same institution. Critics of the gainful employment
standard are quick to disparage it as an attempt to
shut down the sector. But this claim is unfounded.
This analysis shows that gainful employment would
likely force 4 percent of programs to close and restrict
the activities of another 15 percent. But these results
are not set in stone. Colleges will not be judged by the
gainful employment standard for a few years yet, so
they will have time to reduce their costs and student
borrowing, to help students repay their debts, and to
help those students fi nd higher-paying employment.
And any process that can encourage reducing student
costs or improving job placement is a good thing.
EDUCATION SECTOR REPORTS: Are You Gainfully Employed? 15
Notes
1. Sen. Tom Harkin, "Beware of For-Profi t Higher
Education," Forbes, August 11, 2010, http:/ /www.forbes.
com/201 0/ 08/01 /higher-education-student -debt -opinions-
best-colleges-1 0-harkin .html (accessed September 6, 201 0).
2. Jennifer Epstein, "Pushback on Gainful Employment," Inside
Higher Ed, April 22, 2010, http://www.insidehighered.com/
news/201 0/04/22/gainful (accessed September 3, 201 0).
3. Daniel Bennett, "Beating My Head on the Desk Won't Stop the
Insanity," Center for College Affordability and Productivity, May
21 , 2010, http:/ /collegeaffordability.blogspot.com/20 10/05/
beating-my-head-on-desk-wont-stop.html (accessed
September 3, 2010).
4. Many institutions that do not offer a traditional semester-length
course schedule report information to the U.S. Department of
Education by program.
5. Gregory D. Kutz, "For Profit Colleges: Undercover Testing Finds
Colleges Engaged in Deceptive and Questionable Marketing
Practices, " U.S. Government Accountability Office, August 4,
2010, http:// www.gao.gov/products/GA0- 1 0-948T (accessed
September 7, 2010).
6. TESST College of Technology "Medical Assistant, " TESST
College of Technology, http:/ lbit.ly/aJk2aO (accessed
September 7, 2010).
7. Le Cordon Bleu Institute of Culinary Arts, "Le Cordon Bleu
Institute of Culinary Arts in Pittsburgh," http://lecordonbleu-
pittsburgh.com/index.asp?src=141574 (accessed September
7, 2010).
8. Google, "Kaplan Information Technology Degree Google
Search," http://bit.ly/ biWSjk (accessed September 7, 2010).
9. The Art Institute of Pittsburgh Online Division, "Interior Design
Program," http://bit.ly/9gUSbj (accessed September 7, 2010).
10. The Higher Education Act of 1965, Public Law 89-329, 89th
Gong., 1st sess. (November 8, 1965), Page 30, http://ftp.
resource.org/gao.gov/89-329/00004C5 7 .pdf (accessed
September 3, 2010).
11. Author-calculated statistic using data from the U.S. Department
of Education's Integrated Postsecondary Education Data
System.
12. Author-calculated statistic using data from the U.S. Department
of Education's Federal Student Aid Data Center.
13. Author-calculated statistics using data from the U.S.
Department of Education's Integrated Postsecondary
Education Data System. Kelly Field, ''Government Vastly
Undercounts Defaults," Chronicle of Higher Education, July
11, 2010, http:/ /chronicle.com/ article/Many-More-Students-
Are/66223/ (accessed September 3, 201 0).
14. Loans that are replaced by a single consolidation loan are not
considered paid off. Loans that have an in-school or military
deferment or entered repayment after March 31 of a given
fiscal year are excluded.
15. By default, institutions have the rate calculated over the three
years in repayment. If they can prove to the Department of
Education that graduates' salaries increase significantly in the
16 EDUCATION SECTOR REPORTS: Are You Gainfully Employed?
future, then It can oo judged over the fourth, fifth, and six years
after entering repayment.
16. There are three ways a program can end up in this category:
(1) it has a repayment rate above 45 percent and a debt-to-
income ratio above 8 percent and 20 percent, (2) it has a
debt-to-income ratio at or below 8 percent and a repayment
rate below 45 percent, or (3) it has a debt-to-income ratio at or
below 20 percent and a repayment rate below 45 percent.
17.James Kvaal (remarks at the conference, "Reining in For-Profit
Higher Education," New America Foundation, Washington,
D.C., July 30, 2010).
18.1bid.
19. "Program Integrity: Gainful Employment (Notice of Proposed
Rulemaking)," Federal Register 75:142, July 26,2010 Page
43636, http://www.gpo.gov/fdsys/ pkg/FR-201 0-07-26/
pdf/201 0-17845.pdf (accessed September 3, 201 0).
20. Some programs may be spread over multiple years, but the
cost figure represents the total expense to complete that
program.
21. The numbers do not add up to 100 percent due to rounding.
22. Peter Goodman, "In Hard Times, Lured Into Trade School and
Debt," New York Times, March 13, 2010, http://www.nytimes.
com/201 0/03/14/business/14schools.html (accessed
September 13, 201 0).
23. Mark Kantrowitz, "What is Gainful Employment? What is
Affordable Debt?" FinAid.org, March 11, 2010, http://www.
finaid.org/educators/201 00301 gainfulemployment.pdf
(accessed September 3, 2010).
24. Data for all other award levels was discarded because they
required fractions of a year. and it was too difficult to estimate
the appropriate cost amount.
25. Sean Simone, "Tuition and Fee Differentiation at Degree
Granting Postsecondary Education Institutions," Association for
Institutional Research, National Center for Education Statistics,
May 2010, http://www.airweb.org/images/ Simone_Final_
Report_2010.pdf (accessed September 10, 2010), 20.
26.1n total, the analysis considered 6, 7 47 programs at 790
for-profit institutions. Some colleges offered both bachelor's
and associate degrees and so were considered under both
categories, albeit for different programs.
27. Lawrence Gladieux and Laura Perna, "Borrowers Who Drop
Out: A Neglected Aspect of the College Student Loan Trend,"
National Center for Public Policy and Higher Education, May
2005, http://www.highereducation.org/reportslborrowing/
borrowers.pdf (accessed September 3, 201 0).
28. Some colleges argue that excess borrowing is used to fund
things like vacations and car payments. In other cases, the
school may encourage students to borrow more than they
need because it is easier to take out the maximum and then
return money, rather than take out too little and need more.
For example, a recent investigation from ABC News found that
students at a branch of the University of Phoenix were being
told to take out the maximum because it was easier and they
could keep the extra money with no questions asked.
29. "Proprietary Schools: Stronger Department of Education
Oversight Needed to Help Ensure Only Eligible Students
www.educationsector.org
Receive Federal St udent Aid," U.S. Government Accountability
Office, August 2009, http://www.gao.gov/ new.items/ d09600.
pdf (accessed September 10, 2010).
www.educationsector.org EDUCATION SECTOR REPORTS: Are You Gainfully Employed? 17
From: Shireman Bob
To: Yum Georgia
CC: Manheimer Ann
Date: 4/1/2010 6:44:12 PM
Subject: Fw: Request for Follow Up Meeting This Week to Discuss Gainful Employment
From: Hanis Miller
To: Shireman, Bob; Kanter, Martha; Madzelan, Dan; Bergeron, David
Cc: Brian Moran ; Tammy Halligan ; Manheimer, Ann
Sent Thu Apr 01 '17:04:04 2010
Subject: RE: Request for Follow Up Meeting Ths Week to Discuss Gainful Employment
Bob, et.al.: We look forward to the meeting Monday morning at 9 AM at the Department to discuss the gainful
employment issue. Attending as part of the CCA delegation will be, in addition to me, Brian Moran, CCA' s Executive
Vice President for Government Relations, Dr. Jonathan Guryan, University of Chicago, and Dr. Matthew Thompson of
Charles River Associates, the leads on the research we will be presenting, Rebecca Campoverde from Kaplan, Anthony
Guida from EDMC, and Tom Babel from Devry.
If you could please let me know who will be in attendance from the Department, that would be appreciated.
Attached you will find three documents. The first is the Executive Summary of the research findings. I will send you the
full report tomorrow, Friday. Last minute editing on the full report was delayed by the Passover holiday.
Second is the proposal CCA is putting forward as an alternative to the gainful employment proposal put forth by the
Department during the negotiated rulemaking.
Third is a document that summarizes the extensive tools the government already has to take actions against "bad actors"
in our sector.
Please let me know if you have any questions or comments. I look forward to a productive conversation Monday. Enjoy
the spectacular weather.
Harris
From: Shireman, Bob [mailto:Bob.Shireman@ed.gov]
Sent: Monday, March 29, 2010 10:48 AM
To: Harris Miller; Kanter, Martha; Madzelan, Dan; Bergeron, David
Cc: Brian Moran; Tammy Halligan; Manheimer, Ann
Subject: RE: Request for Follow Up Meeting Ths Week to Discuss Gainful Employment
Work with Ann Manheimer on arranging the meeting. Please send your research and alternative prior to the meeting so
we can get concrete and specific in the discussions.
Thanks,
-Bob
From: Harris Miller [mailto:HarrisM@career.org]
Sent: Monday, March 29, 2010 10:04 AM
To: Kanter, Martha; Shireman, Bob; Madzelan, Dan; Bergeron, David
Cc: Brian Moran; Tammy Halligan
Subject: Request for Foil ow Up Meeting This Week to Discuss Gainful Employment
Martha/Bob/Dan/David:
CCA would like to have a follow on meeting with you this week to the one Brian Moran and I had with Dan and David
ten days ago to discuss a) the results of our research (which was not complete when we met) on the impact of the
Department' s gainful employment proposal on students in higher education, and b) an alternative (which we discussed in
general concept with Dan and David) to the Department ofEducation proposal. We can make ourselves available at
almost any time starting tomonow. Please let me know whom I should contact to arrange a meeting. Thanks in advance.
Harris
*****************************************************************
Harris N. Miller
CEO/President
Career College Association
1101 Connecticut Avenue, NW, Suite 900
Washington, DC 20036
hanism@career.org
+ 'I 202 336 6754
Executive Assistant: Jackie McWilliams
j acki em@career. org
+ 1 202 336 6706
www. career.org
To:
From:
Date:
Re:
MEMORANDUM
Interested Parties
Career College Association
March 29, 2010
The Myth That the Department of Education is Powerless to Stop "Bad
Actors" in Postsecondary Education
CCA staff and members have heard during recent visits with Administration
officials, especially at the Department of Education ("the Department"), and less
frequently during Capitol Hill visits, a narrative that runs as follows: Although the
majority of for-profit postsecondary institutions may be high quality institutions serving
their students well, there are a few "bad actors" over which the Department lacks
sufficient regulatory authority. In fact, one of the justifications for the radical proposal
put forth by the Department to change the meaning of "gainful employment" as it has
been accepted and understood for forty years is to stop the conduct of "bad actors." Part
of this same narrative is that prospective students do not have adequate consumer
protections to prevent these bad actors from taking unfair advantage of them.
CCA agrees that any percentage of bad actors in the sector is unacceptable. What
we do not agree with, however, is the mythology that the Department lacks what is, in
fact, extensive authority under current law to investigate and take action against
institutions that are not following the myriad of rules that govern higher education
generally, and for-profit higher education, in particular. In addition to a high degree of
regulatory oversight at the federal level, each of these schools is also overseen by the
other two legs of the regulatory "triad" -state government and accrediting entities that
are approved by the Department itself. For example, over the last decade, the sector has
seen a vast increase in active oversight by state agencies.
This memorandum provides an overview of the layers of quality control and
consumer protection regulation that already exist within the authority of the Department
and other federal and state entities that can and do serve as a check on "bad actors." It is
important to note that some significant enhancements in the law were enacted as part of
the Higher Education Opportunity Act of 2008 ("HEOA"), which amended the Higher
Education Act of 1965 ("HEA"). The final regulations implementing the HEOA are not
effective until july 1, 2010 and, therefore, many important enhancements to the law have
yet to be fully implemented. In addition, during this year's Title IV Program Integrity
Negotiated Rulemaking process, the negotiators reached consensus on changes to
strengthen the misrepresentation regulations. The process of reaching consensus on
enhanced misrepresentation regulations, of which CCA was a part, demonstrates that
higher education and the Department share the goal of weeding out bad actors through
7
common sense enforcement of and enhancements to existing law.
I. Quality Control and "Bad Actors"
As a threshold matter, the Department has never explained clearly how it defines
a "bad actor" school or what specific rules such a school is violating. In reality, real laws
apply to real schools that violate them. For example, if a bad actor is a school that
misrepresents information to a potential student about placement rates or salary
prospects, federal and state consumer protection laws apply. If a bad actor is a school
that "lures" students to enroll through prohibited payments to admission or financial aid
officials, regulations subject that school to penalties and risks to Title IV eligibility. If a
bad actor is a school that charges "too high" a tuition relative to earnings, then students
will default massively and the cohort default rate thresholds will act as a check on Title IV
eligibility. This section describes the quality control regulations in current law put into
place over the last decade to improve quality and thereby eliminate so-called "bad
actors."
The Higher Education Act of 1972 established Title IV eligibility for proprietary
institutions consistent with the Department's goals of expanding postsecondary
education access for a growing and financially needy college-aged population. Since that
time, Congress has established and built upon the following measures in federal law,
some generally applicable to all postsecondary institutions and some specific to
proprietary schools, to ensure school and program quality:
Caps on excessive cohort default rates (Fiscal Year 1989 Appropriations Act);
Thirty day delay in delivery of loans to new borrowers by schools with cohort
default rates of more than 30% (Omnibus Reconciliation Act of 1989);
Prohibition on certification or recertification of an institution as Title IV eligible if
it has lost accreditation during the past 24 months, unless the accreditation has
been restored or the institution demonstrates program integrity (Omnibus
Reconciliation Act of 1989);
Addition of new program integrity provisions, including periodic recertification,
and other requirements for strengthened national accreditation standards and
procedures (Higher Education Amendments of 1992);
Establishment of early 85/15 (now 90/10) ratio for proprietary schools (Higher
Education Amendments of 1992);
Requirements for pro rata refund policy for first-time students who withdraw
(Higher Education Amendments of 1992) and return to Title IV policy as a
successor to that;
Enhancements to federal financial responsibility standards and requirement for
letters of credit for institutions with impaired financial responsibility;
Requirement for annual submission of audited financial statements and financial
aid compliance audits (Higher Education Amendments of 1992);
7
Establishment of incentive compensation prohibitions on student recruitment and
processing of financial aid;
Strengthened federal requirements for ability-to-benefit tests and limit on
percentage of ability to benefit students;
Student satisfactory academic progress requirements (Omnibus Budget
Reconciliation Act of 1989, Omnibus Reconciliation Act of 1990; Higher Education
Technical Amendments of 1991);
Required disclosure of completion and placement rate requirements for
short-term programs;
Establishment of the clock to credit hour conversion rules and minimum academic
year requirements as part ofPell grant eligibility;
Required disclosure of graduation rates (1990 Student Right to Know and Campus
Security Act);
Limitations on branch campuses;
Increased disclosures to borrowers of student loans; and
Disclosure of schools with the greatest tuition increases.
In addition to these measures, institutions must also ensure that the information
disclosed about programs is accurate. The Department may initiate a fine, or a limitation,
suspension, or termination of Title IV eligibility for any substantial misrepresentation
made by an institution regarding the nature of its educational program, its financial
1
charges, or the employability of its graduates. At the recently concluded negotiated
rulemaking, the Department proposed improvements to the misrepresentation
regulation that were agreed upon by consensus by stakeholders, including CCA, that will
increase the oversight of the information provided by institutions. Specifically, the
proposed regulatory language on which the negotiators reached consensus strengthens
the current misrepresentation regulation in several ways, including:
Expanding the class of persons to whom a substantial misrepresentation can be
made to include accrediting agencies and state agencies, as well as prospective
students and the general public;
Holding institutions responsible for any misrepresentation made by companies
with which they contract their services; and
Broadening the definition of what constitutes "substantial misrepresentation."
During negotiated rulemaking, the Department also proposed language that
would support robust state regulation of postsecondary institutions. In addition, it is
important to remember that under current state laws and state attorneys general have
broad grounds to pursue schools in violation of state licensing or consumer protection
requirements. State authorizing agencies and state attorneys general are additional tools
1
See 34 CFR 668.74(c).
7
against bad actors in the sector.
There are also other oversight agencies in addition to the Department that
sanction institutions for misrepresentations or inaccuracies in the information they
provide. Institutions are subject to oversight by the Federal Trade Commission
(proprietary institutions only), the accrediting agencies, the Federal Reserve Board (for
institutional and private educational loans), the Securities and Exchange Commission
(for publicly traded schools), and the Department's Federal Student Aid Office and Office
of Inspector General.
II. Consumer Protections
As stated above, the provision of postsecondary educational services is a highly
regulated activity with numerous controls and enforcement mechanisms currently
available to the Department of Education in statute and regulation that permit it to
identify, investigate, and end Title IV eligibility to institutions of higher education that
violate federal and state law. Consumer protection enhancements applicable to the
sector at the federal and state level have also grown consistently over the last decade,
were further enhanced in the HEOA, and can be strengthened even more, without relying
on a "gainful employment" hook that would damage quality educational service
providers.
A variety of consumer protections exist to inform the educational choices of
prospective and current students. Institutions are required to disclose detailed
information about college cost, student success metrics, and loan repayment. Under the
Higher Education Act and recent amendments to that Act, schools are required to provide
2
a vast array of consumer information, including the following:
The price of attendance, including tuition and fees, estimated cost of books and
supplies, room and board, and transportation costs, and any additional costs for a
program in which the student is enrolled or expresses an interest (HEA, 20 U.S.C.
1092(a)(1)-(2), unchanged by HEOA);
Institutions must make available on their websites by October 29, 2011 a net
price calculator (HEOA Section 111 amended HEA Title I, Part C; added HEA, 20
U.S.C. 1015a(a),(h));
Each institution must make available to prospective and enrolled students
information about the institution's refund policy, requirements and procedures
for withdrawal; and requirements for the return of Title IV, HEA grant or loan aid
(HEA, 20 U.S.C. 1092(a)(1)-(2), unchanged by HEAO);
Academic program and facility information to prospective and enrolled students
(HEOA added 20 U.S.C. 1092(a)(1)(G)(iv));
2
.s.e.e...HEA Section 48S(a)(l) (20 U.S.C. 1092(a)(l)); see also. Report of the National Postsecondary Education
Cooperative, Information Required to be Disclosed Under the Higher Education Act of 1965: Suggestions for
Dissemination, Nov. 2009 (including HEOA amendments).
7
Each institution must make available to prospective and enrolled students: names
of accreditors applicable to the institution and programs, and procedures for
obtaining and reviewing documents describing such accreditation (HEA, 20 U.S.C.
1092(a)(1)-(2), tmchanged by HEAO);
An institution that advertises job placement rates as a means of recruiting
students to enroll must make available to prospective students, at or before the
time the prospective student applies for enrollment: (1) the most recent available
data concerning employment statistics and graduation and (2) any other
information necessary to substantiate the truthfulness of the advertisements; and
(3) relevant state licensing requirements of the state in which the institution is
located for any job for which the course of instruction is designed to prepare
students (HEA, 20 U.S.C. 1094(a)(8), unchanged by HEOA);
Institutions must make available to current and prospective students information
regarding the types of graduate and professional education in which graduates of
the institutions' 4-year degree programs enroll (HEOA, 20 U.S.C. 1092(a)(l)(S));
Detailed financial aid information, including information about all of the
need-based and non-need based federal, state, local, private and institutional aid
programs available to students at the institution (HEAO added HEA
485(a)(l)(M));
The retention rate of certificate and degree seeking, first-time, full-time
undergraduate students as reported to !PEDS (HEOA added HEA 485(a)(l)(U));
The completion or graduation rate of certificate or degree seeking, first-time,
full-time, undergraduate students, including data disaggregated by gender, major
racial and ethnic subgroups, recipients of a Federal Pell Grant, recipients of a
subsidized Stafford Loan who did not receive a Pell Grant, and students who did
not receive either a Pell Grant or a subsidized Stafford Loan (HEOA added HEA
485(a)( 4) and HEA 485(a)(7));
Each institution must publicly disclose and make available to prospective and
enrolled students a statement of the institution's transfer of credit policies and
articulation agreements (HEOA added 20 U.S.C. 1092(h));
Institutions must make available to current and prospective students information
regarding the placement in employment of, and types of employment obtained
by, graduates of the institutions' degree or certificate programs (HEOA added 20
U.S.C. 1092(a)(l)(R));
Textbook cost disclosure on internet course schedule used for preregistration and
registration purposes (HEOA added 20 U.S.C. 1015b);
Detailed entrance and exit counseling for student borrowers, including average
anticipated monthly repayment amount, repayment plan options, and debt
management strategies (HEOA added new HEA 485(1) and 485(b )(l)(A)); and
Detailed private education loan disclosures, including a student self-certification
form. In addition, under the amendments to the Truth in Lending Act, private
lenders must provide three detailed disclosures to borrowers before making
7
private student loans (HEOA added HEA 487(a)(28) and HEA 151-155 (20
U.S.C. 1019, 1019a-1019d)).
The Department makes consumer information easily available to students, parents, and
the public. The Department posts many consumer information items on the College
Navigator website for each institution, including:
Tuition, fees and other estimated student expenses for the last several years, with
the percentage increase from the most recent year to the current one and a
multi-year tuition calendar;
The percentages of students who receive various types of aid and the average
amounts of such aid;
Graduation rates, disaggregated by major racial subgroups;
Retention rates for full-time and part-time students;
Average amount of subsidized and unsubsidized Stafford loans; and
Cohort default rates for the most recent three years.
In addition, the Department provides some of this consumer information direct to
individual students by institution - when the student inputs the school's code on the
FAFSA. It is also relevant that from the standpoint of providing tools to students to
manage student loan debt, Congress first authorized the Income Based Repayment
program in 2007 as part of the College Cost Reduction and Access Act of 2007, and that
program is just now being actively promoted by the Department as a tool to assist
graduates with loan debt management.
In sum, despite the perception created by the Department's "bad actor" narrative
on gainful employment, proprietary postsecondary education is highly regulated at the
federal and state levels and through its accrediting bodies. The gainful employment
proposal is vastly overbroad and misguided in light of the diverse array of tools that
exist under current law to identify and target bad actors and to protect and assist student
consumers with loan debt issues.
7
C
D A Lharles. River

Prepared For:
Harris N. Miller
Career College Association
1101 Connecticut Ave. NW, Suite 900
Washington, DC 20036
Prepared By:
Report on Gainful
Employment: Executive
Summary
Jonathan Guryan, Ph.D.,
Associate Professor of Economics, University of Chicago Booth School of Business
Matthew Thompson, Ph.D.,
Vice President
Charles River Associates
1545 Raymond Diehl Road, Suite 260
Tallahassee, FL 32308
Date: March 29, 2010
Report on Gainful Employment: Executive Summary
March 29, 2010 Charles River Associates
This report addresses the definition of "gainful employment" proposed by the U.S.
Department of Education as a part of negotiated rulemaking.
CONSEQUENCES/IMPACT OF THE PROPOSED GAINFUL EMPLOYMENT REGULATION
To estimate the impact of the proposed regulation on Title IV eligible for-profit postsecondary
institutions, we collected data from Career College Association (CCA) member institutions.
We analyzed data representing approximately 640,000 students and over 10,000 separate
programs from institutions ranging from very small to very large. Using these data, we
compared the median debt among a program's graduates to the maximum allowed loan
implied by the 8-percent rule applied to the appropriate 25th percentile of earnings, calculated
according to the Department of Education's instructions.
Based on these calculations, we estimate that 18 percent of for-profit postsecondary
programs would not satisfy the debt limit requirement of the gainful employment proposal. In
our calculations, larger programs will be more likely to be impacted. Consequently, we
estimate that 33 percent of students in for-profit postsecondary programs would be impacted.
Applying these findings to for-profit enrollment reported in IPEDS, we estimate that
approximately 360,000 students each year enter for-profit postsecondary programs that
would lose eligibility for participation in the federal Title IV financial aid programs. Given
recent growth rates of enrollment in for-profit postsecondary institutions, we estimate that by
2020, approximately 5.4 million students who are on track to attend programs would be
denied access by the proposed regulation. Our estimate is that the impact of the regulation
would be severe on racial and ethnic minority groups, in part because for-profit schools are
proportionately more likely than not for-profit postsecondary schools to serve these students.
We estimate that each year approximately 68,000 non-Hispanic black students and an
additional 79,000 Hispanic students enter for-profit postsecondary schooling in programs that
would be impacted by the proposed regulation and would not be able to attend these
programs. While some capacity may exist in other sectors of higher education to absorb
these students, recent reports indicate that the most likely alternatives-community
colleges-are already oversubscribed in many cases, and are facing further financial
cutbacks as the states that provide much of their funding face severe financial challenges.
Given recent growth rates at for-profit postsecondary institutions, we estimate that by 2020
approximately 1 million non-Hispanic black students and an additional 1 million Hispanic
students are on track to attend programs that would be adversely affected, and would be
denied access as a result.
Based on our estimates, the impact of the regulation would vary across types of programs.
Because the limits on borrowing do not vary with the length of program, longer programs
would be more severely impacted. Whereas approximately 18 percent of students in less
than 2 year programs would be impacted, we estimate that approximately 40 percent of
students in 2- and 4-year programs would be impacted. We also estimate that the impact
would not be limited to a few areas of study, but would impact a wide variety of programs.
For example, we estimate that nearly 14 percent of Health Professional and Related Clinical
Page 1
Report on Gainful Employment: Executive Summary
March 29, 2010 Charles River Associates
Sciences, including Nursing, programs and more than 46 percent of Engineering Related
Technologies/Technicians programs would not currently satisfy the proposed debt limit rule.
A SOLUTION IN SEARCH OF A PROBLEM
We are unaware of any, scientific or otherwise, study by the Department of Education that
outlines what the problem is that the proposed regulation is meant to address. The problem
that the regulation aims to solve has not been clearly stated - neither its nature nor its extent.
As a result, we believe, the proposed regulation is not well-designed to address a specific
problem.
Any regulation aimed at solving a specific problem should be tailored to address that problem
without causing undue harm to other individuals. Based on a few anecdotal references to
complaints of high debt levels by students, posted in the public comments to Negotiated
Rulemaking on the Department's website, we suspect that the Department of Education aims
to protect some students who take on more debt than they can expect to repay. We also
suspect that the number of students who take on more debt than is in their personal interest
is small, but we are not aware of a study of this population that would inform this speculation.
A regulation aimed at protecting these students should focus on identifying these students
and should address these students directly. Because the proposed regulation applies to
schools rather than students, it would affect students who are not over-borrowing along with
any who may be over-borrowing.
Are Default Rates Higher at For-Profit Colleges?
One stated rationale for the proposed regulation is that default rates are higher among
students at for-profit postsecondary schools than among those at not-for-profit colleges. Our
analysis of the Beginning Postsecondary Students Survey data shows that at least half of the
difference in default rates between for-profit and not-for-profit schools is because they serve
different types of students. For-profit schools are more likely to serve low-income students,
racial and ethnic minority students, students who are the first in their immediate family to
attend postsecondary schooling, and students whose families have collected public
assistance. Students in these groups are more likely to default on student loans after
attending both for-profit and not-for-profit colleges.
We estimate that if all schools served students with moderate family resources, loan default
rates at for-profit schools would be cut by more than half and the difference in default rates
between for-profit and not-for-profit schools would also be cut approximately in half. These
estimates are based on regression-based controls for students' family income, dependent
status, race/ethnicity, program completion status, parental education, family welfare receipt,
and Pell eligibility. It is possible that controlling for additional student characteristics, if data
were available, would reduce the default rate gaps even more.
Page2
Report on Gainful Employment: Executive Summary
March 29, 2010 Charles River Associates
RATIONALE FOR THE METRICS OF THE REGULATION
The debt limit portion of the proposed Gainful Employment regulation focuses on the ability of
recent graduates to repay loans in the early years of their post-schooling careers. The choice
of metrics to assure students' ability to pay is logically flawed. Furthermore, the basis of the
debt limit on earnings early in the career stands in contrast with standard economic analysis
of education, which clearly says that the choice of how much to borrow for schooling should
be based on the benefits of schooling, and not on the earnings level at the beginning of a
career. Any proposal aimed at helping students make smart decisions about investments in
education should compare the costs of schooling to the gains that accrue over the full career
as a result of that schooling. It should not compare costs to the level of earnings of recent
graduates.
The 8 Percent Debt Limit
Consider the 8 percent limit on debt payments. First, no scientific or data-driven rationale has
been presented for an 8 percent limit as opposed to any other number. No evidence has
been presented, for example, that loan default rates increase dramatically as student loan
payments cross this threshold. And though no specific rationale has been given for the 8
percent number, to the extent that it is based on rules of thumb from other types of borrowing,
such logic would also be flawed. Borrowing for the purpose of education is inherently
different from borrowing to purchase a home because the economic returns tend to be much
larger. Education is a source of social mobility, so any restriction on student borrowing will
tend to be harmful to students who would have enjoyed large gains from the schooling
investment.
Second, it cannot logically make sense to say that the average student cannot afford to pay 8
percent of her annual earnings to cover student loans for 10 years if those loans paid for
education that raised her earnings more than 8 percent each year for the rest of her working
life. Academic studies consistently find that students with more education on average earn
more, and that these returns to education are large on average, for example, in 2-year
Associate degree programs. Academic estimates of the return to education -the increase in
annual earnings that result from each additional year of schooling - have ranged from 8 to 15
percent, and have been rising steadily over the past 30 years. Our best estimate is that the
return per year of Associate degree schooling is currently more than 1 0 percent. This means
that those who earn a 2-year degree earn between 20 and 25 percent more each year than
those with just a high school education. If the average return to a year of schooling is 10
percent, this means that the average student in a 2-year program could pay 8 percent of her
earnings just with the additional earnings due to the schooling, even after accounting for
taxes.
These average benefits also do not include additional economic benefits of schooling.
Additional benefits of education include increased employment rates, increased likelihood of
health insurance coverage, decreased criminality and improved health.
Page 3
Report on Gainful Employment: Executive Summary
March 29, 2010 Charles River Associates
25th Percentile of Earnings
Many features of the proposed regulation share the flaw that limits on borrowing are based on
the estimated earnings that graduates would experience in the early years after graduation.
This is a flaw because the benefits of education accrue over a long period of time. A policy
aimed at protecting students would compare the benefits of education and the costs of
education. A key feature of education is that the costs are paid up front, both in terms of
foregone earnings and tuition, and the benefits accrue over the entire working life. To focus
exclusively on the short-term benefits is to ignore the long-term benefits and to cause
students to under-invest in education. This under-investment would likely be harmful to many
students affected by the proposed regulation.
One example is the use of the 25
1
h percentile of earnings as an implied ability to repay loans.
The proposed regulation places a limit on the median debt among students at a program.
This limit is based on the 25th percentile of earnings in occupations for which that program
prepares students. Presumably, the 25th percentile is meant to be an estimate of the typical
starting salary of graduates of the program. The use of the 25th percentile is flawed for
multiple reasons. Most fundamentally, the premise of limiting borrowing for education based
on early-career earnings is inappropriate and would be harmful to low-income students who
rely on student loans for access to education beyond high school.
In addition, the choice of the 25th percentile appears to be ad hoc, possibly based on the fact
that it is one of the statistics that the Bureau of Labor Statistics publishes for each occupation.
However, there is no scientific basis for using the 25th percentile of earnings as an estimate of
the early career experience of workers rather than some other percentile (and we reiterate
that limits should not be based on the level of earnings of early career workers in the first
place, but rather an estimate of the individual gains to earnings that result from the
education) . Other percentiles could be calculated fairly easily based on the same data used
by the Bureau of Labor Statistics.
Furthermore, the particular way to calculate the 25th percentile is not innocuous. Small
changes in the way one calculates this number causes large differences in the estimate of
early career earnings. One concern would be that future changes in the method of
calculating this number could have serious consequences. We estimate that differences in
earnings levels resulting from changes in how the 25th percentile is calculated would lead to
large differences in the number of students impacted by the proposed regulation. This
suggests that further consideration should be given to: (a) whether the 25th percentile concept
is appropriate, and (b) whether the method of calculating the student's estimated ability to pay
is overly sensitive to small changes in the future and valid from a scientific standpoint.
1 0-year Repayment
The proposed regulation indicates that annual loan payments should be based on a 1 0-year
repayment period. The use of the 1 0-year repayment length is another way that the
Page4
Report on Gainful Employment: Executive Summary
March 29, 2010 Charles River Associates
regulation would overweight the early costs of education and ignore the future benefits. As
mentioned above, the benefits of additional schooling accrue for the entire working career,
which for most students lasts significantly longer than 1 0 years, in many cases three times as
long. By calculating loan payments using a 1 0-year repayment, the regulation in a sense
imposes a decision rule on students that weights the costs of schooling two- to three-times as
much as the total benefits of schooling. Overweighting of costs in this way would be
shortsighted if it were done by students making education decisions. Any regulation that
imposes such a decision rule on students would cause many students who would otherwise
benefit from additional education to be harmed.
Further Analysis of the Methodology Used to Compute the Metrics of the
Regulation
The proposed regulation is based on a formula that has many moving parts. For each piece
of the debt limit formula, decisions must be made that determine the precise way the value
will be calculated. These decisions are not all straightforward, can have large effects on the
impact of the regulation, and have not to our knowledge been based on scientific research or
subject to sensitivity analyses. One prominent example is the method used to calculate the
25th percentile of earnings. While such a statistic sounds straightforward, it is not. Most
areas of study prepare students for multiple occupations, as defined by the Bureau of Labor
Statistics. There is not a singular method for calculating the 25th percentile of earnings
among workers entering multiple occupations. The method proposed by the Department of
Education during Negotiated Rulemaking indicates that a weighted average of 25
1
h
percentiles within each of the occupations should be used. Taking a weighted average of
25th percentiles within occupations does not, however, give the 25th percentile of earnings
among the workers in those occupations. Take, for example, the case of Business,
Management and Marketing. One of the occupations for which that area of study prepares
students, according to the Department of Education, is Chief Executive. Thus, the
Department of Education's method bases the early career earnings of students finishing
business management degrees in part on the 25
1
h percentile of earnings of chief executives.
We suspect that a recent college graduate has a vanishingly small chance of earning a chief
executive salary in the first few years after finishing school, though some will become chief
executives later in their careers.
Assuming the Department's goal is to generate an estimate of the early earnings of a
program's graduates, there are some types of occupations for which taking the weighted
average of percentiles may make sense, and others for which it does not. Even when a
weighted average may be appropriate, how to weight is an important question that must be
addressed. There is not a simple solution to the general problem of how to estimate the
future earnings of a program's graduates. and more thought and study should be given to
how to address it. One risk that should be addressed is that future decisions about which
occupations for which a program prepares students could have very large effects on the
resulting borrowing limits for that program. Since these decisions may be made for other
Page 5
Report on Gainful Employment: Executive Summary
March 29, 2010 Charles River Associates
reasons, or capriciously, the regulation would be at risk of having unintended consequences
based on future seemingly unrelated decisions.
There has also been no guidance as to what defines a program for the purpose of the
proposed regulation. For our analyses, we define a program as an area of study (i.e. a CIP
code) , at a particular location or campus, for a specific program length (i.e. less than 2-year,
2-year, or 4-year). An alternative would be to define a program based on an institution's
OPEID code rather than a campus. Doing so could create unintended incentives to combine
campuses into OPEID codes. This is another indication that the proposed regulation's
implementation at the school or institution level, rather than at the student level, is
problematic.
CONCLUSION
In summary, the proposed regulation is not currently formulated to address a specific problem
effectively. Secretary of Education Arne Duncan has stated publicly that he wants to ensure
that the Department understands its proposal thoroughly so that it prevents any "unintended
consequences." Our analysis suggests that the "unintended consequences"-cutting off
access to hundreds of thousands of students who want postsecondary education-will be
much more substantial than the intended consequence, which we believe to be-though we
are not certain-reducing the number of students who over borrow.
To start, the Department of Education has not clearly defined what the problem is that the
regulation aims to address. As discussed above, some perceived problems the regulation
may intend to address are not problems at all but rather a reflection of the fact that for-profit
postsecondary schools serve a very different population than not-for-profit postsecondary
schools. If the Department of Education wishes to address the problem that some students
take on excessive debt, the proposed regulation is not well designed to do so. By applying a
rule at the school or program level, many other students would be negatively affected. Our
analysis suggests that 33 percent of students currently in for-profit postsecondary schooling
would be denied access. Many more students would be denied access to postsecondary
schooling than would be protected from excessive borrowing.
Furthermore, it should not be assumed that public postsecondary institutions, particularly
community colleges, would absorb these students. Given the fiscal conditions of the states, it
is not obvious that community colleges will be able to increase capacity to meet the
increasing demand for postsecondary schooling.
Finally, because for-profit schools disproportionately serve racial and ethnic minority students
and students from low-income family backgrounds, the regulation would have the effect of
reducing access to higher education to groups of students that have historically had the
lowest levels of access.
Page6
PROPOSED DISCLOSURE ELEMENTS
AS THEY RELATE TO CERTAIN EDUCATIONAL PROGRAMS
Prospective students should have sufficient information made available to them at the time of the
enrollment process such that they can make an informed decision regarding their enrollment.
Information should include the cost of their program of study, a reasonable projection of
potential earnings in their chosen field upon graduation and throughout their life of employment
in that field, a reasonable estimate of the debt students typically incur to complete their program,
and the options they will have in selecting plans for repaying that debt. Thjs information should
be made available to the prospective student before incurring any financial obligation to the
school. This disclosure would substitute for the debt to earnings ratio proposed to define the
term "gainful employment" as used in provisions of Sections 1001 and 1002 ofthe HEA.
Under the Higher Education Act as most recently reauthorized, prospective students and
borrowers are required to be given much of this information. Among the information that
schools will be required to provide are:
The price of attendance, including tuition and fees, estimated costs of books and
supplies, room and board, and transportation costs, and any additional costs for a
program in which the student is enrolled or expresses an interest;
A net price calculator;
The prices for books;
Detailed financial aid information;
The retention rate of certificate- or degree-seeking, first-time, undergraduate students
as reported to IPEDS;
The completion or graduation rate of certificate- or degree-seeking, first-time, full-
time, undergraduate students; including data disaggregated by gender; major racial
and ethnic subgroup (as defined in IPEDS); recipients of a Federal Pell Grant;
recipients of a subsidized Stafford Loan who did not receive a Pell Grant; and
students who did not receive either a Pel! Grant or a subsidized Stafford Loan;
The placement in employment of, and types of employment obtained by, graduates of
the institution's degree or certificate programs;
The types of graduate and professional education in which graduates of the
institution's four-year degree programs enroll;
Information published by the Department of Education for students at any time that
information regarding loan availability is provided;
Detailed entrance and exit counseling for student borrowers, including average
anticipated monthly repayment amount, repayment plan options, and debt
management strategies; and
Detailed private education loan disclosures, including a student self-certification
form.
The additional information proposed to be disclosed here would complete the information for
prospective students by providing additional disclosure on the occupations which the student
may pursue based on his/her educational program, information relating to salary levels for those
occupations, average student loan indebtedness of graduates of the program, and projected
annual loan repayment amounts for the average level of indebtedness. Regulations pertaining to
the mjsrepresentation of information would apply to these disclosures.
668.41 Reporting and disclosure of information
(h) Prospective students - report on employment obtained by graduates of certain degree and
certificate programs. For a program offered by the institution under 668.8(c)(3) or (d), the
institution must provide a notice to prospective students, prior to the student enrolling in or
entering into a financial obligation to the institution, including but not limited to the following
information:
(i) identification of one or more occupations for which the program helps the student
prepare;
(ii) annual wage and salary information reported at the 25
111
and 75
1
h percentile for the
identified occupation(s) from the Department of Labor' s Occupational Information Network
(O*NET) or a link to O*NET with an explanation that the prospective student can find labor
market and wage and salary information on that site relating to employment in various
occupations;
(iii) wage and salary data for graduates from the most recently completed year for which
data are available, if the institution collects such data for the purposes of this section;
(iv) average federal student loan indebtedness of graduates of the institution with respect
to their attendance at that institution, on a program, degree-level , or institution-wide basis;
(v) average institutional loan indebtedness of graduates of a program, degree-level , or
institution-wide basis, if the institution provides institutional loans to its students as defined in
668.28(a)(5)(i);
(vi) the percentage of graduates who borrowed private student loans with respect to their
attendance at that institution, on a program, degree-level, or institution-wide basis; and
(vii) the expected annual loan repayment amounts for the average federal and
institutional student loan indebtedness, on a standard 10-year repayment plan and at least one
other repayment plan.
[Note - we would request preamble language on several issues, including the following:
1. For a program that does not have a SOC code identified to it or salary information
to a SOC code on the O*NET site, the institution could provide some alternative
information. An institution could simply provide the CIP code for the program and a list of SOC
Codes that are associated with that CIP Code.
2. The institution could identify several occupations but must identify the one that the
largest percentage enter into, if the institution collects that information.
3. Institutions need to be able to clarify for students that the average data provided may
not be accurate for their particular sUuation.
4. For pwposes of allowing institutions to provide actual salary data if they collect it, it
is important not to require institutions to disclose any information they may collect, as it may not
be comprehensive enough to give good consumer information.]
2
From: Jenkins, Harold
To: Woiff,Russell
CC: Finley. Steve
Date: 4/2/2010 9:10:00 AM
Subject: FW: Request for Follow Up Meeting This Week to Discuss Gainful Employment
(b)(S)
Harold
From: Jenkins, Harold
Sent: Friday, April 02, 2010 9:03AM
To: Manheimer, Ann
Cc: Yuan, Georgia; Wolff, Russell; Finley, Steve
Subject: RE: Request for Follow Up Meeting This Week to Discuss Gainful Employment
l(b)(5)
Harold
From: Manheimer, Ann
Sent: Friday, April 02, 2010 8:56AM
To: Harris Miller
Cc: Brian Moran; Tammy Halligan; Bergeron, David; Madzelan, Dan; Kanter, Martha; Shireman, Bob; Wolff, Russell;
Jenkins, Harold; Madzelan, Dan; Picoult, Francine; Finley, Steve; Yuan, Georgia; Smith, Kathleen; Arsenault, Leigh
Subject: RE: Request for Follow Up Meeting Ths Week to Discuss Gainful Employment
Harris- from the Office of the Under Secretruy, Bob Shireman, Leigh Arsenault and I will be attending; from the Office
ofPostsecondruy Education, Dan Madzelan and David Bergeron will be attending, and from the Office of General
Counsel, Harold Jenkins, Steve Finley, Russ Wolff, Steve Finley, and possible, Georgia Yuan, will be attending. We
look forward to hearing your presentation- Ann Manheimer, 260-1488
From: Harris Miller [ mailto:HanisM@career.org]
Sent: Thursday, April 01, 2010 6:04PM
To: Shireman, Bob; Kanter, Martha; Madzelan, Dan; Bergeron, David
Cc: Brian Moran; Tammy Halligan; Manheimer, Ann
Subject: RE: Request for Follow Up Meeting This Week to Discuss Gainful Employment
Bob, et.al.: We look forward to the meeting Monday morning at 9 AM at the Department to discuss the gainful
employment issue. Attending as part of the CCA delegation will be, in addition to me, Brian Moran, CCA's Executive
Vice President for Government Relations, Dr. Jonathan Guryan, University of Chicago, and Dr. Matthew Thompson of
Charles River Associates, the leads on the research we will be presenting, Rebecca Campoverde from Kaplan, Anthony
Guida from EDMC, and Tom Babel from Devry.
If you could please let me know who will be in attendance from the Department, that would be appreciated.
Attached you will find three documents. The first is the Executive Summary of the research findings. I will send you the
full report tomorrow, Friday. Last minute editing on the full report was delayed by the Passover holiday.
Second is the proposal CCA is putting forward as an alternative to the gainful employment proposal put forth by the
Department during the negotiated rulemaking.
Third is a document that summarizes the extensive tools the government already has to take actions against "bad actors"
in our sector.
Please let me know if you have any questions or comments. I look forward to a productive conversation Monday. Enjoy
the spectacular weather.
Harris
From: Shireman, Bob [mailto:Bob.Shireman@ed.gov]
Sent: Monday, March 29, 2010 10:48 AM
To: Harris Miller; Kanter, Martha; Madzelan, Dan; Bergeron, David
Cc: Brian Moran; Tammy Halligan; Manheimer, Ann
Subject: RE: Request for Follow Up Meeting This Week to Discuss Gainful Employment
Work with Ann Manheimer on arranging the meeting. Please send your research and alternative prior to the meeting so
we can get concrete and specific in the discussions.
Thanks,
-Bob
From: Harris Miller [mailto:HarrisM@career.org]
Sent: Monday, March 29, 2010 10:04 AM
To: Kanter, Martha; Shireman, Bob; Madzelan, Dan; Bergeron, David
Cc: Brian Moran; Tammy Halligan
Subject: Request for Follow Up Meeting This Week to Discuss Gainful Employment
Martha/Bob/Dan/David:
CCA would like to have a follow on meeting with you this week to the one Brian Moran and I had with Dan and David
ten days ago to discuss a) the results of our research (which was not complete when we met) on the impact of the
Department's gainful employment proposal on students in higher education, and b) an alternative (which we discussed in
general concept with Dan and David) to the Department of Education proposal. We can make ourselves available at
almost any time starting tomorrow. Please let me know whom I should contact to arrange a meeting. Thanks in advance.
Harris
*****************************************************************
Harris N. Miller
CEO/President
Career College Association
110'1 Connecticut Avenue, NW, Suite 900
Washington, DC 20036
hanism@career.org
+ 1 202 336 6754
Executive Assistant: Jackie McWilliams
j ackiem@career. org
+1 202 336 6706
www.career.org
From: Jenkins, Harold
To: Matinyqci, FrM
Sann Ronald
Woodward, Jennifer
CC:
Date: 4/2/20'10 1:37:44 PM
Subject; FW: Request for Follow Up Meeting This Week to Discuss Gainful Employment
FYI.
From: Harris Miller [ mailto:HanisM@career.org]
Sent: Friday, April 02, 2010 1:26PM
To: Manheimer, Ann
Cc: Brian Moran; Tammy Halligan; Bergeron, David; Madzelan, Dan; Kanter, Martha; Shireman, Bob; Wolff, Russell;
Jenkins, Harold; Madzelan, Dan; Picoult, Francine; Finley, Steve; Yuan, Georgia; Smith, KathJeen; Arsenault, Leigh
Subject: RE: Request for Follow Up Meeting This Week to Discuss Gainful Employment
Ann: Thank you very much. Attached is the full research report for which I sent you and your colleagues the Executive
Summary yesterday. I realize it is a holiday weekend, but I encourage those attending the meeting to read it, if at all
possible. It is excellent work done by top notch scholars. I look forward to seeing you Monday. Harris
From: Manheimer, Ann [mailto:Ann.Manheimer@ed.gov]
Sent: Friday, April 02, 2010 8:56AM
To: Harris Miller
Cc: Brian Moran; Tammy Halligan; Bergeron, David; Madzelan, Dan; Kanter, Martha; Shireman, Bob; Wolff, Russell;
Jenkins, Harold; Madzelan, Dan; Picoult, Francine; Finley, Steve; Yuan, Georgia; Smith, Kathleen; Arsenault, Leigh
Subject: RE: Request for Follow Up Meeting This Week to Discuss Gainful Employment
Harris - from the Office of the Under Secretary, Bob Shireman, Leigh Arsenault and I will be attending; from the Office
of Postsecondary Education, Dan Madzelan and David Bergeron will be attending, and from the Office of General
Counsel, Harold Jenkins, Steve Finley, Russ Wolff, Steve Finley, and possible, Georgia Yuan, will be attending. We
look forward to hearing your presentation- Ann Manheimer, 260-1488
From: Harris Miller [mailto:HanisM@career.org]
Sent Thursday, April 01, 2010 6:04PM
To: Shireman, Bob; Kanter, Martha; Madzelan, Dan; Bergeron, David
Cc: Brian Moran; Tammy Halligan; Manheimer, Ann
Subject: RE: Request for Follow Up Meeting This Week to Discuss Gainful Employment
Bob, et.al.: We look forward to the meeting Monday morning at 9 AM at the Department to discuss the gainful
employment issue. Attending as part of the CCA delegation will be, in addition to me, Brian Moran, CCA's Executive
Vice President for Government Relations, Dr. Jonathan Guryan, University of Chicago, and Dr. Matthew Thompson of
Charles River Associates, the leads on the research we will be presenting, Rebecca Campoverde from Kaplan, Anthony
Guida from EDMC, and Tom Babel from Devry.
If you could please let me know who will be in attendance from the Department, that would be appreciated.
Attached you will find three documents. The first is the Executive Summary of the research findings. I will send you the
full report tomorrow, Friday. Last minute editing on the full report was delayed by the Passover holiday.
Second is the proposal CCA is putting forward as an alternative to the gainful employment proposal put forth by the
Department during the negotiated rulemaking.
Third is a document that summarizes the extensive tools the government already has to take actions against "bad actors"
in our sector.
Please let me know if you have any questions or comments. I look forward to a productive conversation Monday. Enjoy
the spectacular weather.
Harris
From: Shireman, Bob [ mailto:Bob.Shireman@ed.gov]
Sent: Monday, March 29, 2010 10:48 AM
To: Harris Miller; Kanter, Martha; Madzelan, Dan; Bergeron, David
Cc: Brian Moran; Tammy Halligan; Manheimer, Ann
Subject: RE: Request for Follow Up Meeting This Week to Discuss Gainful Employment
Work with Ann Manheimer on arranging the meeting. Please send your research and alternative prior to the meeting so
we can get concrete and specific in the discussions.
Thanks,
-Bob
From: Harris Miller [mailto:HarrisM@career.org]
Sent: Monday, March 29, 2010 10:04 AM
To: Kanter, Shireman, Madzelan, Bergeron, David
Cc: Brian Moran; Tammy Halligan
Subject: Request for Follow Up Meeting This Week to Discuss Gainful Employment
Martha/Bob/Dan/David:
CCA would like to have a follow on meeting with you this week to the one Brian Moran and I had with Dan and David
ten days ago to discuss a) the results of our research (which was not complete when we met) on the impact of the
Department's gainful employment proposal on students in higher education, and b) an alternative (which we discussed in
general concept with Dan and David) to the Department ofEducation proposal. We can make ourselves available at
almost any time starting tomorrow. Please let me know whom I should contact to arrange a meeting. Thanks in advance.
Harris
*****************************************************************
Harris N. Miller
CEO/President
Career College Association
1101 Connecticut Avenue, NW, Suite 900
Washington, DC 20036
harrism@career.org
+ l 202 336 6754
Executive Assistant: Jackie McWilliams
j acki em@career. org
+ 1 202 336 6706
www. career.org
C
D A ( River
.l'l""\.. As.suc1arcs
Prepared For:
Harris N. Miller
Career College Association
1101 Connecticut Ave. NW, Suite 900
Washington DC 20036
Report on
Gainful Employment
Prepared By:
Jonathan Guryan, Ph.D.,
Associate Professor of Economics, University of Chicago Booth School of Business
Matthew Thompson, Ph.D.,
Vice President
Charles River Associates
1545 Raymond Diehl Road, Suite 260
Tallahassee, FL 32308
Date: April 2, 2010
Report on Gainful Employment
April 2, 2010
Disclaimer
This report was prepared at the request of Harris Miller and the Career College Association. It is
based on data and information that were available at the time of the analyses. If additional data
or information become available we may update or modify our report.
Page i
Report on Gainful Employment
April 2 , 2010
Table of Contents
1. WHAT ARE POSSIBLE RATIONALES FOR THIS PROPOSAL? WHAT PROBLEM IS
THE PROPOSED REGULATION AIMING TO SOLVE? .... .... .... .... .... .... .... .... ..... .... .... .... 1
2. ARE THE PROBLEMS AS SEVERE AS ASSUMED? .................................................... ?
2 .1. COMPARING STUDENT CHARACTERISTICS .... .... .... ..... ... .... .... .. .. .... .... .... ..... ... .... .... .. .. .... .... .... 8
2 .2. DEFAULT RATES 14
3. WHAT IS THE RATIONALE FOR SUBSIDIZING LOANS FOR HIGHER EDUCATION?16
3.1. WHAT ARE THE BENEFITS OF HIGHER EDUCATION? .. .... ..... ... .... .... ..... .... ..... ... .... .... .. .. .... .... .. 17
3.2. PRESIDENT 0BAMA RECOGNIZES THE IMPORTANCE OF EDUCATION BOTH AS AN ENGINE OF
ECONOMIC GROWTH FOR THE COUNTRY AND AS A SOURCE OF ECONOMIC PROGRESS FOR
INDIVIDUALS FROM ALL CORNERS OF THE U.S. SOCIAL STRUCTURE ... .. .. .... ..... ... .... .... .... .... ... 18
4. WHAT ARE THE BENEFITS OF AN ASSOCIATE DEGREE EDUCATION IN
PARTICULAR? ... ..... ..... ... .... .... .... .... .... ..... .... .... .... ... ..... ..... .... .... .... .... ..... ..... ... .... .... .... ... 20
4 .1. THE PROPOSAL'S FOCUS ON ANNUAL INCOME IGNORES THE OTHER BENEFITS OF EDUCATION:21
5. WHAT EFFECTS MIGHT THE PROPOSED REGULATION HAVE? .... .. ..................... 24
5 .1. WHAT ARE THE EFFECTS OF INCREASED ACCESS TO FUNDING FOR HIGHER EDUCATION AND HOW
DO THESE VARY ACROSS DIFFERENT TYPES OF STUDENTS? .. .... .... .. .. .... .... .... ..................... .. 24
5.2. HOW TO MEASURE THE 25 PERCENTILE OF EARNINGS? .... .. .. .... ........... ............ ... .. .... .... .... . 24
5.3. PRELIMINARY ANALYSIS BASED ON DATA SUBMITTED BY CCA MEMBER INSTITUTIONS .... .... ... 30
6. WHAT ALTERNATIVE REGULATIONS OR POLICIES MIGHT BE SUGGESTED TO
ADDRESS THE PROBLEM AT HAND? ...... ........ ...... .................................. ........ .......... 36
6 .1. FURTHER CRITICISMS OF THE PROPOSED REGULATION ........................................................ 36
6 .2. TO DIRECTLY ADDRESS THE PROBLEM, REGULATION SHOULD FOCUS ON DISCLOSURE AND
ENSURING THAT STUDENTS MAKE INFORMED DECISIONS .... .... .... .. .. .... ..... .................... ..... .... 37
7. CONCLUSION .... .... .... .... .... ............. .... .... ........ ..... .... .... .... .... .... .... ..... .... .... .... .... ......... ... 38
8. APPENDIX A ...... .... .... ..... .... .... .... .... .... .... .... .... .... .... .... ..... .... .... .... .... .... .... .... .... .... .... .... . 39
Page ii
Report on Gainful Employment
April 2, 2010
This report addresses the definition of"gainful employment" proposed by the U.S. Department of
Education as a part of negotiated rulemaking. At present the proposal would define gainful employment
so that a program's students would be required to have a median debt level no greater than 8 percent of
the 25th percentile of annual earnings among individuals working in occupations for which that program
prepares students. The 25th percentile of earnings would be calculated from Bureau of Labor Statistics
(BLS) data, and is presumably meant to be an estimate of the typical starting annual earnings for
someone finishing that program. Annual loan payments would be calculated from the median debt level
based on a 1 0-year repayment plan using the interest rate on unsubsidized Stafford loans. For programs
that do not meet the 8 percent loan-to-income cut-off, an alternative is to have a 90 percent repayment
rate for recent graduates.
In this document, six basic areas are covered:
1. What are possible rationales for this proposal? What problem is the proposed regu-
lation aiming to solve?
2. Are the problems as severe as assumed?
3. What is the rationale for subsidizing loans for higher education?
4. What are the benefits of an Associate degree education in particular?
5. What effects might the proposed regulation have?
6. What alternative regulations or policies might be suggested to address the problem
at hand?
1. WHAT ARE POSSIBLE RATIONALES FOR THIS PROPOSAL?
WHAT PROBLEM IS THE PROPOSED REGULATION AIMING TO
SOLVE?
Presumably, the motivation behind the proposed regulation is to protect students from taking on "too
much debt". Taking on excessive debt may lead to an inability to repay the debt, resulting in default.
There may be a belief that some students agree to borrow so much there is little chance they will be able
to repay the loan in the future. There may also be a belief among policymakers that, regardless of
whether the loans are eventually paid back, some amounts of debt are too high per se.
1
The proposal's focus on for-profit schools implies that there is a belief that the problems of excessive debt
burden and high default rates are either specific to, or more severe at, for-profit schools. There has been
no analysis of whether differences in debt levels or differences in default or delinquency rates across
types of schools are the result of actions by the schools or due to differences in the types of students that
the schools serve.
1
Further, the proposed legislation assumes that individuals do not have the ability to determine appropriate levels of personal debt without
government guidance. One argument that the Department of Education may advance is that students do not have all of the
necessary information to make informed decisions, and thus government guidance is necessary.
Page 1
Report on Gainful Employment
April 2, 2010
Later in this document, we address whether in fact it appears that default rates are higher at for-profit
schools, and whether some of this difference is a result of serving students from different backgrounds.
The focus on for-profit schools suggests that another possible motivation for the proposed regulation is
that the Department of Education believes the cost offor-profit programs is too high. This sense that the
cost to students is greater at for-profit programs is surely based almost exclusively on a comparison of
tuitions. However, the full cost of schooling must also include foregone earnings if the student must
cease working to attend school, and other less obvious but very real costs. Since for-profit programs tend
to offer more flexibility both in terms of the timing and location of schooling, these types of costs tend to
be lower. Compare the costs for a student currently earning $30,000 per year who could continue to work
while completing a 2-year for-profit program, but who would have to stop working to attend a community
college program because it conflicts with his work schedule. Even if the tuition at the community college
were significantly less, the total cost to the student (tuition plus any foregone earnings) is likely lower at
the for-profit program.
It may be useful to discuss why policy makers should think of high default or debt levels as being
something students should be protected from. Consider high default rates. The negative effects of loan
default may include future difficulty securing loans. Without these costs, a defaulted loan is similar, from
the student's standpoint, to a grant. If there were no long-term penalty from defaulting it would be in the
student's interest to borrow monies he will not pay back. Thus, to the extent that the regulation's intent is
to protect students, it should be made clear that it should protect students from the penalties associated
with default, not from the funding stream that makes an education possible. One might imagine an
alternative regulation that was aimed at reducing the number of students who strategically take out loans
with no intention of repaying. But, it would seem that such a regulation would focus on the process by
which students are approved for loans, and on lenders rather than schools.
Next consider the concept of "too much debt". It is important to consider the purpose of the debt before
deeming it excessive. As we will discuss later in this document, the standard economic analysis of higher
education treats it as an investment. Since education so consistently yields high returns in the form of
increased earnings, lower unemployment rates, lower crime rates and even better health and longevity, it
can be a smart worthwhile investment to borrow even large amounts to be educated.
The question of how much debt is too much can be answered in different ways. The Department of
Education proposal focuses on the ability to make the associated loan payments relative to annual
income. Another way to view the decision is to ask whether taking on the debt and getting the education
increases the present value of a student's lifetime earnings. Those with more education tend to earn
more per year. This is of course a benefit.2 This benefit should be weighed against the costs. A
significant cost of education typically is to forego earnings while in school. The other main cost is tuition.
In Table 1 below, we calculate how much debt a student can take on such that comparing all of the costs
and benefits getting more education leads to an increase in lifetime earnings net of the debt costs.
Before we turn to these calculations, it may be helpful to consider both the benefits of education and the 8
percent proposed debt limit together. It is widely accepted among academic economists that each
2 There are other benefits of education that will be discussed later. The following calculation is conservative in that it ignores non-income
benefits of education.
Page2
Report on Gainful Employment
April 2, 2010
additional year of education increases earnings by some percentage. This percentage has varied from
about 7 to 15 percent over the past 40 years. Recent estimates peg this number currently near 15
percent. This means each year of schooling causes a student's annual earnings to be higher by 15
percent every year that she works. In the first year after school, her earnings are 15 percent higher than
they would be if she had not gone to school.
It cannot make sense to have a limit on debt payments that is less than the earnings return to
education. It does not make logical sense to say that she cannot afford to spend more than 8 percent of
her earnings to have that 15 percent bump in pay. If the government "protects" her from making these
debt payments, she loses the 15 percent annual bonus, and her expenditures are 8 percent lower. It
does students no favors to decrease their earnings by more than you decrease their required
expenditures. If she could not afford to make the debt payments, she certainly cannot afford not to make
them. She has less discretionary income in the world without the debt payments than she does in the
world with the debt payments.
This argument ignores the other costs of education, namely the foregone earnings while in school. Thus,
this argument does not imply that it is always a good investment to pay up to 15 percent of your income
per year to finance the tuition for a year of schooling. However, it very clearly shows that it is wrong to
say that someone is unable to pay more than 8 percent of her annual income to finance schooling. Thus,
to the extent that the proposal is meant to protect students from taking on debt payments they cannot
afford, it is misguided.
Here we consider not just whether students can afford these levels of debt, but how much one should be
willing to borrow to finance a 2-year program if the goal is to maximize lifetime earnings net of costs.
Each row of Table 1 shows the calculation for slightly different situations. In all cases, we consider a
student deciding whether to get two additional years of schooling at age 18. Consider the first row. Here
we evaluate the choice of a student who would earn $30,000 per year with a High School Degree, and
$34,992 per year if she spends two years earning her Associates Degree. This corresponds to an 8
percent increase in earnings for each year of schooling (i.e. an 8 percent "return to education"). This is a
fairly conservative estimate that may have been appropriate 30 years ago before returns to education
increased so dramatically. Column 2 denotes the rate at which the student discounts earnings and costs
that will come in the future. Much of the benefits of education come far in the future so how much these
are discounted are important. Column 3 shows the increase in lifetime earnings associated with the extra
education, in net present value. In other words, this is how much more someone with an Associate
Degree will earn over her lifetime than someone with a High School Degree, properly discounting to
account for the fact that much of the benefits will come many years in the future. Column 4 shows the
annual loan payments associated with the maximum debt someone could take on to cover tuition costs
and still not erase the amount in column 3. Column 5 shows the ratio of that annual debt payment to
annual earnings with an Associate Degree.
The calculation is repeated for more appropriate 1 o and 15 percent schooling returns, and for a 3 percent
discount rate. A 1 o percent return to schooling is closer to estimates of the return per year to a 2-year
Page 3
Report on Gainful Employment
April 2, 2010
college education (e.g. Kane and Rouse, 1995)3 which are based on the experiences of people who
graduated high school in the 1970s and 1980s. Estimates of contemporaneous returns to schooling tend
to be closer to 15 percent per year of education (e.g. Goldin and Katz, 2008).
Table 1: How much is really too much debt?
(1) (2) (3) (4) (5) (6) (7)
Maximum
annual debt
payments
Net present for 10 years Debt to
Return value of such that income% Annual Debt to
per year lifetime education a with max- payments if income%
of school- Discount earnings good in- imum 10 paid off over if paid in
ing rate difference vestment year loan 20 years 20 years
8% 5% $24,696.74 $3,135.44 9.00% $1,904.92 5.40%
10% 5% $45,783.80 $5,812.60 16.00% $3,531.42 9.70%
15% 5% $100,194.22 $12,720.42 35.00% $7,728.23 21 .30%
8% 3% $57,967.21 $6,786.99 19.40% $4,471.15 12.80%
10% 3% $88,196.71 $10,326.35 28.40% $6,802.83 18.70%
15% 3% $166,197.14 $19,458.89 53.60% $12,819.19 35.30%
Note: The calculations are for an individual who would earn $30,000 per year with a high school degree and $34,992, $36,300, or
$39,675 with a 2-year associate degree and an 8, 10, or 15 percent return to education per year of schooling.
A number of things should be noted from the calculations:
The increase in lifetime earnings associated with two additional years of education can be quite
large, even for someone who would have earned $30,000 per year without that schooling. For a 10
percent return to education, using a 5 percent discount rate, the lifetime earnings benefit is more
than $45,000. This number is net of the cost of foregone earnings during the two years while she is
in school.
4
This calculation implies that someone who is trying to maximize her lifetime earnings should be
willing to pay $45,000 for those two years of education. Paying anything less than that in tuition, the
schooling will benefit her over the course of her life.
3 This paper, co-authored by current member of the President's Council of Economic Advisors (CEA) Cecilia Rouse, shows that community
college and other 2-year programs yield approximately the same returns per credit hour as 4-year colleges.
4 If an individual is able to continue working while completing her two year education this benefit increases by as much as $60,000, the current
cost of foregone earnings.
Page 4
Report on Gainful Employment
April 2, 2010
The net present value of the increase in lifetime earnings depends significantly on how future
earnings streams are discounted. The more the future is discounted, the lower the benefits of
education are. Using a 3 percent discount rate, the lifetime earnings benefit of a 2-year degree is
almost $90,000 if the return is 1 o percent per year of schooling.
The return to education matters enormously in determining how much individuals should be willing to
pay for schooling. If the return is 15 percent per year, the present value increase in earnings from a
2-year program is more than $100,000 even using a 5% discount rate.
The Department of Education proposal essentially almost completely discounts (i.e. ignores) all
future benefits of education by focusing on the ability to pay in the years immediately following the
finish of school. This is present in the choice of estimated starting salary (i.e. the 25
1
h percentile
annual earnings) as the ability to pay, in the focus on repayment rates among recent graduates, and
in the use of the 10-year repayment schedule to calculate loan payments. The proposed regulation
is misguided in that it is not a function of the benefits of education. If the returns to education were to
continue to rise, as they have for the past 30 years, students would be restricted from borrowing
more to get this valuable training. Only the children of the rich (i.e. those who could afford to pay
tuition without borrowing) would be able to get this valuable education.
Using a 5 percent discount rate, and assuming a 10 percent return to schooling, a student who would
earn $30,000 per year with a high school education would earn $45,784 more over her lifetime if she
gets an Associate Degree. This calculation accounts both for the fact that she would spend two
years early in her life earning nothing while she is in school, and that the higher earnings associated
with education will come in the future. If she is able to work while in school then the increase in
lifetime earnings is even greater.
The student described above could pay close to $45,000 in tuition for the two years of schooling and
still end up ahead. If she borrowed to cover all of these tuition costs, her annual loan payments
would be $5,813 for the ten years she spends repaying, and in this time her loan payments would be
16% of her annual earnings (double the Department of Education limit).
While it would surely be difficult to make the payments during the 1 0-year repayment period, the
calculation shows that even taking on this high level of debt is a good investment for the student.
Any restriction on borrowing that is more stringent than the levels shown in Table 1 will lead the
student described to earn less over her lifetime.
One reason the loan payments in the table may appear high, even though taking on this much debt
is a good investment for the students, is that the repayment horizon is shorter than the time during
which the benefits of education accrue. The proposed regulation's use of a 1 0-year repayment rate
is another way in which it ignores the future benefits of education. If the student were to pay back
over 20 years instead of 10, the loan payment to income ratio for the student described above would
be 9. 7 percent rather than 16 percent. For the 8 percent return to education calculation, the 20-year
loan payment to income ratio would be 5.4 percent rather than 9.0 percent.
For a 15 percent return to education, the calculations indicate that one should be willing to pay
approximately 20 percent of his income for a 20-year repayment period. The reason this increases
lifetime earnings is that 15 of that 20 percent is accounted for by the earnings increase resulting from
Page 5
Report on Gainful Employment
April 2, 2010
the schooling. For the working years after the loan is repaid, the 15 percent benefit is enjoyed with
no cost.
A major part of the costs of education considered in the calculation is the foregone earnings the
student gives up if she attends school full-time. Schools that allow students to work and earn money
while in school are therefore less costly, even if the tuition charged is the same. The calculations
above would indicate a higher debt ceiling if foregone earnings were not considered as a cost.
The calculation above assumes that the difference in earnings between college and high school
educated individuals is the same at all ages. If instead the earnings of college educated students
start similar to those of high school educated students but grow faster, the role of discounting the
future is even more important.
Illustration: If the 8 percent loan limits were applied to medical school
Doctors spend much of their early years earning pay that is lower than they will earn in the long run, but
continuing to train on the job. Much of the earnings benefits of a medical education come when doctors
are far into their career. For such occupations, restricting debt levels to an amount that can be repaid
given early-career earnings would preclude borrowing for extremely valuable investments. Furthermore,
the lifetime benefits of a medical degree are quite large. Thus, for many it is worth making the investment
of time and large sums of money to obtain the degree. This is the case even though it is typical for
doctors to leave medical school with significant debt.
Here we show that if loan payments for medical school were limited to be 8 percent of the early earnings
of doctors, medical education would be largely restricted to students who could pay tuition costs without
much borrowing. In this case doctors would largely be drawn from wealthy families. The vast majority of
racial minority students and students whose parents have less than a college education would not be
allowed to become doctors. In fact more than half of the households in the U.S. do not have a net worth
high enough to pay the amount of medical school tuition that would not be covered by loans.
To complete the calculation, we use the median medical school tuition for non-resident programs from the
Association of American Medical Colleges web site. We ignore the costs of a medical education incurred
during residency years. The median four-year tuition at medical school is $168,840. If someone financed
80 percent of this, it would require borrowing $135,072. To pay this loan back over the 1 0-year horizon at
the unsubsidized Stafford loan rates would require an annual loan payment of $19,054 (or $1 ,588 per
month). To satisfy the proposed 8 percent criteria, someone would have to earn an annual salary of
$238,173. Payscale.com indicates that the typical annual salary of a doctor with 1-4 years of experience
is, in fact, $140,000. Given these calculations, the median medical school would not meet the proposed
standards.
We can also ask who would be able to afford medical school tuition if borrowing were only allowed up to
the limit implied by the 8 percent standard. Using the $140,000 annual salary, the maximum debt that
leaves annual payments no more than 8 percent of annual earnings is $79,397. Thus, to be able to
attend medical school, students and their families would have to find the funds to cover more than half of
the tuition costs, or $89,443.
Page6
Report on Gainful Employment
April 2, 2010
Table 2: Paying for Medical School
Median non-resident medical school tuition, 2009-10
Total tuition, 4 years non-resident
Debt required to pay 80 percent of tuition
Annual loan payment with 10 year repayment
Annual earnings required to satisfy 8 percent rule
Median salary of Doctors with 1-4 years experience
Maximum allowed debt to satisfy 8 percent rule
Remaining tuition that would have to be paid without borrow-
ing
Source: AAMC.org, Payscale.com.
(http :1/services.aamc.org/tsfreports/report_ median .cfm?yea r _of_ study=201 0)
$42,210
$168,840
$135,072
$19,054
$238,173
$140,000
$79,397
$89,443
How many families in the U.S. have the ability to pay nearly $90,000 without borrowing? The Survey of
Consumer Finances, sponsored by the Federal Reserve Board, indicates that in 2007, about one-third of
U.S. households had total net worth less than $90,000. This means that students living in about this
fraction of households would be precluded from becoming doctors. Many households who have slightly
more than $90,000 in net worth hold a good deal of that wealth as housing equity, meaning that they
would have to choose between owning a home and allowing their child to become a doctor. For minority
students, the problem would be even worse. The median net worth of non-whites and Hispanics in 2007
was $28,200. Thus, the vast majority of non-white and Hispanic students would likely not be able to
become doctors, regardless of their potential for success.
2. ARE THE PROBLEMS AS SEVERE AS ASSUMED?
The Department of Education's proposal presumably aims to address a specific perceived problem. In
this section we explore whether that problem is as severe as might be thought given summary measures
comparing for-profit and not-for-profit schools.
What is the problem that the proposal aims to address? We suspect that the Department of Education
sees the following problem: (a) a significant number of students take on more debt than they can afford to
repay upon entering the labor force and (b) this problem is more severe for students who attend for-profit
schools. We infer the latter because the proposed regulation treats for-profit schools differently than not-
for-profit schools. Here we address whether it is the case that default rates are strongly related to the for-
profit status of schools. The data strongly suggest that a large portion of the difference in default rates
between for-profit and not-for-profit schools is because for-profit and not-for-profit schools serve very
different student populations. For-profit schools are more likely to serve low-income, minority students.
Page?
Report on Gainful Employment
April 2, 2010
without parental support. They are more likely to serve students who are the first in their families to
attend college. Because of their access to outside resources, these groups of students are more likely to
default at both for-profit and not-for-profit schools. We estimate that if for-profit and not-for-profit schools
served the same population of students that the default rates would be significantly closer across the
types of schools. One might effectively argue that it is bad public policy to punish institutes of higher
education for serving students from groups who historically have not had wide access to schooling.
Furthermore. in part because the student populations are different, persistence rates at for-profit and not-
for-profit schools are different. Students who complete higher education programs are more likely to find
the jobs for which those programs prepare students. Some of the difference in default rates across types
of schools is accounted for by the higher completion rates at not-for-profit schools (particularly 4-year
programs). It would also seem to be bad policy to punish schools for the decisions by students not to
complete. At the very least, such a policy punishes the students who work to complete the program by
restricting access for all students, not just those who fail to complete.
Here we lay out the data on which the former conclusions are based. First, consider the difference in
characteristics of students by the type of school attended. Specifically, the data below come from the
Beginning Postsecondary Students Longitudinal Study, 1996 cohort. This is a survey administered by the
U.S. Department of Education that follows a nationally representative sample of students who entered
postsecondary education for the first time in 1996. We focus on this cohort of students rather than more
recent data because a follow-up survey has been done that allows for measurement of default rate by this
cohort.
2.1. COMPARING STUDENT CHARACTERISTICS
A series of figures shows that the demographic and socioeconomic background of students attending for-
profit and not-for-profit are different in systematic ways. All comparisons are based on the fi rst institution
attended, so those who begin at community college and transfer to a 4-year institution are categorized as
community college students. We first examine the characteristics of students who entered postsecondary
education in 1996, because these are the students for whom we can track default rates in the NCES data
(BPS) . For comparison, we also show comparable figures (Appendix A) calculated from the 2008
National Postsecondary Student Aid Study (NPSAS:08).
Figure 1 below shows the average age at which students begin postsecondary schooling. Beginning
students at for-profit schools are on average 25 years old. Students at not-for-profit public and private
programs of 2 years or less are on average 22-23 years old. In contrast, students entering 4-year not-for-
profit colleges are significantly younger, less than 19 years old on average. This difference shows that
the set of students entering these different programs are not the same. Though most of the differences
between for-profit and not-for-profit college students have held up since 1996, this is one that has
changed. While for-profit students continue to be older when they begin, the average age difference is
much closer today than it was in 1996.
Page 8
Report on Gainful Employment
April 2, 2010
Figure 1:
Average age at which students first enroll in postsecondary schooling
26.0
25.0
24.0
(I) 23.0
0)
<U
(I)
0)
22.0
<U
Qj
~
21.0
20.0
19.0
18.0
For-profit Private not-for- Private not-for- Publi c <=2 Publi c 4-year
profit <=2 profit 4-year
Age first enrolled
Notes: Calculated from the Beginning Postsecondary Students Longitudinal Survey, 1996 Cohort.
Next, we tum to the average income of students and/or their parents (Figure 2) . Specifically, we consider
the total income of either parents (for those who are dependents) or students (for those who are
independent at the time they apply) . Students at for-profit schools have the lowest income, prior to
entering college, of any group. VVhereas the average prior family or individual income of students
entering for-profit schools was $22,165 (in 1996 dollars), it was 71 percent higher among public 2-year-or-
less students, 154 percent higher (i.e. more than 2 %times) among public 4-year students, and 178
percent higher (i.e. close to 3 times) among private 4-year students.
Page 9
Report on Gainful Employment
April 2, 2010
Figure 2:
Average income of parents or independent students prior to school entry
$70.000 ~
$60.000
$50.000
Q)
E
0
$40.000
(.)
c
Q)
0'1
!
$30.000
,_
~
<{
$20.000
$10.000
$0
For-profit Private not- Private not- Publi c <=2 Publi c 4-year
for-profit <=2 f or-profit 4-year
Average income. parents or independent. 1994
Notes: Calculated from the Beginning Postsecondary Students Longitudinal Survey, 1996 Cohort.
For-profit students do not just come from families with lower average income, they are also more likely to
come from families that are quite poor. Consider the percent of students whose families collected AFDC
(Aid for Families with Dependent Children, the precursor to TANF, and commonly known as "welfare").
Figure 3 shows a full 16 percent of students who began at for-profit schools in 1996 came from families
collecting AFDC. This was more than double the rate of students attending not-for-profit 2-year-or-less
programs. At publ ic and private not-for-profit 4-year programs the corresponding rates are 2.6 and 1.6
percent.
Page 10
Report on Gainful Employment
April 2, 2010
Figure 3:
Percent of students from families who received AFDC prior to school entry
18
16
14
_,_
12
_,_
- 10 c -
<1>
0
lii
a.. 8
~ -
6
-'-
4
_.___
2 - '- -
0
L::_::_::_::_::_::_:
1;.;.:.:-:-:-:-.' L .I
"
For-profit Pri vate not-for- Privat e not-for- Publi c <=2 Publi c 4-year
profit <=2 profit 4-year
Percent receieved AFDC. 94-95
Notes: Calculated from the Beginning Postsecondary Students Longitudinal Survey, 1996 Cohort.
Students attending for-profit schools are also significantly more likely to be single parents at the time they
begin school. A full 28.5 percent of for-profit students were single parents prior to beginning school in
1996. At not-for-profit 2-year-or-less programs, less than 12 percent were single parents upon entering.
At 4-year programs single parents were extremely rare, less than 2 percent of students (Figure 4 below).
One interesting thing that has changed is that single parents are more likely today to attend college than
they were in 1996, but they are still significantly more likely to attend for-profit schools.
Page 11
Report on Gainful Employment
April 2, 2010
Figure 4:
Percent of students who were single parents prior to school entry
30 ~
25
20
c
<ll
15
0
....
<ll
Q_
10
For-profit Private not-for- Private not-for- Publi c <=2 Publi c 4-yeal'
profit <=2 profit 4-year
Percent single parent. 1995-96
Notes: Calculated from the Beginning Postsecondary Students Longitudinal Survey, 1996 Cohort.
Students at for-profit schools are also the first in their immediate family to attend college. Figure 5 shows
the fraction of students at each school type who have at least one parent with at least some college
education. The figure shows that these rates are lower at for-profit schools, particularly when compared
with 4-year not-for-profit schools.
Page 12
Report on Gainful Employment
April 2, 2010
Figure 5:
Percent of students whose parents attended at least some college
80.0
70 0
For profit
Not for profit
r--
-- --
,__.
r--
600
,.-
500
-
f.--
r---
c
4>
40.0 e
4>
Q.
r----",.-
-- -
r---
r--
300 - - '----- c-

20.0
r - - -- - -- -- r--- --
r--
100
I - -- i- - -- -- -- -
00

r
bJj _
-
til:;.

uf;
Prrvate for Pnvate lor Prrune for Public less- Public 2-year Public 4-year Prrvale not-fo1- Pnvate not-for-
prolit,less-thao- prol>l, 2-yeat profit. 4-year than-2-year 2-year profn 4-year
2-year
Percent whose pare11ts have some college or more
Notes: Calculated from the Beginning Postsecondary Students Longitudinal Survey, 1996 Cohort.
Finally, Figure 6 shows that for-profit schools are much more likely to serve students from racial and
ethnic minority groups. The fraction of students at for-profit schools who are either Black or Hispanic was
43.9, 37.2 and 51.4 percent at less than 2-year, 2-year and 4-year programs respectively. Private not-for-
profit less-than-2-year programs also are likely to serve Black or Hispanic students (35.9 percent) .
However, the share of students who are Black or Hispanic at all other not-for-profit school groups was
less than 25 percent.
Page 13
Report on Gainful Employment
April 2, 2010
Figure 6:
Percent of students who are Black or Hispanic
60 0
500
40.0
c:
30.0
"' 0..
2().0 ...._..
100
00
For profit
Not for profit
Pmatetor
prol'lt, tess4han.
2,yea
Puva!e not-for PriVate noHot Pt<Yale not-tor Pubht tess. Publie2-year Publit .t-year
profit tess-tha.,. profit 2-year profil4-year than 2Yel
2-year

Notes: Calculated from the Beginning Postsecondary Students Longitudinal Survey, 1996 Cohort.
2.2. D EFAULT RATES
Using the same Beginning Postsecondary Students Longitudinal Survey data, we also compared 6-year
default rates of students at different types of schools. The default rate is computed as the fraction of
students in the 1996 entering cohort with any student borrowing who ever default by 2001. This is
calculated in the same way as Mark Kantrowitz did in his recent report.
The light blue bars in Figure 7 below show the raw default rates of students starting in different types of
schools. All for-profit students are considered together. Because sample sizes are small in particular
groups, less-than-2-year and 2-year schools are combined. Without adjusting for the differences in
student background across the different school groups, the 6-year default rate is significantly higher at
for-profit schools than at not-for-profit schools. At for profit schools, almost 25 percent of the 1996 cohort
borrowers defaulted on at least one loan at some point by 2001. The default rate is 17.1 percent at
private not-for-profit 2-year-or-less programs, and 8.5 percent at public not-for-profit 2-year-or-less
programs. The rates at not-for-profit 4-year programs are both around 6.3 percent.
Page 14
Report on Gainful Employment
April 2, 2010
Figure 7:
6-year default rates by type of school, controlling for student characteristics
25.0%
20.0%
15.0%
10.0%
5.0%
0.0%
For-profit Pri vate NFP 2- Pri vate NFP 4- Publi c 2-year or Publi c 4-year
year or less year less
Default rates
Default rates. controlling for race. gender. fami ly income. persi stence. Peii/AFDC receipt. single
parenthood
Notes: Calculations from the Beginning Postsecondary Students Survey: 1996 cohort and 2001 follow-up. The light bars show the
fraction of students beginning in each type of school in 1996 who ever defaulted by 2001 . The dark bars show the default rates after
controlling for race, gender, persistence and completion, Pell grant receipt in 95-96, family AFDC receipt in 94-95, parent or own
income (if dependent), and dependency status.
The dark blue bars show our estimates of what the default rates would be if all schools had similar
entering student bodies. To estimate this, we run a regression of individual students' default status on
controls for their race/ethnicity, gender, family income, dependency status, whether they persisted or
completed their program, and whether they received a Pell grant. The dark blue bars show what we
estimate the default rates would be if all schools served white male students who are dependent, whose
parents earn between $60-75K per year, who completed their programs and who did not receive a Pell
grant.
The estimates show that if all schools served similar students with similar backgrounds and who
completed their programs, the differences in default rates between for-profit and not-for-profit schools
would narrow considerably. Whereas the difference in 6-year default rates between for-profit and public
4-year schools is 18.3 percentage points (24.6- 6.3) without controlling for the differences in student
characteristics, this difference is almost cut in half to 9.6 percentage points (12.2- 2.6) when adjustments
for student characteristics are made. This measure of default also overestimates the difference because
students in shorter programs have more time to default by 2001 . Furthermore, this narrowing is what
Page 15
Report on Gainful Employment
April 2, 2010
happens when adjustments are made for the characteristics that are observed in the BPS data. One
might suspect that there are other dimensions along which for-profit and not-for-profit students are
different. It is possible that if we had data on those characteristics and were able to adjust for them that
the difference in default rates would narrow more.
In addition to narrowing the difference in default rates between for-profrt and not-for-profit students,
adjusting for student characteristics also lowers the default rate at for-profit considerably. These
estimates imply that the default rate at for-profrt schools would be cut in half (from 24.6 to 12.2 percent) if
for-profit schools served wealthier non-minority students, and students who were more likely to complete.
3. WHAT IS THE RATIONALE FOR SUBSIDIZING LOANS FOR
HIGHER EDUCATION?
The standard economic analysis of education considers the choice of an individual whether to get an
additional year of schooling.
5
In this standard way of thinking, individuals weigh the costs and benefits of
schooling. The costs are the earnings foregone if one attends school full time, and tuition/fees. The
benefits include increased earnings in future years. Individuals choose to get more education so long as
the benefits are larger than the costs.
Education is an investment, meaning that the costs are paid up front and the benefits come in the future.
To properly weigh the costs and benefits, one must discount benefits that will not be realized for many
years. To simplify things, use the interest paid on savings accounts or the expected return on personal
investments as the discount rate.
Now consider the education choice of two students: one who has enough personal or family wealth to pay
tuition costs out of savings, the other who must borrow to finance the tuition costs.
For someone who would pay tuition costs out of savings, the decision comes down to comparing the
present value of increased lifetime earnings (the benefits) to the foregone earnings while in school and
the tuition (the costs) . If the benefits are greater than the costs, then the student should continue in her
schooling. If the costs are larger than the benefits, she should end her schooling and begin working.6
Compare this decision with someone who must borrow to pay the tuition costs. This student must
consider as costs the additional interest payments associated with the loan. Those payments must be
paid in the future. If the interest rate on the loan were equal to the interest rate used for discounting (in
this case the interest paid on savings) , then the decision would be the same for both students. Since the
unsubsidized interest rate charged on student loans is typically higher than the interest rate paid on
savings accounts, the cost of furthering education is higher for this student.
5 The standard reference is Human Capital by Gary Becker (University of Chicago), who won the Nobel Prize in Economics for this and other
work.
6 Vllhile it is necessary to consider as a cost the interest she does not earn on the money she takes out of saving to pay tuition, these interest
payments are discounted because they would have happened in the future. If we use the savings account interest rate as the
discount rate, the discounting eliminates this from consideration.
Page 16
Report on Gainful Employment
April 2, 2010
In short, because borrowing interest rates are higher than savings interest rates, the cost of schooling is
higher for those who must borrow to pay for higher education. Because these students almost by
definition come from poorer families, this problem creates access differences that relate to wealth,
socioeconomic status, and race. Subsidies for student loans are meant to narrow the difference between
borrowing and saving interest rates so that the costs of education are less related to family wealth.
Any restriction of access to debt financing for higher education will have the effect of decreasing access
more for poor and minority students. This is completely at odds with the intent and spirit of the Higher
Education Act.
Notice that the economic analysis of the schooling decision does not depend on the level of earnings.
Instead, it focuses on the increase in earnings resulting from the schooling. The proposal's focus on the
ability of students to pay back their loans quickly leads it to focus on the level of earnings.
This will have the effect of differentially punishing students with poor labor market prospects and who
would gain the most from higher education. Students with poor labor market prospects would have low
earnings, and likely high unemployment rates, without any higher education. Among these students, the
ones who would benefit greatly from additional focused schooling may end up in occupations with low
earnings. But, these earnings may be much higher than the student's personal alternative. The proposal
would limit how much this student could borrow based on the low level of earnings, and not based on the
large gains that would be realized from the doors opened by education.
3.1. WHAT ARE THE BENEFITS OF HIGHER EDUCATION?
Education is widely recognized as a source of social mobility. Though the US is regarded as a "land of
opportunity," correlations in earnings between fathers and sons are actually quite high. To understand
how much social mobility there is in the U.S., consider a family of four right at the poverty threshold.
Based on the best current estimates, it would on average take the descendants 5 or 6 generations before
their income is within 5 percent of the national average.?
What's more, studies find less social mobility among families with low net worth, suggesting that the
inability to borrow restricts social mobility. In other words, restrictions on borrowing (coming from poorly
functioning credit markets and high interest rates) makes being bom into wealth or poverty quite
determinative of earnings in adulthood.
One large reason for the effect of net worth and borrowing constraints on intergenerational mobility is
likely access to schooling. As an example consider the economic progress made by African Americans
during the past century. While legislative changes such as the Civil Rights Act of 1964 have led to
significant progress in reducing discrimination, economic research suggest very strongly that
improvements in educational opportunities for blacks have been more important. Until recently
successive cohorts of blacks have obtained progressively more education, and in turn their earnings have
caught up to whites.
7 Mazumder, Bhashkar, "Fortunate Sons: New Estimates of lntergenerational Mobility in the United States Using Social Security Earnings
Data," Review of Economics and Statistics 2005.
Page 17
Report on Gainful Employment
April 2, 2010
One of the most important ways in which the labor market in the U.S. has changed in the past 30 years is
that the benefits of education have increased dramatically. Those with more education have always
earned more on average than those with less. But, the difference in earnings between those with and
without a college education has increased sharply since 1980. In 1980, 4-year college graduates earned
about 40 percent more (about 1 o percent per year of additional schooling) than comparable high school
graduates. By 2005, the benefit of a college education was more than 70 percent (almost 18 percent per
year of schooling) . It is more important than it has been since the 1920s to be educated, and more
important than ever to get education beyond high school.
The changes that have led to this dramatic increase in the monetary benefits to education have also led
to very high levels of inequality. The difference in earnings and economic well-being between the rich
and the poor is also as large as it has been since the early part of the 20
1
h century. And, how much
education you have is significantly determinative of which side of that inequality you lie on. Those with
more education have benefited from the rich getting richer in the past 30 years. Those with less
education have been hurt terribly by the poor getting poorer during that same time.
All of this argues strongly that is as important as it has ever been to assure that all students who will
benefit have access to higher education. The social costs of restricted access are larger than they have
been in almost a century.
3.2. PRESIDENT 0BAMA RECOGNIZES THE IMPORTANCE OF EDUCATION BOTH AS AN ENGINE OF
ECONOMIC GROWTH FOR THE COUNTRY AND AS A SOURCE OF ECONOMIC PROGRESS FOR
INDIVIDUALS FROM ALL CORNERS OF THE U.S. SOCIAL STRUCTURE.
In a Washington Post column, published on July 12, 2009, President Obama called for increase in 5
million students with certificate or associate degree in the next 1 o years. Here are two quotes from what
he wrote:
"In an economy where jobs requiring at least an associate's degree are projected to grow twice as fast as
jobs requiring no college experience, it's never been more essential to continue education and training after
high school. That's why we've set a goal of leading the world in college degrees by 2020. Part of this goal
will be met by helping Americans better afford a college education." Barack Obama, The Washington Post,
July 12, 2009.
we believe it's time to reform our community colleges so that they provide Americans of all ages a chance
to team the skills and knowledge necessary to compete for the jobs of the future. Our community colleges
can serve as 21st-century job training centers, working with local businesses to help workers team the skills
they need to fill the jobs of the future. We can reallocate funding to help them modernize their facilities,
increase the quality of online courses and ultimately meet the goal of graduating 5 million more Americans
from community colleges by 2020." Barack Obama, The Washington Post, July 12, 2009.
What the President calls for cannot be done without increasing the capacity of community colleges and
for profit schools. Even with increased federal support, community colleges will face funding problems as
states continue to deal with severe fiscal problems. By decreasing access to the specific programs the
President wants to increase, the proposal will make the President's goal almost surely unattainable.
The for-profit education sector will be essential in helping the President to achieve this goal. For
example, consider that enrollment at for-profit colleges has grown significantly more than at non-profit
Page 18
Report on Gainful Employment
April 2, 2010
schools. In other words, as the demand for higher education has increased in recent years- likely as a
response to the increased returns described above- the non-profit sector has not been able to meet that
need. The for-profit sector has. Consider the growth rates of enrollment in public, private not-for-profit
and private for-profrt colleges over the past 5, 1 o and 20 years.
Overall enrollment in higher education has grown by more than 2 percent per year during the past 10
years. Capacity at neither public nor private not-for-profit colleges has grown fast enough to keep up with
this increased demand. Note that some of this demand comes from natural population growth, while
some comes from an increased desire to get a college education due to the high returns. Capacity
growth in the not-for-profit (public or private) sectors has not matched the overall increase in demand for
higher education over either the last 5, 10 or 20 years.
Table 3: Five, ten and twenty year enrollment growth by type of
institution, through 2007
Private Private
Not-for- For-
Total Public profit profit
Total percent growth in enroll-
ment:
20 years
39.78% 32.80% 33.60% 438.23%
10 years
25.79%
21 .10% 18.80% 225.60%
5 years
9.85% 5.80% 9.40% 99.60%
Average annual growth rate:
20 years
1.70%
1.40% 1.50% 8.80%
10 years
2.30%
1.90% 1.80% 13.70%
5 years
1.90% 1.10% 1.80% 14.80%
Source: Digest of Education Statistics, 2008, Table 188.
During that same time, the private for-profit sector has grown to meet the needs of students not-for-profit
schools cannot serve. Enrollment growth rates have been significantly higher. Though the private for-
profit sector is smaller than the other two sectors, the significantly higher growth rates have ensured that
overall enrollment could increase.
To meet President Obama's call for 5 million more Associate degree or certificate holders from 2-year
programs, capacity will have to increase in some or all of these sectors. Fiscal difficulties in the states are
likely to restrict community colleges and state colleges from meeting this need, even with increased
funding called for by the President. Given the historical role of for-profit schools in meeting increased
demand, it is likely that these schools will be the most able to expand quickly in response to this need.
Restrictions on student borrowing will curtail the for-profit sector from meeting these needs, and will make
it less likely that the President's goal will be met.
Page 19
Report on Gainful Employment
April 2, 2010
4. WHAT ARE THE BENEFITS OF AN ASSOCIATE DEGREE
EDUCATION IN PARTICULAR?
The Bureau of Labor Statistics administers a survey each month called the Current Population Survey.
This is the survey that is used to calculate the official unemployment rate that is released each month as
a barometer of the health of the economy. In that survey, individuals are asked questions about their
employment status, earnings, educational attainment and demographic information. Since 1996,
individuals have also been asked about their access to health insurance.
The following table presents estimates of the difference in various outcomes between those with an
Associate degree and those with exactly a high school education. We look separately at "academic" and
"vocational/occupational" Associate degrees. All estimates control for individuals' years of labor market
experience, and for changes over time in the outcomes that affect all individuals in the same way.
Positive effects on annual earnings: The way to read the table is the following. The first row shows
how much more those with an Associate degree earn on an annual basis than those with a high school
degree. For example, males with a Vocational/Occupational Associate degree earn 23.2 percent more
each year than males with a high school degree. If the average Associate program were 2 years, this
would correspond to an 11 .6 percent earnings return to each year of schooling.
Table 4: The Benefits of an Associate Degree Education
Benefit of Associate degree or some college,
relative to a high school education for:
Annual earnings
Percent employed full time
Percent employed
Percent with employer sponsored health insurance
Percent with any health insurance
Males
Associate:
Vocational/ Associate:
Occupational Academic
23.20%
3.50%
2.60%
10.10%
9.10%
27.40%
2.10%
2.50%
8.70%
9.30%
Females
Associate:
Vocational/ Associate:
Occupational Academic
27.20% 30.30%
-2.00% -0.70%
0.20% 0.50%
6.40% 8.40%
5.30% 6.70%
Note: Regression estimates from the Current Population Survey. The table shows the difference in various outcomes between those with an
Associate degree and those with exactly a high school education. Two types of Associate degrees are considered, vocational/occupational and
academic. These self-reported by the respondents to the Current Population Survey according to the guidelines described by the Bureau of
Labor Statistics.
Men who go on to get a vocational Associate degree earn 23.2 percent more each year than comparable
high school graduates. The return to an academic Associate degree is slightly larger for men -a
27.4percent increase in annual earnings. For women, the returns are even higher. Women who get a
vocational Associate degree earn 27.2 percent more than high school graduate women, and women who
Page 20
Report on Gainful Employment
April 2, 2010
get an academic Associate degree earn 30.3 percent more than women who stop schooling after
completing high school.
Positive effects on employment for men: The benefits of an Associate degree extend beyond the
earnings of those who work. Men with a vocational Associate degree are 3.5 percent more likely to be
employed than high school graduates. Men with an academic Associate degree are 2.1 percent more
likely to be employed than high school graduates. For women, there is no corresponding effect on
employment. Women with Associate degrees are actually less likely to be employed full-time than high
school graduates, but equally likely to be employed. Though there is no effect on employment for
women, it appears (based on the earnings effects and the health insurance effects discussed below) that
those who work are in better jobs than they would be if they did not get the additional education.
Positive effects on health insurance: An Associate degree education also helps to reduce the number
of uninsured. There is of course great interest currently in access to health insurance. For both men and
women, those with an Associate degree are significantly more likely to have health insurance. As
compared with high school graduates, men with a vocational Associate degree are 10.1 percent more
likely to have employer provided health insurance and 9.1 percent more likely to have health insurance of
any kind. Men with an academic Associate degree are 8.7 percent more likely to have employer provided
health insurance and 9.3 percent more likely to have health insurance of any kind. Women with a
vocational Associate degree are 6.4 percent more likely to have employer provided health insurance and
5.3 percent more likely to have health insurance of any kind. Men with an academic Associate degree
are 8.4 percent more likely to have employer provided health insurance and 6.7 percent more likely to
have health insurance of any kind.
4.1. THE PROPOSAL'S FOCUS ON ANNUAL INCOME IGNORES THE OTHER BENEFITS OF EDUCATION:
Academic research has shown that there are many benefits of education beyond the large increases in
annual income.
Reduced unemployment: The proposed regulation is based on the 25
1
h percentile of earnings among
those working. But, individuals with more education are less likely to be unemployed. It is perfectly
rational for a student to be willing to pay more than 8 percent of her annual income to avoid joblessness.
Insulation from recessions: Recessions typically hit the least educated the most severely. Consider the
current recession and the unemployment rates of people with different levels of education, in January
2008, January 2009 and January 2010, shown below.
Page 21
Report on Gainful Employment
April 2, 2010
Table 5: Unemployment Rates by Schooling Level
Jan- Jan- Jan-
08 09 10
Less than HS 7.7% 12.4% 15.2%
High School 4.6% 8.1% 10.1%
Associate or Some College 3.6% 6.4% 8.5%
Bachelor or more 2.1% 3.9% 4.9%
Source: The Employment Situation, Bureau of Labor Statistics.
First, notice that even in normal economic conditions, such as January 2008, unemployment rates are
strongly related to schooling levels. In addition to earning less, those with fewer years of schooling are
much more likely to experience unemployment. This is true when comparing students with a high school
education to students with some college and/or an Associate degree.
Second, notice that the increase in unemployment rates that has happened during the current severe
recession has impacted all groups, but the less educated more than others. Those with a high school
education saw their unemployment rates increase from 4.6 to 10.1 percent over the last two years, a 5.5
percentage point increase. In comparison, those with some college or an Associate degree saw their
unemployment rate increase by 4.9 percentage points. It is typical that recessions are differentially
burdensome on the least educated.
Furthermore, the most recent employment numbers for February 201 o show that while the overall
unemployment remained steady at 9.7 percent, this obscures very different experienced for more and
less educated Americans. For those with less than a high school degree and high school graduates, the
unemployment rate rose by 0.4 percentage points (15.2 to 15.6 percent for high school dropouts; 10.1 to
10.5 percent for high school graduates) . For those with a Bachelor's degree or more, the unemployment
rate remained essentially unchanged, rising from 4.9 to 5.0 percent. The only education group for which
the unemployment rate fell significantly this month was those with an Associate degree or some college.
For this group, the unemployment rate fell by 0.5 percentage points, from 8.5 to 8.0 percent).
Increased access to health care and health insurance: As shown above, those with more education
are more likely to have both employer-provided health insurance and any health insurance at all. This
relationship is partly explained by the positive effect of schooling on the likelihood of being employed, and
partly explained by the effect of education on income. Those with more income are more likely to be able
to afford health insurance.
For this reason, and possibly because more educated people make more informed decisions regarding
the management of their own health, individuals wffh more education tend to be healthier. Their mortality
rates are lower, they are less likely to smoke, more likely to exercise, more likely to engage in
preventative care, more likely to properly manage chronic conditions such as diabetes.
Page 22
Report on Gainful Employment
April 2, 2010
"In 1999, the age-adjusted mortality rate of high school dropouts ages 28-64 was more than twice as
large as the mortality rate for those with some college (Lyert et al. 2001 , table 26)."8
Comparing across countries, those with higher average education levels have longer life-expectancy
at birth (i.e. people live longer in countries with more educated populations) .
Some argue that this relationship is just an association, that either health causes people to get more
education, or that there is a third factor that causes both health and education to rise together. There are,
however, policy experiments that suggest this relation may be causal -that increasing education may
cause improvements in health.
Various studies show that laws that require children to complete more years of schooling (increases in the
compulsory schooling age) lead to improvements in health when those kids become adults.9
Educated mothers also have healthier babies: One study shows that increases in the number of
colleges nearby increases the likelihood that women attend college, and in turn makes them more likely
to have healthy babies. The study finds that the women were more likely to obtain prenatal care, and less
likely to smoke and drink alcohol during pregnancy. They were also more likely to be married at the time
they gave birth and had fewer children.10
Reduced criminality: There is evidence that obtaining more education makes it less likely that someone
will engage in crime. A study by Lochner and Moretti (2004) finds that compulsory schooling laws reduce
the likelihood that people become incarcerated. The effect is large for whites, and even larger for blacks.
They estimate that there is an additional15-25 percent benefit to each year of education in the form of
reduced crime that is not accounted for by the increased earnings that educated people enjoy. In other
words if we consider the reduced cost of crime imposed on society because of education, the total benefit
of education should be 15-25 percent larger than the increase in earnings that results from the additional
schooling.
General fulfillment: In addition to the benefits of education that are easily measurable, surely education
and learning brings some direct satisfaction. To the extent that students enjoy learning new ideas and
new skills, these are real benefits and they are not accounted for as a benefit of education if we just focus
on monetary earnings. People buy things all the time that bring them pleasure but no monetary return.
For example, consider vacations, televisions, tickets to sporting events, clothing, food. None of these
purchases increase earnings, but no one would dispute that these are reasonable things to buy.
To point out that the proposed regulation is misguided as a way to protect students from borrowing too
much, consider the following. Would there be support for a regulation that restricted individuals from
spending more than 8 percent of their annual earnings on food? This may sound ridiculous, but the logic
is quite similar to the proposal's. Through its effects on schools, the proposed restriction intends to
protect students from spending more than 8 percent of their annual earnings to be educated. Put this
8 Quoted in "Education and Health: Evaluating Theories and Evidence" chapter 2 in Making Americans Healthier: Social and Economic Policy
as Health Policy, ed. Robert F. Schoeni, James S. House and George A. Kaplan, Russell Sage Foundation, 2008.
9 See e.g. Lleras-Muney, 2005; Oreopolous, 2003; Arendt, 2005; Spasojevic, 2003.
10 Currie and Moretti, 2003.
Page 23
Report on Gainful Employment
April 2, 2010
way, the declaration that this is too much to spend on education is not very different logically from a
declaration that it would be too much to spend on any other good that people need or enjoy, such as food
or clothing.
5. WHAT EFFECTS MIGHT THE PROPOSED REGULATION HAVE?
5.1. WHAT ARE THE EFFECTS OF INCREASED ACCESS TO FUNDING FOR HIGHER EDUCATION AND
HOW DO THESE VARY ACROSS DIFFERENT TYPES OF STUDENTS?
There is relatively good scientific evidence of how college costs and the ability to borrow affects access to
higher education. The evidence is divided into two types: (1) estimates of the effect of reducing the price
of higher education, e.g. through grants, on college attendance, and (2) estimates of the effect of
increasing access to borrowing, e.g. through subsidized loans, on college attendance.
While estimates of the effect of eligibility for Pell grants are mixed, various studies of other sources of
grants find a significant effect of reducing the cost of college on college attendance. Studies of the G.l.
Bill and the Social Security student benefit find large effects of these grants on the likelihood that those
who are eligible go to college. Dynarski (2003) for example finds that an extra $1 ,000 grant (i.e. reduction
in tuition) increases college attendance by 4 percentage points.
There are fewer good studies of the effect of access to loans on college attendance. Reyes (1995)
shows that when loan eligibility changed differentially across income groups in the early 1980's, college
enrollment rates increased for the groups for whom loan eligibility increased. Dynarski (2005) finds
positive but smaller effects of loan eligibility on college attendance based on a study of changes in
eligibility induced by the Higher Education Amendments of 1992.
A more recent experimental study may be directly relevant.11 A group of researchers simplified the
FAFSA and worked with H&R Block to automatically fill out the form using information already entered
from individuals' 1040 tax forms. For randomly selected households, H&R Block pre-populated the
FAFSA form and offered to assist the family in filling out the form. Relative to a randomly selected
comparison group, the assistance increased college enrollment significantly both for recent high school
graduates and for older independent students with no college experience. There was no effect on a
second treatment group who were just given information about the FAFSA but no assistance. These
results show that barriers to the availability of financing restrict access to higher education. Based in part
on this research, the President and Secretary of Education recently announced that the FAFSA form will
be drastically simplified.12
5.2. HOW TO MEASURE THE 25TH PERCENTILE OF EARNINGS?
The proposed regulation places a limit on the median debt among students at a program. This limit is
based on the 25
1
h percentile of earnings in occupations for which that program prepares students.
11 Bettinger, Eric, Bridget Terry Long and Philip Oreopolous, "Increasing Postsecondary Enrollment Among Low-Income Families: A Project to
Improve Access to College Information and Financial Aid" (http://gseacademic.harvard.edul-longbr/FAFSA_Project_-
_Bettinger _Long_ Oreopoulos_ -_Description_1-09.pdf)
12 http://www2. ed .gov /news/pressrelea ses/2009/06/06242009 .html
Page 24
Report on Gainful Employment
April 2, 2010
Presumably, the 25th percentile is meant to be an estimate of the typical starting salary of graduates of the
program. However, it is clear that the proposal has the potential to act as a limit on tuitions that for-profit
schools will be able to charge. The extent of these limits will depend on how the 25th percentile of
earnings for a given area of study (CIP code) is determined. In Appendix A of the proposed regulations,
the Department of Education has provided a step-by-step method for calculating what it considers to be
the 25th percentile of earnings for a particular CIP code.
The Department proposes using the wage, earnings and employment data that are regularly collected by
the Bureau of Labor Statistics. However, since the employment information is reported by occupation
based on the Standard Occupational Classification (SOC) system, rather than area of study (CIP codes) ,
areas of study must be linked to one (or more) occupations. According to the Department's methodology
the 25th percentile of earnings for each program (based on the 6-digit CIP code) can be calculated using
the following method:
First, determine all occupations based on the SOC codes available from the 0-Net crosswalk
(http://online.onetcenter.org/crosswalk/CIP/) that are associated with each 6-digit CIP code.
Next, for each soc code determine employment and annual 25th percentile wages using data
from the Bureau of Labor Statistics (BLS) available at http://www.bls.gov/oes/current/oes_stru.htm.
Finally, based on the above values calculate for each CIP code the weighted average of the
annual 25th percentile wages using the total employment of each soc code as the weights.
According to the Department, this weighted average represents the 25th percentile of earnings for
each 6-digit CIP code.
While the Department's calculations are clear and concise, they are neither simple nor correct. The
Department's choice of how to calculate the 25th percentile is also far from innocuous. Below, we
describe three ways in which the calculation of expected earnings can be quite sensitive to choices
concerning the method. All of these choices are made either explicitly or implicitly, and all of them can
have significant effects both on the earnings levels and on the ultimate impact of the proposed regulation.
These examples also point out that future changes in seemingly technical inputs, such as which
occupations are matched in the CIP to soc crosswalk, have the potential to have large impacts on
programs and students.
In calculating the earnings measure the Department makes assumptions regarding the occupations for
which a graduate is likely to enter and the relative importance of each of these occupations in determining
earnings. In addition, the calculated earnings measure is not the 25th percentile across the SOC codes.
Modifying either the assumptions or the method for calculating earnings can have substantial impacts on
whether a program meets the 25th percentile/8 percent rule. Even without explicit changes from the
Department, programs may change from meeting the proposed regulation to not meeting the proposed
regulation because of future changes in BLS coding or employment patterns.
The correspondence between CIP codes and BLS occupation codes is important:
Table 6 below shows the weighted average for the Culinary Arts/Chef Training area of study based on the
25th percentiles of the occupations that are assigned to that CIP code. According to the CIP to SOC
crosswalk that is used by the Department to determine the "25th Percentile" this area of study includes
four occupations: Chefs and Head Cooks; Cooks, Private Household; Cooks, Restaurant; and Cooks, All
Page 25
Report on Gainful Employment
April 2, 2010
Other. While Culinary Arts programs are designed to train Chefs as defined in the first occupational
category, the Department's definition appears to include those individuals working as cooks at fast food
restaurants and cafeterias, and short order cooks. While the majority of students who complete Culinary
Arts/Chef Training programs do not work at fast food restaurants, these workers' low earnings would be
used to estimate graduates' ability to afford student loan payments.
Table 6: Department of Education
Calculation of the 25th Percentile
Weighted 25th Number
Area of Study Average Percentile Employed
12.0503 Culinary Arts/Chef Training $19,278
35-1011.00 Chefs and Head Cooks $29,050 98,040
35-2013.00 Cooks, Private Household $19,030 960
35-2014.00 Cooks, Restaurant $18,230 899,620
35-2019.00 Cooks, All Other $18,390 17,340
52.0201 Business Administration/Management $62,379
11-1011 .00 Chief Executives $102,080 301 ,930
11-1021.00 General and Operations Managers $62,900 1,697,690
11-2022.00 Sales Managers $65,350 333,910
11-3011.00 Administrative Services Managers $52,240 246,930
11-3051 .00 Industrial Production Managers $64,390 154,030
11-3071 .01 Transportation Managers $59,830 96,300
11-9021.00 Construction Managers $60,650 220,550
11-9151 .00 Social and Community Service Managers $42,110 117,150
11-9199.00 Managers, All Other $64,440 365,460
13-1051 .00 Cost Estimators $42,720 218,400
13-1111.00 Management Analysts $54,890 535,850
25-1011 .00 Business Teachers, Postsecondary $46,400 69,690
51.3501 Massage Therapy/Therapeutic Massage $45,777
25-1071 .00 Health Specialties Teachers, Postsecondary $54,850 125,100
31-9011.00 Massage Therapists $23,630 51 ,250
The choice to use a weighted average of 2Sh percentiles is important:
More general areas of study are mapped to many occupations. To calculate the 25th percentile of
earnings among graduates of a program, the Department's method takes a weighted average of the 25
1
h
percentiles in each of the assigned occupations. Taking a weighted average of 25th percentiles within
occupations does not, however, give the 25th percentile of earnings among the workers in those
occupations. Take, for example, the case of Business Administration/Management (shown in the table
above). One of the occupations for which that area of study prepares students, according to the
Department of Education, is Chief Executive. Thus, the Department of Education's method bases the
Page 26
Report on Gainful Employment
April 2, 2010
early career earnings of students finishing with Business Management degrees in part on the 25th
percentile of earnings of Chief Executives. We suspect that a recent college graduate has a vanishingly
small chance of earning a Chief Executive salary in the first few years after finishing school , though some
will become Chief Executives later in their careers. An important implication of this example is that the
allowable debt levels would be very sensitive to future decisions concerning which occupations match to
each CIP code. Removing 'Chief Executive' from the set of occupations for which a Business
Administration/Management program prepares students, for example, would significantly lower the
estimated earnings, and in turn the allowable debt.
As another example, consider trying to calculate the 25
1
h percentile of earnings among workers in two
equally large states: a very high-wage state and a very low-wage state. To make the illustration clear,
imagine the extreme situation in which the lowest-paid worker in the high wage state earns twice as much
as the highest-paid worker in the low-wage state. The average of the 25th percentiles will fall somewhere
in the range between the highest-paid worker from the low-wage state and the lowest-paid worker from
the high-wage state. However, the 25
1
h percentile earner among all the workers in both states is
someone in the middle of the pack in the low-wage state.
The choice to take a weighted average of percentiles may be appropriate in some situations, and the
determination depends on the way the relevant occupations are defined. Consider, for example, that a
student leaving a program has a 50 percent chance of entering occupation X and a 50 percent chance of
entering occupation Y. This student will remain in either of these parallel occupations for his career. In
this case, the average of the 25
1
h percentiles in occupations X and Y may be an appropriate estimate of
his early career earnings experience.
In contrast, consider a student leaving a program who will enter the entry-level occupation A after which
he will eventually progress to occupation B. In this case, the weighted average of 25th percentiles does
not tell us about his experience early in his career.
In some cases occupations are defined by the Bureau of Labor Statistics to correspond to a typology
represented by occupations X andY, and in other cases (as illustrated by some of the examples in the
table above) occupations are defined to correspond more closely to occupations A and B.
The estimate of earnings does not distinguish by degree level:
Furthermore, the CIP code is not specific to a level of degree, but rather just to the area of study.
Therefore, an individual with an Associate's degree in Business Administration/Management will have the
same CIP code as an individual with a Master' s degree in Business Administration/Management. Thus,
the Department's assessment of earnings (and measure of affordable debt) will be the same for these two
individuals.
How to weight information from different occupations is important:
Assuming the Department's goal is to generate an estimate of the early earnings of a program's
graduates given that they may be prepared for multiple occupations, a weighted average of percentiles
may make sense. Even when a weighted average may be appropriate, how to weight is an important
question that must be addressed. The Department's current approach of using total employment in the
full labor market (and not specific to either degree earners or for-profit students) is likely inappropriate in
many situations. For example, the table above shows the occupations associated with Massage
Page 27
Report on Gainful Employment
April 2, 2010
Therapy/Therapeutic Massage. In this case the teachers of health specialties (which includes Massage
Therapy and other health specialties) receives more than 70 percent of the weighted average when it
likely represents a much smaller percentage of Massage Therapy graduate placements.
From a mathematical point of view, the problem is that percentiles are not linear. As a result the average
of percentiles within groups is not the percentile of the full population. It is therefore possible that the
Department of Education's method for calculating the 25th percentile of earnings would not survive the
rulemaking process. For this reason, in our calculations of impact below we present estimates that are
based on an alternative method of calculating the 25th percentile among the workers in the occupations
that match to an area of study.
What is the 2Sh Percentile?
As noted above the Department's calculated earnings measure is not the 25
1
h percentile of the
occupations that are assigned to a CIP code. An arguably more appropriate measure of the 25th
percentile can be obtained by sorting the individual earnings information of all individuals in occupations
assigned to a given CIP code and determining the earnings at the 25th percentile of that set of workers.
To do this, we first obtained a crosswalk between CIP codes and BLS occupation codes from the National
Center for Education Statistics (NCES), a division of the U.S. Department of Education. We then merged
this information with earnings data from the Current Population Survey (CPS) March Annual Demographic
File. Each March, the CPS includes more-detailed questions about annual earnings and demographics.
For each CIP code, we sorted the annual earnings of individuals in the occupations that were matched to
that CIP code, and calculated the 25th percentile of annual earnings. The table below compares the
difference in "25th percentiles" based on the alternative methods of selected CIP codes. In addition, for
each CIP code we have computed the implied maximum debt allowed based on an 8 percent limit on
annual loan payments (assuming a 1 0-year repayment schedule at 6.8 percent interest). As is clear from
the table, the maximum debt can vary substantially depending on the calculation of the 25th percentile.
Page 28
Report on Gainful Employment
April 2, 2010
Table 7: Comparison of 25m Percentile Earnings and Maximum Debt Level f or
Selected CIP Codes
Full-Time Earners De artment of Education
CIP 25th Maximum 25th Maximum
Code CIP Description Percentile Debt Percentile Debt
10.0202 Radio & Television Broadcasting $27,000 $15,312 $27,207 $15,430
Technology/Technician
$7.940 1
12.0503 Culinary Arts/Chef Training $14,000 $19,278 $10,933
14.0901 Computer Engineering, General $48,000 $27,222 1 $73,752 $41 ,826
14.0903 Computer Software Engineering $47,000 $26,655 ' $73,791 $41 ,848
(New)
14.1001 Electrical, Electronics & Communi- $45,000 $25,520 $75,437 $42,782
cations Engineering
15.1301 Drafting & Design Technolo- $35,000 $19,849 $35,266 $20,000
gy!Technician, General
15.1399 Drafting/ Design Engineering Tech- $35,000 $19,849 $35,130 $19,923
nologies/Technicians, Other (New)
31.0501 Health & Physical Education, Gen- $32,000 $18,148 ! $19,927 $11 ,301
era I
31 .0504 Sport & Fitness Administra- $32,000 $18,148 $18,989 $10,769
lion/Management
43.0203 Fire Science/Firefighting $45,000
$25,520 I
$31 ,532 $17,883
47.0201 Heating, Air Conditioning, Ventilation $29,700 $16,843 $31 ,070 $17,620
& Refrigeration Maintenance Tech-
nology!Technician
51 .2001 Pharmacy (PharmD, BS/BPharm) $41 ,000 $23,252 1 $80,585 $45,701
51.3501 Massage Therapy/Therapeutic Mas- $36,000 $20,416 $45,777 $25,961
sage
52.0201 Business Administra- $45,000 $25,520 l $62,379 $35,376
!ion/Management
I
52.0408 General Office Occupations & Cieri- $23,000 $13,044 $23,239 $13,179
cal Services
52.1902 Fashion Merchandising $35,000 $19,849 I
$36,460 $20,677
52.1904 Apparel and Accessories Marketing $35,000 $19,849 $36,460 $20,677
Operations
As shown above, the particular way to calculate the 25th percentile is not innocuous. Small changes in
the way one calculates this number causes large differences in the estimate of early career earnings.
One concern would be that future changes in the method of calculating this number could have serious
consequences. We estimate that differences in earnings levels resulting from changes in how the 25th
percentile is calculated would lead to large differences in the number of students impacted by the
proposed regulation. This suggests that further consideration should be given to: (a) whether the 25th
percenti le concept is appropriate, and (b) whether the method of calculating the student's estimated
ability to pay is overly sensitive to small changes in the future and valid from a scientific standpoint.
Page 29
Report on Gainful Employment
April 2, 2010
5.3. PRELIMINARY ANALYSIS BASED ON DATA SUBMITTED BY CCA MEMBER INSTITUTIONS
To estimate the impact of the proposed regulation on the Title IV eligible for-profit postsecondary
institutions, we collected data from Career College Association (CCA) member institutions. Specifically,
we collected student/loan level data from each institution based on the population included in their 2006,
2007 and 2008 Cohort Default Rate calculation. These data include information on student loans and
default status on all students entering repayment during a given cohort year for 3 years after entering
repayment, and are the actual data that the institution's cohort default rate is based upon. We also
received individual level demographic data from each institution including race, gender, program of study
(CIP code) , OPEID, campus information, total loan amounts (both public and private) , and length of
program. In all, we received data from 17 different institutions, representing approximately 450
campuses, 640,000 students and over 10,000 separate programs from institutions ranging from very
small to very large.
In order to determine the impact of the proposed regulation on the CCA schools for which we received
data, we calculated both the 25th percentile of earnings based on the methodology used by the
Department of Education, and the median debt of graduating students from each of the schools and
programs with available data. In addition, we calculated the 25th percentile of earnings for each CIP
code based on full-time earners in the CPS data. This alternative 25
1
h percentile of earnings calculation
was done according to the method described in the previous section.
The median total loan amounts (public and private) accrued by graduating students were calculated for
each school, OPEID, campus, program length, and 6-digit CIP code. The Department of Education's
methodology requires that students who do not take any loans (public or private) should be included in
the median calculation as having accrued o loans. Since the data we have available for the CCA schools
only include students who have taken some form of government loan, we needed to impute the number of
students not taking any loans. Also, some schools did not provide data on the private loans taken by
students so we needed to impute the value of private loans in these instances. We do not have data on
students who do not take any public loans, but take private loans. We have not included any adjustment
for these individuals.1 3
In order to account for students not taking any loans, we used I PEDS data to calculate the average
percent of students in private, for-profit institutions that do not take any loans (approximately 20 percent).
Since the population that we observe in the data are only 80 percent of the total population that should
be included in the calculations we use the 37 .5th percentile of total loans amounts instead of the 50th
percentile as this would impute a total of 20 percent of the total population as having o loans (since they
would all be below the median).
In cases where no data was available to assess the amount of private loans taken, we multiplied the
value of public loans by 1.47 since the average percent of total loans that were public (based on
NPSAS:2008 data from the NCES) was approximately 70 percent.
13 For most schools we do not have information on loans students took at prior postsecondary institutions. As a result, we underestimate
median total debt, thus possibly underestimate the fraction of programs that would be impacted by the proposed debt limit rule.
Page 30
Report on Gainful Employment
April 2, 2010
Based on the 25th percentile of earnings determined above, we calculated the maximum amount of debt
that could be accrued using the 8 percent rule proposed by the Department of Education (assuming a 10-
year repayment schedule at 6.8percent interest). Comparing the maximum debt value with the median
debt actually accrued from the students in each program, we determined the programs which would
currently be impacted by the proposed regulation (i.e. the programs whose median debt was higher than
the maximum allowed under the regulation's guidelines).
As shown in Table 8, our analysis implies that approximately 18 percent of the programs we examined
would be impacted by the 8-percent/25
1
h-percentile rule when using the Department of Education's
income calculation. Using the CPS full-time earners results in nearly 25 percent of the programs being
impacted. The impact is disproportionately on larger programs as nearly 34 percent of students are
impacted using the Department's approach, and almost 50 percent are impacted using the CPS full-time
earners. In the sample of students analyzed, approximately 29 percent of black students and 35 percent
of Hispanic students would be in programs impacted by the proposed regulation. In addition, 25 percent
of women are in programs that would be affected. All of these percentages are higher when the
alternative measure of the 25
1
h percentile of earnings is used to evaluate programs.
Table 8: Percent of CIPs Impacted by Proposed Regulation
(Assuming 8 Percent Debt Ratio Using the 25th Percentile of Income)
Median Loan Based on Graduates
Total Percent of Percent of Percent of Percent of
Number of Programs Students Females Blacks
Programs Impacted Impacted Impacted Impacted
Department of Education 25
111
Percentile 10,725 18.19% 33.72% 24.79% 28.91%
CPS 25
111
Percentile 10,695 24.58% 49.10% 39.95% 44.91%
Source: Data provided by CCA member institutions.
Percent of Percent of
Hispanics Asians
Impacted Impacted
34.89% 44.26%
47.40% 65.14%
As shown in Table 9 below, if the sample of schools and programs used in the analysis is representative
of the full set of for-profit schools and programs- and we caution that not enough analysis has been done
yet to ascertain whether this is a reasonable assumption- these estimates imply that each year 361 ,000
students, including 68,300 non-Hispanic black students, 78,500 Hispanic, and 179,000 women, would
Page 31
Report on Gainful Employment
April 2, 2010
enter for-profit postsecondary programs that would lose eligibility for participation in the federal Title IV
financial aid programs.14
While some capacity may exist in other sectors of higher education to absorb these students, recent
reports indicate that the most likely alternatives-community colleges-are already oversubscribed in
many cases, and are facing further financial cutbacks as the states that provide much of their funding
face severe financial challenges. Given recent growth rates at for-profit postsecondary institutions, we
estimate that by 2020 approximately 1 million non-Hispanic black students and an additional 1 million
Hispanic students are on track to attend programs that would be adversely affected, and would be denied
access as a result.15
14
The annual flow of students in for-profit programs is estimated from the 12-month enrollment reported in the IPEDS. Since the IPEDS
figures provide the stock of students enrolled at a given point in time we divide the number of students enrolled in a 4-year program
by 4, the number of students enrolled in a 2-year program by 2, and then add those results to the number enrolled in less than 2-
year programs to obtain an estimate of the flow of students Into for-profit schools. This is likely an underestimate of the flow
because all students do not stay enrolled for the full length of the program and institutions are categorized base on the longest
program offered (so, some students recorded in a 4-year program are enrolled in something less than four years).
15 Estimates based on the CPS full-time earners and estimates of impacted students by state are provided in the appendix.
Page 32
Report on Gainful Employment
April 2, 2010
Year
Table 9: Estimated Number of Students Impacted by 2020
Median Loan Based on Graduates
25th Percentile Based on Department of Education Calculation
Number of
Total Number of African- Number of
Number of Female American Hispanic
Students Students Students Students
Impacted Impacted Impacted Impacted
Number of
Asian
Students
Impacted
Using the Department of Education's 25th percentile of annual earnings and 8% Debt-to-Earnings Ratio
2011 361,172 179,149 68,348 78,545
2012 392,955 194,914 74, 363 85,456
2013 427,535 212,066 80,907 92,977
2014 465,158 230,728 88,027 101,159
2015 506,092 251,032 95,773 110,060
2016 550,628 273,123 104,201 119,746
2017 599,084 297,158 113,371 130,283
2018 651,803 323,307 123,347 141,748
2019 709,162 351,759 134,202 154,222
2020 771,568 382,713 146,012 167,794
Total Students Impacted 5,435,157 2,695,948 1,028,550 1,181,990
Note: The number of impacted students assumes that the CCA data is representative of all for-profit
schools, that for-profit schools will cont inue to grow at 8.8% per year (the growth rate over the last five
years), and the relat ive student composition does not change during this period.
15,875
17,272
18,792
20,445
22,245
24,202
26,332
28,649
31,170
33,913
238,895
Based on our estimates, the impact of the regulation would vary across types of programs. Because the
limits on borrowing do not vary with the length of program, longer programs would be more severely
impacted. Whereas approximately 18 percent of students in less than 2 year programs would be
impacted, we estimate that approximately 40 percent of students in 2- and 4-year programs would be
impacted. Table 10 below shows the percent of programs and students impacted by program length. the
results based on the CPS 25
1
h percentile are also provided in Table 10 below.
Page 33
Report on Gainful Employment
April 2, 2010
Table 10: Percent of CIPs Impacted by Proposed Regulation
(Assuming 8 Percent Debt Ratio Using the 25th Percentile of Income)
Median Loan Based on Graduates
Total Percent Percent Percent Percent Percent Percent
Number of of of of of of
of Programs Students Females Blacks Hispanics Asians
Program Type Programs Impacted Impacted Impacted Impacted Impacted Impacted
Department of Education 25
1
h Percentile
Less than 2
Years 2,335 10.66% 18.79% 7.94% 17.06% 13.43% 18.08%
2 Year 4,493 18.52% 39.04% 27.86% 35.71% 52.76% 56.35%
4 Year 2,892 22.23% 40.93% 39.19% 34.98% 53.71% 51 .43%
CPS 25
1
h Percentile
Less than 2
Years 2,335 15.59% 25.88% 16.22% 24.84% 17.65% 31 .86%
2 Year 4,494 24.81% 50.03% 37.95% 49.88% 68.54% 74.62%
4 Year 2,853 28.74% 68.79% 68.47% 65.58% 76.91% 78.01%
Source: Data provided by CCA member institutions.
We also estimate that the impact would not be limited to a few areas of study, but would impact a wide
variety of programs. Table 11 below reports the results aggregated to general CIP categories for
categories for which we have data on at least 1 oo programs. For example, we estimate that nearly 14
percent of Health Professional and Related Clinical Sciences, including Nursing, programs and more than
46 percent of Engineering Related Technologies/Technicians programs would not currently satisfy the
proposed debt limit rule as defined by the Department. If the alternative measure of 25
1
h percentile
earnings were to be adopted, the percent of impacted programs and students substantially would be
higher.
Page 34
Report on Gainful Employment
April 2, 2010
Table 11 : Percent of CIPs Impacted by Proposed Regulation
(Assuming 8 Percent Debt Ratio Using the 25th Percentile of Income)
Median Loan Based on Graduates
Total Percent Percent Percent Percent Percent Percent
Number of of of of of of
of Programs Students Females Blacks Hispanics Asians
CIP CIP Description Programs Impacted Impacted Impacted Impacted Impacted Impacted
Department of Education Income Calculation
9 Communications, Journalism, & Related Fields 111 32.43% 70.89% 77.00% 68.40% 68.05% 86.67%
10 Graphic Communications 219 28.31% 51.27% 45.19% 53.47% 57.24% 55.81%
11 Computer & Information Sciences & Support 1,390 19.35% 32.80% 25.12% 35.15% 42.05% 41 .66%
Services
12 Personal & Culinary Services 542 27.31% 88.82% 85.02% 80.35% 88.81% 94.00%
13 Education 192 26.04% 51 .56% 55.46% 43.28% 26.43% 30.10%
15 Engineering Related Technologies/Technicians 535 46.73% 81 .70% 66.91% 80.37% 75.63% 91.35%
22 Law, Legal Services, & Legal Studies 331 9.97% 21 .91% 22.36% 8.37% 5.28% 2.56%
42 Psychology 185 33.51% 70.71% 69.90% 75.13% 54.15% 78.81%
43 Protective Services 806 9.43% 13.97% 13.79% 11.48% 26.57% 18.02%
47 Mechanic & Repair Technology 160 39.38% 80.84% 73.46% 77.70% 77.64% 87.86%
50 Visual & Performing Arts 1,342 22.35% 56.10% 57.61% 50.60% 58.39% 63.86%
51 Health Professions & Related Clinical Sciences 2,322 13.48% 15.31% 15.57% 14.54% 7.43% 17.53%
52 Business, Management, Marketing, & Related 2,356 11.50% 9.31% 9.24% 9.83% 12.59% 13.52%
Support Services
CPS Full-Time Earners
9 Communications, Journalism, & Related Fields 111 31.53% 71.84% 76.45% 69.71% 71.25% 86.67%
10 Graphic Communications 219 39.27% 84.21% 84.07% 80.68% 91 .13% 91 .73%
11 Computer & Information Sciences & Support 1,390 29.57% 62.37% 57.27% 64.20% 74.15% 75.99%
Services
12 Personal & Culinary Services 537 32.03% 92.81% 89.96% 88.14% 94.45% 96.90%
13 Education 192 32.29% 60.30% 55.85% 95.85% 93.21% 90.29%
15 Engineering Related Technologies/Technicians 535 48.79% 84.17% 68.27% 82.38% 78.42% 94.01%
22 Law, Legal Services, & Legal Studies 356 14.89% 28.66% 28.87% 22.11% 14.76% 7.69%
42 Psychology 185 36.22% 73.48% 72.81% 77.02% 56.45% 81 .78%
43 Protective Services 806 18.24% 39.90% 40.76% 38.17% 54.61% 70.43%
47 Mechanic & Repair Technology 160 41.88% 81 .05% 73.72% 77.85% 77.82% 88.04%
50 Visual & Performing Arts 1,342 26.01% 66.07% 68.47% 68.60% 66.59% 75.50%
51 Health Professions & Related Clinical Sciences 2,322 20.24% 26.89% 26.89% 24.38% 13.56% 38.07%
52 Business, Management, Marketing, & Related 2,298 18.41% 32.04% 30.91% 36.13% 38.00% 44.06%
Support Services
Source: Data provided by CCA member institutions.
Page 35
Report on Gainful Employment
April 2, 2010
6. WHAT ALTERNATIVE REGULATIONS OR POLICIES MIGHT BE
SUGGESTED TO ADDRESS THE PROBLEM AT HAND?
6.1 . FURTHER CRITICISMS OF THE PROPOSED REGULATION
Thus far this report has focused primarily on the first provision of the Department of Education's proposal,
which would limit median debt to be no more than 8 percent of the 25th percentile of earnings in specified
occupations. Programs that fail this test could retain gainful employment status by meeting alternative
tests.
Schools would be allowed to show that the graduates of the program at their school in particular earn
more than the 25th percentile upon entering the workforce. This provision would seem to address some
of the concerns raised above. However, to properly conduct a survey of graduates would be costly, and
some of these costs would be passed on to students. Furthermore, there is no guidance as to how such
a survey would need to be conducted. Would schools be required to show that the respondents were a
representative sample of all graduates? How would this be determined? How large a sample would the
estimated earnings need to be based on? Should the survey focus on the earnings in the appropriate
occupations, as specified by the Department of Education, or would earnings in other occupations count?
How would students who chose to take jobs in other higher-paying occupations be treated? How would
students who chose to take jobs in other occupations because they were unable to find work in the
specified occupation be treated? These and other questions would need to be answered. Many of these
questions highlight that the implementation of this part of the proposal would be messy at best, and quite
possibly arbitrary.
Programs that failed the 8 percent test could also retain gainful employment status by showing that they
maintained 90 percent repayment rates. As others have noted, this would not be based on default
behavior as defined in the Cohort Default Rate calculation. Students who are not current in their
payments, even though they have not yet reached the point of default, would count against a school 's
clean record. Students in deferment or forbearance would also apparently count against a school's
repayment rate. It is difficult to know how many programs would satisfy this standard. Most problematic,
the data necessary to calculate this rate is not readily available to schools. It is therefore almost
impossible to analyze whether the 90 percent standard is appropriate. Furthermore, without the data
underlying this calculation, it is not possible for schools to monitor problems, or to affect the behavior that
leads to low repayment rates. It is also not clear whether the Department of Education based the
standard on any analysis of data.
One argument described above is that it cannot be good policy to have limits on student loan payments
that are less than the benefits to earnings from schooling. The argument is the following. Standard
estimates of the return to education suggest that a student who completes a 2-year program earns 20
percent more per year, every year she works. If the debt payment limit were less than 20 percent, she
could make the loan payments out of her 20 percent schooling bonus, and still have money left over.
Page 36
Report on Gainful Employment
April 2, 2010
If the Department of Education were to consider a debt limit that is approximately equal to the return to
education, several additional factors would need to be considered. For example:
Programs are different lengths. Longer programs have larger returns. Would the department
institute different loan limitations for 1-, 2- and 4-year programs? How would the limit be set if
there were variation within a program in how long students attended (or how many credits
students earned)?
The return to education changes over time. It has risen dramatically in the past 30 years. How
would the Department of Education decide what the return is in each year?
6.2. To DIRECTLY ADDRESS THE PROBLEM, REGULATION SHOULD FOCUS ON DISCLOSURE AND
ENSURING THAT STUDENTS MAKE INFORMED DECISIONS
A comparison of expected returns to education with the costs of education is what students are doing
when they decide whether to get a higher education, and whether to take on loans to finance that
education. If the problem the policy is trying to solve is that students are not doing this well- that they
are not making informed, considered decisions based on comparisons of expected benefits and costs-
then the regulation should address this problem.
To make this case, it may be necessary first to refocus the discussion on which problem is in need of
solving. Whereas the current proposal appears to be based on the perception that students take on too
much debt, we bel ieve this is misguided. As argued above, standard economic analysis clearly indicates
that the amount of debt should be dictated by the benefits of the investment, not by the level of income. It
may not be in the students' interest to be restricted from taking on large amounts of debt; that debt may
be the key to a better future. The important thing is to make sure that students make informed and
intelligent decisions about whether loans are right for them, and whether the benefits of the schooling
they wish to finance are large enough to repay the debt they take on.
This focus on making smart informed decisions leads directly to a policy based on provision of
information, and assistance analyzing the consequences of borrowing.
One way that this problem could be addressed directly is through different forms of disclosure and
education:
Increased scrutiny could be placed on lenders to ensure that every student who takes on a loan is
made aware ofthe costs associated with the loan, the magnitude of the annual or monthly payment,
and the length of the payback period.
Disclosure could also include mandated information regarding typical earnings of workers in the
occupation for which the student is preparing. For example, lenders could be required to show
students the 25
1
h percentile or median of annual earnings in the appropriate occupation.
This could be extended further so that students would be shown a mock budget based on an
estimate of their earnings in the appropriate occupation, their loan payments, and a standardized set
of necessary expenses. This could be done either in a standardized paper form, or an online
application could be developed to allow students to enter various earnings and expense values to
see how they fit into the budget.
Page 37
Report on Gainful Employment
April 2, 2010
7. CONCLUSION
In summary, the proposed regulation is not currently formulated to address a specific problem effectively.
Secretary of Education Arne Duncan has stated publicly that he wants to ensure that the Department
understands its proposal thoroughly so that it prevents any "unintended consequences." Our analysis
suggests that the "unintended consequences" -cutting off access to hundreds of thousands of students
who want postsecondary education-will be much more substantial than the intended consequence,
which we believe to be-though we are not certain-reducing the number of students who over borrow.
To start, the Department of Education has not clearly defined what the problem is that the regulation aims
to address. As discussed above, some perceived problems the regulation may intend to address are not
problems at all but rather a reflection of the fact that for-profit postsecondary schools serve a very
different population than not-for-profit postsecondary schools. If the Department of Education wishes to
address the problem that some students take on excessive debt, the proposed regulation is not well
designed to do so. By applying a rule at the school or program level, many other students would be
negatively affected. Our analysis suggests that 33 percent of students currently in for-profit
postsecondary schooling would be denied access. Many more students would be denied access to
postsecondary schooling than would be protected from excessive borrowing.
Furthermore, it should not be assumed that public postsecondary institutions, particularly community
colleges, would absorb these students. Given the fiscal conditions of the states, it is not obvious that
community colleges will be able to increase capacity to meet the increasing demand for postsecondary
schooling.
Finally, because for-profit schools disproportionately serve racial and ethnic minority students and
students from low-income family backgrounds, the regulation would have the effect of reducing access to
higher education to groups of students that have historically had the lowest levels of access.
Page 38
Report on Gainful Employment
April 2, 2010
8. APPENDIX A
Appendix Figure 1 :
Average age at which students first enroll in postsecondary schooling, 2008
23
22.5
22
21.5
Q)
21 0)
ro
Q)
0)
20.5
ro
.....
Q)
>
20
<(
19.5
19
18.5
18
For-profit Pri vat e not-for- Privat e not-for- Public <=2 Publi c 4-year
profit <=2 profit 4-year
Age first enroll ed
Notes: Calculated from the National Postsecondary Student Aid Study, 2008.
Page 39
Report on Gainful Employment
April 2, 2010
Appendix Figure 2:
Average income of parents or independent students prior to school entry, 2008
90,000.00
80.000.00
70.000.00
<I>
60.000.00
E
0
(.)
50.000.00
c
<I>
Ol
40.000.00
ttl
...
~
<(
30.000.00
20.000.00
10.000.00
0.00
For-profit Private not-
for-profit <=2
Private not-
for-profit 4-
year
Pllblic <=2
Average income, parents or independent. 2007
Notes: Calculated from the National Postsecondary Student Aid Study, 2008.
Pllblic 4-year
Page 40
Report on Gainful Employment
April 2, 2010
Appendix Figure 3:
Percent of students from families who received AFDC prior to school entry, 2008
18 ~ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
16
14
12
c 10
Q)
0
.....
f 8
6
4
2
0
For-profit Pri vate not-for- Private not-for- Publi c <=2
profit <=2 profit 4-year
Percent receieved Food Stamps. 2007-08
Notes: Calculated from the National Postsecondary Student Aid Study, 2008.
Publi c 4-year
Page 41
Report on Gainful Employment
April 2, 2010
Appendix Figure 4:
Percent of students who were single parents prior to school entry, 2008
35 ~ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
30
25
c 20
Q)
()
....
Q)
a. 15
10
5
0
For-profit Pri vate not-for- Private not-for- Publi c <=2
profit <=2 profit 4-year
Percent single parent. 2007-08
Notes: Calculated from the National Postsecondary Student Aid Study, 2008.
Publi c 4-year
Page 42
Report on Gainful Employment
April 2, 201 0
Appendix Figure 5:
Percent of students whose parents attended at least some college, 2008
000
For profit
BOO
700
60.0
50.0 ..
E
..
.,
..
0.
400
300
200
0.0
r:>rrvale for-protot r:>uvate lo .. prolrl
2 yea.s or more 2yt&r
Not for profit
Pubhc 4-year
ooMoctorate
4-)'l)at PrrYal e not-for
doctorate prole 4
year
Notes: Calculated from the National Postsecondary Student Aid Study, 2008.
Prwa1e not-fo.-.
protrt 4-yr

Ptl'l$18 not-lor
pro(J1 4year
doctorate
Page 43
Report on Gainful Employment
April 2, 201 0
Appendix Figure 6:
Percent of students who are Black or Hispanic, 2008
60.0 ~
50.0
40.0
c
~ 30.0
"' 0.
10.0
0.0
For profit
Not for profit
Private for- Private for- PriVate not-for- Private not-for- Private not-for- Public less- Public 2-year Public 4-year Public 4-year
profitless-than profrt 2years or profrt less than proflt4Yr pro1it4-year thar)-2-year nondoctol'31e doctorate
2 - ~ a r more 4-year nondoctorale doctorate
Percent Black or Hispanic
Notes: Calculated from the National Postsecondary Student Aid Study, 2008
Page 44
Report on Gainful Employment
April 2, 201 0
Appendix Table 1
Estimated Number of Students Impacted by 2020
Median Loan Based on Graduates
25th Percentile Based on Full-Time Earners (CPS)
Number of
Total Number of African- Number of Number of
Number of Female American Hispanic Asian
Students Students Students Students Students
Year Impacted Impacted Impacted Impacted Impacted
Using the 25th percentile of annual earnings and 8% Debt-to-Earnings Ratio
2011 525,924 288,663 106,169 106,729 23,365
2012 572,206 314,065 115,511 116,121 25,421
2013 622,560 341,703 125,676 126,340 27,658
2014 677,345 371,773 136,736 137,458 30,092
2015 736,952 404,489 148,769 149,554 32,740
2016 801,803 440,084 161,860 162,715 35,621
2017 872,362 478,811 176,104 177,034 38,756
2018 949,130 520,946 191,601 192,613 42,167
2019 1,032,653 566,790 208,462 209,563 45,877
2020 1,123,527 616,667 226,807 228,004 49,914
Total Students
Impacted 7,914,462 4,343,989 1,597,695 1,606,132 351 ,613
Note: The number of impacted students assumes that the CCA data is representative of all for-profit
schools, that for-profit schools wi ll continue to grow at 8.8% per year (the growth rate over the last five
years), and the relative student composition does not change during this period.
Page 45
Report on Gainful Employment
April 2, 201 0
Appendix Table 2
Estimated Annual Number of Students Impacted by State
Median Loan Based on Graduates
25th Percentile Based on Department of Education Calculation
Number of
Total Number of African- Number of Number of
Number of Female American Hispanic Asian
Students Students Students Students Students
State lmeacted lmeacted lmeacted lmeacted lmeacted
Using the 25th percentile of annual earnings and 8% Debt-to-Earnings Ratio
AL 1,134 630 470 13 11
AR 1,204 718 297 22 19
AZ 33,998 15,928 3,377 3,160 924
CA 44,910 22,958 3,935 16,101 6,713
co 7,593 3,363 981 1,004 248
CT 5,648 2,169 870 1,015 120
DC 4,681 2,266 2,150 368 172
FL 25,202 12,307 5,601 7,738 487
GA 10,324 5,727 4,979 491 242
lA 6,684 3,705 292 117 50
IL 18,988 9,294 4,290 2,372 677
IN 4,978 2,732 1,115 195 37
KS 1,681 893 270 108 58
KY 3,292 1,865 574 41 41
LA 3,885 2,242 1,581 57 42
MA 5,765 2,893 520 742 240
MD 4,297 1,960 2,155 188 94
Ml 8,949 5,099 3,043 232 109
MN 9,406 5,038 1,309 325 365
MO 5,510 2,953 1,349 106 80
MS 1,111 667 576 7 23
NC 2,307 1,116 863 69 44
NH 1,284 760 17 55 18
NJ 9,118 4,366 1,969 2,294 430
NM 1,612 926 102 650 27
NV 2,722 1,344 411 472 265
NY 18,845 8,797 4,928 4,439 1,129
OH 11,686 6,147 3,401 291 129
OK 2,456 1,273 417 163 49
OR 2,901 1,579 107 183 148
PA 16,909 7,653 3,334 896 299
Rl 1,839 936 189 324 38
sc 1,523 878 655 46 19
TN 5,682 2,911 1,848 124 67
TX 29,176 14,832 5,846 12,457 718
UT 2,675 1,380 33 231 83
VA 7,032 3,720 2,854 328 250
WA 3,964 2,107 269 261 490
WI 2,004 1,116 707 100 46
wv 3,226 941 321 215 80
Total Students Impacted 361 ,172 179,149 68,348 78,545 15,875
Note: The number of impacted students assumes that the CCA data is representative of all for-profit schools. States
with less than 1,000 students who are impacted and Puerto Rico are not reported above.
Page 46
Report on Gainful Employment
April 2, 201 0
Appendix Table 3
Estimated Annual Number of Students Impacted by State
Median Loan Based on Graduates
25th Percentile Based on Full-Time Earners (CPS)
Number of
Total Number of African- Number of Number of
Number of Female American Hispanic Asian
Students Students Students Students Students
State lmeacted lmeacted lmeacted lmeacted lmeacted
Using the 25th percentile of annual earnings and 8% Debt-to-Earnings Ratio
AL 1,652 1,016 730 18 17
AR 1,753 1,157 461 29 28
AZ 49,506 25,665 5,246 4,294 1,360
CA 65,396 36,992 6,113 21,879 9,881
co 11 ,056 5,419 1,525 1,364 365
CT 8,224 3,495 1,351 1,379 177
DC 6,816 3,650 3,340 500 253
FL 36,698 19,830 8,700 10,515 717
GA 15,034 9,228 7,733 667 355
HI 1,029 548 23 27 1,002
lA 9,733 5,970 454 158 74
IL 27,650 14,976 6,664 3,223 996
IN 7,249 4,402 1,732 265 54
KS 2,449 1,438 420 146 86
KY 4,794 3,005 892 56 61
LA 5,657 3,613 2,455 78 62
MA 8,395 4,662 808 1,008 353
MD 6,256 3,159 3,348 256 139
ME 1,279 924 26 13 19
Ml 13,032 8,217 4,727 315 161
MN 13,697 8,118 2,033 441 538
MO 8,024 4,759 2,095 144 117
MS 1,618 1,075 895 10 34
NC 3,359 1,798 1,341 94 64
NE 1,181 739 149 41 15
NH 1,870 1,224 26 74 26
NJ 13,277 7,036 3,058 3,117 633
NM 2,347 1,492 158 884 39
NV 3,964 2,166 638 641 391
NY 27,441 14,174 7,654 6,031 1,662
OH 17,017 9,905 5,283 395 189
OK 3,576 2,051 647 222 72
OR 4,225 2,544 166 249 218
PA 24,622 12,331 5,179 1,217 441
Rl 2,679 1,507 293 440 55
sc 2,217 1,415 1,018 62 28
TN 8,274 4,691 2,871 168 99
TX 42,485 23,899 9,081 16,926 1,057
UT 3,896 2,223 52 314 123
VA 10,239 5,994 4,434 446 369
WA 5,773 3,395 418 354 722
WI 2,918 1,798 1,098 135 68
vw 4,697 1,517 499 292 118
Total Students Impacted 525,924 288,663 106,169 106,729 23,365
Note: The number of impacted students assumes that the CCA data is representative of all for-profit schools. States
with less than 1,000 students who are impacted and Puerto Rico are not reported above.
Page 47
From: Manheimer Ann
To: Jenkins, Harold
Arsenault Leigh
Wolff> Russell
Finley Steve
Picoult Francine
Yuan, Georgia
CC: Sellers Fred
Kolotos, John
Date: 4/2/2010 9:00:16 AM
Subject: FW: Request for Follow Up Meeting This Week to Discuss Gainful Employment
Forwarding so you can access the attachments - not sure you could on the last message - Ann
From: Harris Miller [mailto:HarrisM@career.org]
Sent: Thursday, April 01, 2010 6:04PM
To: Shireman, Kanter, Madzelan, Dan; Bergeron, David
Cc: Brian Tammy Halligan; Manheimer, Ann
Subject: RE: Request for Follow Up Meeting This Week to Discuss Gainful Employment
Bob, et.al.: We look forward to the meeting Monday morning at 9 AM at the Department to discuss the gainful
employment issue. Attending as part of the CCA delegation will be, in addition to me, Brian Moran, CCA' s Executive
Vice President for Government Relations, Dr. Jonathan Guryan, University of Chicago, and Dr. Matthew Thompson of
Charles Ri ver Associates, the leads on the research we will be presenting, Rebecca Campoverde from Kaplan, Anthony
Guida from EDMC, and Tom Babel from Devry.
If you could please let me know who will be in attendance from the Department, that would be appreciated.
Attached you will find three documents. The first is the Executive Summary of the research findings. I will send you the
full report tomorrow, Friday. Last minute editing on the full report was delayed by the Passover holiday.
Second is the proposal CCA is putting forward as an alternative to the gainful employment proposal put forth by the
Department during the negotiated rulemaking.
Third is a document that summarizes the extensive tools the government already has to take actions against "bad actors"
in our sector.
Please let me know if you have any questions or comments. I look forward to a productive conversation Monday. Enjoy
the spectacular weather.
Ranis
From: Shireman, Bob [mailto:Bob.Shireman@ed.gov]
Sent: Monday, March 29, 2010 10:48 AM
To: Ranis Miller; Kanter, Martha; Madzelan, Dan; Bergeron, David
Cc: Brian Moran; Tammy Halligan; Manheimer, Ann
Subject: RE: Request for Follow Up Meeting This Week to Discuss Gainful Employment
Work with Ann Manheimer on arranging the meeting. Please send your research and alternative prior to the meeting so
we can get concrete and specific in the discussions.
Thanks,
-Bob
From: Harris Miller [mailto:HanisM@career.org]
Sent: Monday, March 29,2010 10:04 AM
To: Kanter, Martha; Shireman, Bob; Madzelan, Dan; Bergeron, David
Cc: Brian Moran; Tammy Halligan
Subject Request for Follow Up Meeting This Week to Discuss Gainful Employment
Martha/Bob/Dan/David:
CCA would like to have a follow on meeting with you this week to the one Brian Moran and I had with Dan and David
ten days ago to discuss a) the results of our research (which was not complete when we met) on the impact of the
Department's gainful employment proposal on students in higher education, and b) an alternative (which we discussed in
general concept with Dan and David) to the Department ofEducation proposal. We can make ourselves available at
almost any time starting tomorrow. Please let me know whom l should contact to arrange a meeting. Thanks in advance.
Harris
*****************************************************************
Harris N. Miller
CEO/President
Career College Association
1101 Connecticut Avenue, NW, Suite 900
Washington, DC 20036
hanism@career. org
+ l 202 336 6754
Executive Assistant: Jackie McWilliams
j acki em@career. org
+ 1 202 336 6706
www. career.org
To:
From:
Date:
Re:
MEMORANDUM
Interested Parties
Career College Association
March 29, 2010
The Myth That the Department of Education is Powerless to Stop "Bad
Actors" in Postsecondary Education
CCA staff and members have heard during recent visits with Administration
officials, especially at the Department of Education ("the Department"), and less
frequently during Capitol Hill visits, a narrative that runs as follows: Although the
majority of for-profit postsecondary institutions may be high quality institutions serving
their students well, there are a few "bad actors" over which the Department lacks
sufficient regulatory authority. In fact, one of the justifications for the radical proposal
put forth by the Department to change the meaning of "gainful employment" as it has
been accepted and understood for forty years is to stop the conduct of "bad actors." Part
of this same narrative is that prospective students do not have adequate consumer
protections to prevent these bad actors from taking unfair advantage of them.
CCA agrees that any percentage of bad actors in the sector is unacceptable. What
we do not agree with, however, is the mythology that the Department lacks what is, in
fact, extensive authority under current law to investigate and take action against
institutions that are not following the myriad of rules that govern higher education
generally, and for-profit higher education, in particular. In addition to a high degree of
regulatory oversight at the federal level, each of these schools is also overseen by the
other two legs of the regulatory "triad" -state government and accrediting entities that
are approved by the Department itself. For example, over the last decade, the sector has
seen a vast increase in active oversight by state agencies.
This memorandum provides an overview of the layers of quality control and
consumer protection regulation that already exist within the authority of the Department
and other federal and state entities that can and do serve as a check on "bad actors." It is
important to note that some significant enhancements in the law were enacted as part of
the Higher Education Opportunity Act of 2008 ("HEOA"), which amended the Higher
Education Act of 1965 ("HEA"). The final regulations implementing the HEOA are not
effective until july 1, 2010 and, therefore, many important enhancements to the law have
yet to be fully implemented. In addition, during this year's Title IV Program Integrity
Negotiated Rulemaking process, the negotiators reached consensus on changes to
strengthen the misrepresentation regulations. The process of reaching consensus on
enhanced misrepresentation regulations, of which CCA was a part, demonstrates that
higher education and the Department share the goal of weeding out bad actors through
7
common sense enforcement of and enhancements to existing law.
I. Quality Control and "Bad Actors"
As a threshold matter, the Department has never explained clearly how it defines
a "bad actor" school or what specific rules such a school is violating. In reality, real laws
apply to real schools that violate them. For example, if a bad actor is a school that
misrepresents information to a potential student about placement rates or salary
prospects, federal and state consumer protection laws apply. If a bad actor is a school
that "lures" students to enroll through prohibited payments to admission or financial aid
officials, regulations subject that school to penalties and risks to Title IV eligibility. If a
bad actor is a school that charges "too high" a tuition relative to earnings, then students
will default massively and the cohort default rate thresholds will act as a check on Title IV
eligibility. This section describes the quality control regulations in current law put into
place over the last decade to improve quality and thereby eliminate so-called "bad
actors."
The Higher Education Act of 1972 established Title IV eligibility for proprietary
institutions consistent with the Department's goals of expanding postsecondary
education access for a growing and financially needy college-aged population. Since that
time, Congress has established and built upon the following measures in federal law,
some generally applicable to all postsecondary institutions and some specific to
proprietary schools, to ensure school and program quality:
Caps on excessive cohort default rates (Fiscal Year 1989 Appropriations Act);
Thirty day delay in delivery of loans to new borrowers by schools with cohort
default rates of more than 30% (Omnibus Reconciliation Act of 1989);
Prohibition on certification or recertification of an institution as Title IV eligible if
it has lost accreditation during the past 24 months, unless the accreditation has
been restored or the institution demonstrates program integrity (Omnibus
Reconciliation Act of 1989);
Addition of new program integrity provisions, including periodic recertification,
and other requirements for strengthened national accreditation standards and
procedures (Higher Education Amendments of 1992);
Establishment of early 85/15 (now 90/10) ratio for proprietary schools (Higher
Education Amendments of 1992);
Requirements for pro rata refund policy for first-time students who withdraw
(Higher Education Amendments of 1992) and return to Title IV policy as a
successor to that;
Enhancements to federal financial responsibility standards and requirement for
letters of credit for institutions with impaired financial responsibility;
Requirement for annual submission of audited financial statements and financial
aid compliance audits (Higher Education Amendments of 1992);
7
Establishment of incentive compensation prohibitions on student recruitment and
processing of financial aid;
Strengthened federal requirements for ability-to-benefit tests and limit on
percentage of ability to benefit students;
Student satisfactory academic progress requirements (Omnibus Budget
Reconciliation Act of 1989, Omnibus Reconciliation Act of 1990; Higher Education
Technical Amendments of 1991);
Required disclosure of completion and placement rate requirements for
short-term programs;
Establishment of the clock to credit hour conversion rules and minimum academic
year requirements as part ofPell grant eligibility;
Required disclosure of graduation rates (1990 Student Right to Know and Campus
Security Act);
Limitations on branch campuses;
Increased disclosures to borrowers of student loans; and
Disclosure of schools with the greatest tuition increases.
In addition to these measures, institutions must also ensure that the information
disclosed about programs is accurate. The Department may initiate a fine, or a limitation,
suspension, or termination of Title IV eligibility for any substantial misrepresentation
made by an institution regarding the nature of its educational program, its financial
1
charges, or the employability of its graduates. At the recently concluded negotiated
rulemaking, the Department proposed improvements to the misrepresentation
regulation that were agreed upon by consensus by stakeholders, including CCA, that will
increase the oversight of the information provided by institutions. Specifically, the
proposed regulatory language on which the negotiators reached consensus strengthens
the current misrepresentation regulation in several ways, including:
Expanding the class of persons to whom a substantial misrepresentation can be
made to include accrediting agencies and state agencies, as well as prospective
students and the general public;
Holding institutions responsible for any misrepresentation made by companies
with which they contract their services; and
Broadening the definition of what constitutes "substantial misrepresentation."
During negotiated rulemaking, the Department also proposed language that
would support robust state regulation of postsecondary institutions. In addition, it is
important to remember that under current state laws and state attorneys general have
broad grounds to pursue schools in violation of state licensing or consumer protection
requirements. State authorizing agencies and state attorneys general are additional tools
1
See 34 CFR 668.74(c).
7
against bad actors in the sector.
There are also other oversight agencies in addition to the Department that
sanction institutions for misrepresentations or inaccuracies in the information they
provide. Institutions are subject to oversight by the Federal Trade Commission
(proprietary institutions only), the accrediting agencies, the Federal Reserve Board (for
institutional and private educational loans), the Securities and Exchange Commission
(for publicly traded schools), and the Department's Federal Student Aid Office and Office
of Inspector General.
II. Consumer Protections
As stated above, the provision of postsecondary educational services is a highly
regulated activity with numerous controls and enforcement mechanisms currently
available to the Department of Education in statute and regulation that permit it to
identify, investigate, and end Title IV eligibility to institutions of higher education that
violate federal and state law. Consumer protection enhancements applicable to the
sector at the federal and state level have also grown consistently over the last decade,
were further enhanced in the HEOA, and can be strengthened even more, without relying
on a "gainful employment" hook that would damage quality educational service
providers.
A variety of consumer protections exist to inform the educational choices of
prospective and current students. Institutions are required to disclose detailed
information about college cost, student success metrics, and loan repayment. Under the
Higher Education Act and recent amendments to that Act, schools are required to provide
2
a vast array of consumer information, including the following:
The price of attendance, including tuition and fees, estimated cost of books and
supplies, room and board, and transportation costs, and any additional costs for a
program in which the student is enrolled or expresses an interest (HEA, 20 U.S.C.
1092(a)(1)-(2), unchanged by HEOA);
Institutions must make available on their websites by October 29, 2011 a net
price calculator (HEOA Section 111 amended HEA Title I, Part C; added HEA, 20
U.S.C. 1015a(a),(h));
Each institution must make available to prospective and enrolled students
information about the institution's refund policy, requirements and procedures
for withdrawal; and requirements for the return of Title IV, HEA grant or loan aid
(HEA, 20 U.S.C. 1092(a)(1)-(2), unchanged by HEAO);
Academic program and facility information to prospective and enrolled students
(HEOA added 20 U.S.C. 1092(a)(1)(G)(iv));
2
.s.e.e...HEA Section 48S(a)(l) (20 U.S.C. 1092(a)(l)); see also. Report of the National Postsecondary Education
Cooperative, Information Required to be Disclosed Under the Higher Education Act of 1965: Suggestions for
Dissemination, Nov. 2009 (including HEOA amendments).
7
Each institution must make available to prospective and enrolled students: names
of accreditors applicable to the institution and programs, and procedures for
obtaining and reviewing documents describing such accreditation (HEA, 20 U.S.C.
1092(a)(1)-(2), tmchanged by HEAO);
An institution that advertises job placement rates as a means of recruiting
students to enroll must make available to prospective students, at or before the
time the prospective student applies for enrollment: (1) the most recent available
data concerning employment statistics and graduation and (2) any other
information necessary to substantiate the truthfulness of the advertisements; and
(3) relevant state licensing requirements of the state in which the institution is
located for any job for which the course of instruction is designed to prepare
students (HEA, 20 U.S.C. 1094(a)(8), unchanged by HEOA);
Institutions must make available to current and prospective students information
regarding the types of graduate and professional education in which graduates of
the institutions' 4-year degree programs enroll (HEOA, 20 U.S.C. 1092(a)(l)(S));
Detailed financial aid information, including information about all of the
need-based and non-need based federal, state, local, private and institutional aid
programs available to students at the institution (HEAO added HEA
485(a)(l)(M));
The retention rate of certificate and degree seeking, first-time, full-time
undergraduate students as reported to !PEDS (HEOA added HEA 485(a)(l)(U));
The completion or graduation rate of certificate or degree seeking, first-time,
full-time, undergraduate students, including data disaggregated by gender, major
racial and ethnic subgroups, recipients of a Federal Pell Grant, recipients of a
subsidized Stafford Loan who did not receive a Pell Grant, and students who did
not receive either a Pell Grant or a subsidized Stafford Loan (HEOA added HEA
485(a)( 4) and HEA 485(a)(7));
Each institution must publicly disclose and make available to prospective and
enrolled students a statement of the institution's transfer of credit policies and
articulation agreements (HEOA added 20 U.S.C. 1092(h));
Institutions must make available to current and prospective students information
regarding the placement in employment of, and types of employment obtained
by, graduates of the institutions' degree or certificate programs (HEOA added 20
U.S.C. 1092(a)(l)(R));
Textbook cost disclosure on internet course schedule used for preregistration and
registration purposes (HEOA added 20 U.S.C. 1015b);
Detailed entrance and exit counseling for student borrowers, including average
anticipated monthly repayment amount, repayment plan options, and debt
management strategies (HEOA added new HEA 485(1) and 485(b )(l)(A)); and
Detailed private education loan disclosures, including a student self-certification
form. In addition, under the amendments to the Truth in Lending Act, private
lenders must provide three detailed disclosures to borrowers before making
7
private student loans (HEOA added HEA 487(a)(28) and HEA 151-155 (20
U.S.C. 1019, 1019a-1019d)).
The Department makes consumer information easily available to students, parents, and
the public. The Department posts many consumer information items on the College
Navigator website for each institution, including:
Tuition, fees and other estimated student expenses for the last several years, with
the percentage increase from the most recent year to the current one and a
multi-year tuition calendar;
The percentages of students who receive various types of aid and the average
amounts of such aid;
Graduation rates, disaggregated by major racial subgroups;
Retention rates for full-time and part-time students;
Average amount of subsidized and unsubsidized Stafford loans; and
Cohort default rates for the most recent three years.
In addition, the Department provides some of this consumer information direct to
individual students by institution - when the student inputs the school's code on the
FAFSA. It is also relevant that from the standpoint of providing tools to students to
manage student loan debt, Congress first authorized the Income Based Repayment
program in 2007 as part of the College Cost Reduction and Access Act of 2007, and that
program is just now being actively promoted by the Department as a tool to assist
graduates with loan debt management.
In sum, despite the perception created by the Department's "bad actor" narrative
on gainful employment, proprietary postsecondary education is highly regulated at the
federal and state levels and through its accrediting bodies. The gainful employment
proposal is vastly overbroad and misguided in light of the diverse array of tools that
exist under current law to identify and target bad actors and to protect and assist student
consumers with loan debt issues.
7
C
D A Lharles. River

Prepared For:
Harris N. Miller
Career College Association
1101 Connecticut Ave. NW, Suite 900
Washington, DC 20036
Prepared By:
Report on Gainful
Employment: Executive
Summary
Jonathan Guryan, Ph.D.,
Associate Professor of Economics, University of Chicago Booth School of Business
Matthew Thompson, Ph.D.,
Vice President
Charles River Associates
1545 Raymond Diehl Road, Suite 260
Tallahassee, FL 32308
Date: March 29, 2010
Report on Gainful Employment: Executive Summary
March 29, 2010 Charles River Associates
This report addresses the definition of "gainful employment" proposed by the U.S.
Department of Education as a part of negotiated rulemaking.
CONSEQUENCES/IMPACT OF THE PROPOSED GAINFUL EMPLOYMENT REGULATION
To estimate the impact of the proposed regulation on Title IV eligible for-profit postsecondary
institutions, we collected data from Career College Association (CCA) member institutions.
We analyzed data representing approximately 640,000 students and over 10,000 separate
programs from institutions ranging from very small to very large. Using these data, we
compared the median debt among a program's graduates to the maximum allowed loan
implied by the 8-percent rule applied to the appropriate 25th percentile of earnings, calculated
according to the Department of Education's instructions.
Based on these calculations, we estimate that 18 percent of for-profit postsecondary
programs would not satisfy the debt limit requirement of the gainful employment proposal. In
our calculations, larger programs will be more likely to be impacted. Consequently, we
estimate that 33 percent of students in for-profit postsecondary programs would be impacted.
Applying these findings to for-profit enrollment reported in IPEDS, we estimate that
approximately 360,000 students each year enter for-profit postsecondary programs that
would lose eligibility for participation in the federal Title IV financial aid programs. Given
recent growth rates of enrollment in for-profit postsecondary institutions, we estimate that by
2020, approximately 5.4 million students who are on track to attend programs would be
denied access by the proposed regulation. Our estimate is that the impact of the regulation
would be severe on racial and ethnic minority groups, in part because for-profit schools are
proportionately more likely than not for-profit postsecondary schools to serve these students.
We estimate that each year approximately 68,000 non-Hispanic black students and an
additional 79,000 Hispanic students enter for-profit postsecondary schooling in programs that
would be impacted by the proposed regulation and would not be able to attend these
programs. While some capacity may exist in other sectors of higher education to absorb
these students, recent reports indicate that the most likely alternatives-community
colleges-are already oversubscribed in many cases, and are facing further financial
cutbacks as the states that provide much of their funding face severe financial challenges.
Given recent growth rates at for-profit postsecondary institutions, we estimate that by 2020
approximately 1 million non-Hispanic black students and an additional 1 million Hispanic
students are on track to attend programs that would be adversely affected, and would be
denied access as a result.
Based on our estimates, the impact of the regulation would vary across types of programs.
Because the limits on borrowing do not vary with the length of program, longer programs
would be more severely impacted. Whereas approximately 18 percent of students in less
than 2 year programs would be impacted, we estimate that approximately 40 percent of
students in 2- and 4-year programs would be impacted. We also estimate that the impact
would not be limited to a few areas of study, but would impact a wide variety of programs.
For example, we estimate that nearly 14 percent of Health Professional and Related Clinical
Page 1
Report on Gainful Employment: Executive Summary
March 29, 2010 Charles River Associates
Sciences, including Nursing, programs and more than 46 percent of Engineering Related
Technologies/Technicians programs would not currently satisfy the proposed debt limit rule.
A SOLUTION IN SEARCH OF A PROBLEM
We are unaware of any, scientific or otherwise, study by the Department of Education that
outlines what the problem is that the proposed regulation is meant to address. The problem
that the regulation aims to solve has not been clearly stated - neither its nature nor its extent.
As a result, we believe, the proposed regulation is not well-designed to address a specific
problem.
Any regulation aimed at solving a specific problem should be tailored to address that problem
without causing undue harm to other individuals. Based on a few anecdotal references to
complaints of high debt levels by students, posted in the public comments to Negotiated
Rulemaking on the Department's website, we suspect that the Department of Education aims
to protect some students who take on more debt than they can expect to repay. We also
suspect that the number of students who take on more debt than is in their personal interest
is small, but we are not aware of a study of this population that would inform this speculation.
A regulation aimed at protecting these students should focus on identifying these students
and should address these students directly. Because the proposed regulation applies to
schools rather than students, it would affect students who are not over-borrowing along with
any who may be over-borrowing.
Are Default Rates Higher at For-Profit Colleges?
One stated rationale for the proposed regulation is that default rates are higher among
students at for-profit postsecondary schools than among those at not-for-profit colleges. Our
analysis of the Beginning Postsecondary Students Survey data shows that at least half of the
difference in default rates between for-profit and not-for-profit schools is because they serve
different types of students. For-profit schools are more likely to serve low-income students,
racial and ethnic minority students, students who are the first in their immediate family to
attend postsecondary schooling, and students whose families have collected public
assistance. Students in these groups are more likely to default on student loans after
attending both for-profit and not-for-profit colleges.
We estimate that if all schools served students with moderate family resources, loan default
rates at for-profit schools would be cut by more than half and the difference in default rates
between for-profit and not-for-profit schools would also be cut approximately in half. These
estimates are based on regression-based controls for students' family income, dependent
status, race/ethnicity, program completion status, parental education, family welfare receipt,
and Pell eligibility. It is possible that controlling for additional student characteristics, if data
were available, would reduce the default rate gaps even more.
Page2
Report on Gainful Employment: Executive Summary
March 29, 2010 Charles River Associates
RATIONALE FOR THE METRICS OF THE REGULATION
The debt limit portion of the proposed Gainful Employment regulation focuses on the ability of
recent graduates to repay loans in the early years of their post-schooling careers. The choice
of metrics to assure students' ability to pay is logically flawed. Furthermore, the basis of the
debt limit on earnings early in the career stands in contrast with standard economic analysis
of education, which clearly says that the choice of how much to borrow for schooling should
be based on the benefits of schooling, and not on the earnings level at the beginning of a
career. Any proposal aimed at helping students make smart decisions about investments in
education should compare the costs of schooling to the gains that accrue over the full career
as a result of that schooling. It should not compare costs to the level of earnings of recent
graduates.
The 8 Percent Debt Limit
Consider the 8 percent limit on debt payments. First, no scientific or data-driven rationale has
been presented for an 8 percent limit as opposed to any other number. No evidence has
been presented, for example, that loan default rates increase dramatically as student loan
payments cross this threshold. And though no specific rationale has been given for the 8
percent number, to the extent that it is based on rules of thumb from other types of borrowing,
such logic would also be flawed. Borrowing for the purpose of education is inherently
different from borrowing to purchase a home because the economic returns tend to be much
larger. Education is a source of social mobility, so any restriction on student borrowing will
tend to be harmful to students who would have enjoyed large gains from the schooling
investment.
Second, it cannot logically make sense to say that the average student cannot afford to pay 8
percent of her annual earnings to cover student loans for 10 years if those loans paid for
education that raised her earnings more than 8 percent each year for the rest of her working
life. Academic studies consistently find that students with more education on average earn
more, and that these returns to education are large on average, for example, in 2-year
Associate degree programs. Academic estimates of the return to education -the increase in
annual earnings that result from each additional year of schooling - have ranged from 8 to 15
percent, and have been rising steadily over the past 30 years. Our best estimate is that the
return per year of Associate degree schooling is currently more than 1 0 percent. This means
that those who earn a 2-year degree earn between 20 and 25 percent more each year than
those with just a high school education. If the average return to a year of schooling is 10
percent, this means that the average student in a 2-year program could pay 8 percent of her
earnings just with the additional earnings due to the schooling, even after accounting for
taxes.
These average benefits also do not include additional economic benefits of schooling.
Additional benefits of education include increased employment rates, increased likelihood of
health insurance coverage, decreased criminality and improved health.
Page 3
Report on Gainful Employment: Executive Summary
March 29, 2010 Charles River Associates
25th Percentile of Earnings
Many features of the proposed regulation share the flaw that limits on borrowing are based on
the estimated earnings that graduates would experience in the early years after graduation.
This is a flaw because the benefits of education accrue over a long period of time. A policy
aimed at protecting students would compare the benefits of education and the costs of
education. A key feature of education is that the costs are paid up front, both in terms of
foregone earnings and tuition, and the benefits accrue over the entire working life. To focus
exclusively on the short-term benefits is to ignore the long-term benefits and to cause
students to under-invest in education. This under-investment would likely be harmful to many
students affected by the proposed regulation.
One example is the use of the 25
1
h percentile of earnings as an implied ability to repay loans.
The proposed regulation places a limit on the median debt among students at a program.
This limit is based on the 25th percentile of earnings in occupations for which that program
prepares students. Presumably, the 25th percentile is meant to be an estimate of the typical
starting salary of graduates of the program. The use of the 25th percentile is flawed for
multiple reasons. Most fundamentally, the premise of limiting borrowing for education based
on early-career earnings is inappropriate and would be harmful to low-income students who
rely on student loans for access to education beyond high school.
In addition, the choice of the 25th percentile appears to be ad hoc, possibly based on the fact
that it is one of the statistics that the Bureau of Labor Statistics publishes for each occupation.
However, there is no scientific basis for using the 25th percentile of earnings as an estimate of
the early career experience of workers rather than some other percentile (and we reiterate
that limits should not be based on the level of earnings of early career workers in the first
place, but rather an estimate of the individual gains to earnings that result from the
education) . Other percentiles could be calculated fairly easily based on the same data used
by the Bureau of Labor Statistics.
Furthermore, the particular way to calculate the 25th percentile is not innocuous. Small
changes in the way one calculates this number causes large differences in the estimate of
early career earnings. One concern would be that future changes in the method of
calculating this number could have serious consequences. We estimate that differences in
earnings levels resulting from changes in how the 25th percentile is calculated would lead to
large differences in the number of students impacted by the proposed regulation. This
suggests that further consideration should be given to: (a) whether the 25th percentile concept
is appropriate, and (b) whether the method of calculating the student's estimated ability to pay
is overly sensitive to small changes in the future and valid from a scientific standpoint.
1 0-year Repayment
The proposed regulation indicates that annual loan payments should be based on a 1 0-year
repayment period. The use of the 1 0-year repayment length is another way that the
Page4
Report on Gainful Employment: Executive Summary
March 29, 2010 Charles River Associates
regulation would overweight the early costs of education and ignore the future benefits. As
mentioned above, the benefits of additional schooling accrue for the entire working career,
which for most students lasts significantly longer than 1 0 years, in many cases three times as
long. By calculating loan payments using a 1 0-year repayment, the regulation in a sense
imposes a decision rule on students that weights the costs of schooling two- to three-times as
much as the total benefits of schooling. Overweighting of costs in this way would be
shortsighted if it were done by students making education decisions. Any regulation that
imposes such a decision rule on students would cause many students who would otherwise
benefit from additional education to be harmed.
Further Analysis of the Methodology Used to Compute the Metrics of the
Regulation
The proposed regulation is based on a formula that has many moving parts. For each piece
of the debt limit formula, decisions must be made that determine the precise way the value
will be calculated. These decisions are not all straightforward, can have large effects on the
impact of the regulation, and have not to our knowledge been based on scientific research or
subject to sensitivity analyses. One prominent example is the method used to calculate the
25th percentile of earnings. While such a statistic sounds straightforward, it is not. Most
areas of study prepare students for multiple occupations, as defined by the Bureau of Labor
Statistics. There is not a singular method for calculating the 25th percentile of earnings
among workers entering multiple occupations. The method proposed by the Department of
Education during Negotiated Rulemaking indicates that a weighted average of 25
1
h
percentiles within each of the occupations should be used. Taking a weighted average of
25th percentiles within occupations does not, however, give the 25th percentile of earnings
among the workers in those occupations. Take, for example, the case of Business,
Management and Marketing. One of the occupations for which that area of study prepares
students, according to the Department of Education, is Chief Executive. Thus, the
Department of Education's method bases the early career earnings of students finishing
business management degrees in part on the 25
1
h percentile of earnings of chief executives.
We suspect that a recent college graduate has a vanishingly small chance of earning a chief
executive salary in the first few years after finishing school, though some will become chief
executives later in their careers.
Assuming the Department's goal is to generate an estimate of the early earnings of a
program's graduates, there are some types of occupations for which taking the weighted
average of percentiles may make sense, and others for which it does not. Even when a
weighted average may be appropriate, how to weight is an important question that must be
addressed. There is not a simple solution to the general problem of how to estimate the
future earnings of a program's graduates. and more thought and study should be given to
how to address it. One risk that should be addressed is that future decisions about which
occupations for which a program prepares students could have very large effects on the
resulting borrowing limits for that program. Since these decisions may be made for other
Page 5
Report on Gainful Employment: Executive Summary
March 29, 2010 Charles River Associates
reasons, or capriciously, the regulation would be at risk of having unintended consequences
based on future seemingly unrelated decisions.
There has also been no guidance as to what defines a program for the purpose of the
proposed regulation. For our analyses, we define a program as an area of study (i.e. a CIP
code) , at a particular location or campus, for a specific program length (i.e. less than 2-year,
2-year, or 4-year). An alternative would be to define a program based on an institution's
OPEID code rather than a campus. Doing so could create unintended incentives to combine
campuses into OPEID codes. This is another indication that the proposed regulation's
implementation at the school or institution level, rather than at the student level, is
problematic.
CONCLUSION
In summary, the proposed regulation is not currently formulated to address a specific problem
effectively. Secretary of Education Arne Duncan has stated publicly that he wants to ensure
that the Department understands its proposal thoroughly so that it prevents any "unintended
consequences." Our analysis suggests that the "unintended consequences"-cutting off
access to hundreds of thousands of students who want postsecondary education-will be
much more substantial than the intended consequence, which we believe to be-though we
are not certain-reducing the number of students who over borrow.
To start, the Department of Education has not clearly defined what the problem is that the
regulation aims to address. As discussed above, some perceived problems the regulation
may intend to address are not problems at all but rather a reflection of the fact that for-profit
postsecondary schools serve a very different population than not-for-profit postsecondary
schools. If the Department of Education wishes to address the problem that some students
take on excessive debt, the proposed regulation is not well designed to do so. By applying a
rule at the school or program level, many other students would be negatively affected. Our
analysis suggests that 33 percent of students currently in for-profit postsecondary schooling
would be denied access. Many more students would be denied access to postsecondary
schooling than would be protected from excessive borrowing.
Furthermore, it should not be assumed that public postsecondary institutions, particularly
community colleges, would absorb these students. Given the fiscal conditions of the states, it
is not obvious that community colleges will be able to increase capacity to meet the
increasing demand for postsecondary schooling.
Finally, because for-profit schools disproportionately serve racial and ethnic minority students
and students from low-income family backgrounds, the regulation would have the effect of
reducing access to higher education to groups of students that have historically had the
lowest levels of access.
Page6
PROPOSED DISCLOSURE ELEMENTS
AS THEY RELATE TO CERTAIN EDUCATIONAL PROGRAMS
Prospective students should have sufficient information made available to them at the time of the
enrollment process such that they can make an informed decision regarding their enrollment.
Information should include the cost of their program of study, a reasonable projection of
potential earnings in their chosen field upon graduation and throughout their life of employment
in that field, a reasonable estimate of the debt students typically incur to complete their program,
and the options they will have in selecting plans for repaying that debt. Thjs information should
be made available to the prospective student before incurring any financial obligation to the
school. This disclosure would substitute for the debt to earnings ratio proposed to define the
term "gainful employment" as used in provisions of Sections 1001 and 1002 ofthe HEA.
Under the Higher Education Act as most recently reauthorized, prospective students and
borrowers are required to be given much of this information. Among the information that
schools will be required to provide are:
The price of attendance, including tuition and fees, estimated costs of books and
supplies, room and board, and transportation costs, and any additional costs for a
program in which the student is enrolled or expresses an interest;
A net price calculator;
The prices for books;
Detailed financial aid information;
The retention rate of certificate- or degree-seeking, first-time, undergraduate students
as reported to IPEDS;
The completion or graduation rate of certificate- or degree-seeking, first-time, full-
time, undergraduate students; including data disaggregated by gender; major racial
and ethnic subgroup (as defined in IPEDS); recipients of a Federal Pell Grant;
recipients of a subsidized Stafford Loan who did not receive a Pell Grant; and
students who did not receive either a Pel! Grant or a subsidized Stafford Loan;
The placement in employment of, and types of employment obtained by, graduates of
the institution's degree or certificate programs;
The types of graduate and professional education in which graduates of the
institution's four-year degree programs enroll;
Information published by the Department of Education for students at any time that
information regarding loan availability is provided;
Detailed entrance and exit counseling for student borrowers, including average
anticipated monthly repayment amount, repayment plan options, and debt
management strategies; and
Detailed private education loan disclosures, including a student self-certification
form.
The additional information proposed to be disclosed here would complete the information for
prospective students by providing additional disclosure on the occupations which the student
may pursue based on his/her educational program, information relating to salary levels for those
occupations, average student loan indebtedness of graduates of the program, and projected
annual loan repayment amounts for the average level of indebtedness. Regulations pertaining to
the mjsrepresentation of information would apply to these disclosures.
668.41 Reporting and disclosure of information
(h) Prospective students - report on employment obtained by graduates of certain degree and
certificate programs. For a program offered by the institution under 668.8(c)(3) or (d), the
institution must provide a notice to prospective students, prior to the student enrolling in or
entering into a financial obligation to the institution, including but not limited to the following
information:
(i) identification of one or more occupations for which the program helps the student
prepare;
(ii) annual wage and salary information reported at the 25
111
and 75
1
h percentile for the
identified occupation(s) from the Department of Labor' s Occupational Information Network
(O*NET) or a link to O*NET with an explanation that the prospective student can find labor
market and wage and salary information on that site relating to employment in various
occupations;
(iii) wage and salary data for graduates from the most recently completed year for which
data are available, if the institution collects such data for the purposes of this section;
(iv) average federal student loan indebtedness of graduates of the institution with respect
to their attendance at that institution, on a program, degree-level , or institution-wide basis;
(v) average institutional loan indebtedness of graduates of a program, degree-level , or
institution-wide basis, if the institution provides institutional loans to its students as defined in
668.28(a)(5)(i);
(vi) the percentage of graduates who borrowed private student loans with respect to their
attendance at that institution, on a program, degree-level, or institution-wide basis; and
(vii) the expected annual loan repayment amounts for the average federal and
institutional student loan indebtedness, on a standard 10-year repayment plan and at least one
other repayment plan.
[Note - we would request preamble language on several issues, including the following:
1. For a program that does not have a SOC code identified to it or salary information
to a SOC code on the O*NET site, the institution could provide some alternative
information. An institution could simply provide the CIP code for the program and a list of SOC
Codes that are associated with that CIP Code.
2. The institution could identify several occupations but must identify the one that the
largest percentage enter into, if the institution collects that information.
3. Institutions need to be able to clarify for students that the average data provided may
not be accurate for their particular sUuation.
4. For pwposes of allowing institutions to provide actual salary data if they collect it, it
is important not to require institutions to disclose any information they may collect, as it may not
be comprehensive enough to give good consumer information.]
2
From: Kanter Martha
To: Private -Duncan. Ame
CC: Yuan. Georgia
Date: 4/3/2010 6:00:20 PM
Subject: FW: Request for Follow Up Meeting This Week to Discuss Gainful Employment
(b)(5)
From: Harris Miller [HarrisM@career.org]
Sent: Friday, April 02, 2010 1:25PM
To: Manheimer, Ann
Cc: Brian Moran; Tammy Halligan; Bergeron, David; Madzelan, Dan; Kanter, Martha; Shireman, Bob; Wolff, Russell;
Jenkins, Harold; Madzelan, Dan; Picoult, Francine; Finley, Steve; Yuan, Georgia; Smith, Kathleen; Arsenault, Leigh
Subject: RE: Request for Follow Up Meeting This Week to Discuss Gainful Employment
Ann: Thank you very much. Attached is the full research report for which I sent you and your colleagues the Executive
Summary yesterday. I realize it is a holiday weekend, but I encourage those attending the meeting to read it, if at all
possible. It is excellent work done by top notch scholars. I look forward to seeing you Monday. Harris
From: Manheimer, Ann [ mailto:Ann.Manheimer@ed.gov]
Sent: Friday, April 02, 2010 8:56AM
To: Harris Miiier
Cc: Brian Moran; Tammy Halligan; Bergeron, David; Madzelan, Dan; Kanter, Martha; Shireman, Bob; Wolff, Russell;
Jenkins, Harold; Madzelan, Dan; Picoult, Francine; Finley, Steve; Yuan, Georgia; Smith, Kathleen; Arsenault, Leigh
Subject: RE: Request for Follow Up Meeting This Week to Discuss Gainful Employment
Harris- from the Office of the Under Secretary, Bob Shireman, Leigh Arsenault and I will be attending; from the Office
of Postsecondary Education, Dan Madzelan and David Bergeron will be attending, and from the Office of General
Counsel, Harold Jenkins, Steve Finley, Russ W ol:ff, Steve Finley, and possible, Georgia Yuan, will be attending. We
look forward to hearing your presentation- Ann Manheimer, 260-1488
From: Harris Miller [mailto:HarrisM@career.org]
Sent Thursday, April 01, 2010 6:04 PM
To: Shireman, Bob; Kanter, Martha; Madzelan, Dan; Bergeron, David
Cc: Brian Moran; Tammy Halligan; Manheimer, Ann
Subject: RE: Request for Follow Up Meeting This Week to Discuss Gainful Employment
Bob, et.al.: We look forward to the meeting Monday morning at 9 AM at the Department to discuss the gainful
employment issue. Attending as part of the CCA delegation will be, in addition to me, Brian Moran, CCA's Executive
Vice President for Government Relations, Dr. Jonathan Guryan, University of Chicago, and Dr. Matthew Thompson of
Charles River Associates, the leads on the research we will be presenting, Rebecca Campoverde from Kaplan, Anthony
Guida from EDMC, and Tom Babel from Devry.
If you could please let me know who will be in attendance from the Department, that would be appreciated.
Attached you will find three documents. The first is the Executive Summary of the research findings. I will send you the
full report tomorrow, Friday. Last minute editing on the full report was delayed by the Passover holiday.
Second is the proposal CCA is putting forward as an alternative to the gainful employment proposal put forth by the
Department during the negotiated rulemaking.
Third is a document that summarizes the extensive tools the government already has to take actions against "bad actors"
in our sector.
Please let me know if you have any questions or comments. I look forward to a productive conversation Monday. Enjoy
the spectacular weather.
Harris
From: Shireman, Bob [mailto:Bob.Shireman@ed.gov]
Sent: Monday, March 29, 2010 10:48 AM
To: Harris Miller; Kanter, Martha; Madzelan, Dan; Bergeron, David
Cc: Brian Moran; Tammy HaJligan; Manheimer, Ann
Subject: RE: Request for Follow Up Meeting Ths Week to Discuss Gainful Employment
Work with Ann Manheimer on arranging the meeting. Please send your research and alternative prior to the meeting so
we can get concrete and specific in the discussions.
Thanks,
-Bob
From: Harris Miller [mailto:HarrisM@career.org]
Sent: Monday, March 29, 2010 10:04 AM
To: Kanter, Martha; Shireman, Bob; Madzelan, Dan; Bergeron, David
Cc: Brian Moran; Tammy HaJligan
Subject: Request for Follow Up Meeting This Week to Discuss Gainful Employment
Martha/Bob/Dan/David:
CCA would like to have a follow on meeting with you this week to the one Brian Moran and I had with Dan and David
ten days ago to discuss a) the results of our research (which was not complete when we met) on the impact of the
Department' s gainful employment proposal on students in higher education, and b) an alternative (which we discussed in
general concept with Dan and David) to the Department ofEducation proposal . We can make ourselves available at
almost any time starting tomorrow. Please let me know whom I should contact to arrange a meeting. Thanks in advance.
Harris
*****************************************************************
Harris N. Mlll er
CEO/President
Career College Association
1101 Connecticut Avenue, NW, Suite 900
Washington, DC 20036
harrism@career.org
+ 1 202 336 6754
Executive Assistant: Jackie McWilliams
j ackiem@career. org
+ 1 202 336 6706
www. career.org
[cid:image00l .jpg@OlCAD267.F04D6970]
C
D A ( River
.l'l""\.. As.suc1arcs
Prepared For:
Harris N. Miller
Career College Association
1101 Connecticut Ave. NW, Suite 900
Washington DC 20036
Report on
Gainful Employment
Prepared By:
Jonathan Guryan, Ph.D.,
Associate Professor of Economics, University of Chicago Booth School of Business
Matthew Thompson, Ph.D.,
Vice President
Charles River Associates
1545 Raymond Diehl Road, Suite 260
Tallahassee, FL 32308
Date: April 2, 2010
Report on Gainful Employment
April 2, 2010
Disclaimer
This report was prepared at the request of Harris Miller and the Career College Association. It is
based on data and information that were available at the time of the analyses. If additional data
or information become available we may update or modify our report.
Page i
Report on Gainful Employment
April 2 , 2010
Table of Contents
1. WHAT ARE POSSIBLE RATIONALES FOR THIS PROPOSAL? WHAT PROBLEM IS
THE PROPOSED REGULATION AIMING TO SOLVE? .... .... .... .... .... .... .... .... ..... .... .... .... 1
2. ARE THE PROBLEMS AS SEVERE AS ASSUMED? .................................................... ?
2 .1. COMPARING STUDENT CHARACTERISTICS .... .... .... ..... ... .... .... .. .. .... .... .... ..... ... .... .... .. .. .... .... .... 8
2 .2. DEFAULT RATES 14
3. WHAT IS THE RATIONALE FOR SUBSIDIZING LOANS FOR HIGHER EDUCATION?16
3.1. WHAT ARE THE BENEFITS OF HIGHER EDUCATION? .. .... ..... ... .... .... ..... .... ..... ... .... .... .. .. .... .... .. 17
3.2. PRESIDENT 0BAMA RECOGNIZES THE IMPORTANCE OF EDUCATION BOTH AS AN ENGINE OF
ECONOMIC GROWTH FOR THE COUNTRY AND AS A SOURCE OF ECONOMIC PROGRESS FOR
INDIVIDUALS FROM ALL CORNERS OF THE U.S. SOCIAL STRUCTURE ... .. .. .... ..... ... .... .... .... .... ... 18
4. WHAT ARE THE BENEFITS OF AN ASSOCIATE DEGREE EDUCATION IN
PARTICULAR? ... ..... ..... ... .... .... .... .... .... ..... .... .... .... ... ..... ..... .... .... .... .... ..... ..... ... .... .... .... ... 20
4 .1. THE PROPOSAL'S FOCUS ON ANNUAL INCOME IGNORES THE OTHER BENEFITS OF EDUCATION:21
5. WHAT EFFECTS MIGHT THE PROPOSED REGULATION HAVE? .... .. ..................... 24
5 .1. WHAT ARE THE EFFECTS OF INCREASED ACCESS TO FUNDING FOR HIGHER EDUCATION AND HOW
DO THESE VARY ACROSS DIFFERENT TYPES OF STUDENTS? .. .... .... .. .. .... .... .... ..................... .. 24
5.2. HOW TO MEASURE THE 25 PERCENTILE OF EARNINGS? .... .. .. .... ........... ............ ... .. .... .... .... . 24
5.3. PRELIMINARY ANALYSIS BASED ON DATA SUBMITTED BY CCA MEMBER INSTITUTIONS .... .... ... 30
6. WHAT ALTERNATIVE REGULATIONS OR POLICIES MIGHT BE SUGGESTED TO
ADDRESS THE PROBLEM AT HAND? ...... ........ ...... .................................. ........ .......... 36
6 .1. FURTHER CRITICISMS OF THE PROPOSED REGULATION ........................................................ 36
6 .2. TO DIRECTLY ADDRESS THE PROBLEM, REGULATION SHOULD FOCUS ON DISCLOSURE AND
ENSURING THAT STUDENTS MAKE INFORMED DECISIONS .... .... .... .. .. .... ..... .................... ..... .... 37
7. CONCLUSION .... .... .... .... .... ............. .... .... ........ ..... .... .... .... .... .... .... ..... .... .... .... .... ......... ... 38
8. APPENDIX A ...... .... .... ..... .... .... .... .... .... .... .... .... .... .... .... ..... .... .... .... .... .... .... .... .... .... .... .... . 39
Page ii
Report on Gainful Employment
April 2, 2010
This report addresses the definition of"gainful employment" proposed by the U.S. Department of
Education as a part of negotiated rulemaking. At present the proposal would define gainful employment
so that a program's students would be required to have a median debt level no greater than 8 percent of
the 25th percentile of annual earnings among individuals working in occupations for which that program
prepares students. The 25th percentile of earnings would be calculated from Bureau of Labor Statistics
(BLS) data, and is presumably meant to be an estimate of the typical starting annual earnings for
someone finishing that program. Annual loan payments would be calculated from the median debt level
based on a 1 0-year repayment plan using the interest rate on unsubsidized Stafford loans. For programs
that do not meet the 8 percent loan-to-income cut-off, an alternative is to have a 90 percent repayment
rate for recent graduates.
In this document, six basic areas are covered:
1. What are possible rationales for this proposal? What problem is the proposed regu-
lation aiming to solve?
2. Are the problems as severe as assumed?
3. What is the rationale for subsidizing loans for higher education?
4. What are the benefits of an Associate degree education in particular?
5. What effects might the proposed regulation have?
6. What alternative regulations or policies might be suggested to address the problem
at hand?
1. WHAT ARE POSSIBLE RATIONALES FOR THIS PROPOSAL?
WHAT PROBLEM IS THE PROPOSED REGULATION AIMING TO
SOLVE?
Presumably, the motivation behind the proposed regulation is to protect students from taking on "too
much debt". Taking on excessive debt may lead to an inability to repay the debt, resulting in default.
There may be a belief that some students agree to borrow so much there is little chance they will be able
to repay the loan in the future. There may also be a belief among policymakers that, regardless of
whether the loans are eventually paid back, some amounts of debt are too high per se.
1
The proposal's focus on for-profit schools implies that there is a belief that the problems of excessive debt
burden and high default rates are either specific to, or more severe at, for-profit schools. There has been
no analysis of whether differences in debt levels or differences in default or delinquency rates across
types of schools are the result of actions by the schools or due to differences in the types of students that
the schools serve.
1
Further, the proposed legislation assumes that individuals do not have the ability to determine appropriate levels of personal debt without
government guidance. One argument that the Department of Education may advance is that students do not have all of the
necessary information to make informed decisions, and thus government guidance is necessary.
Page 1
Report on Gainful Employment
April 2, 2010
Later in this document, we address whether in fact it appears that default rates are higher at for-profit
schools, and whether some of this difference is a result of serving students from different backgrounds.
The focus on for-profit schools suggests that another possible motivation for the proposed regulation is
that the Department of Education believes the cost offor-profit programs is too high. This sense that the
cost to students is greater at for-profit programs is surely based almost exclusively on a comparison of
tuitions. However, the full cost of schooling must also include foregone earnings if the student must
cease working to attend school, and other less obvious but very real costs. Since for-profit programs tend
to offer more flexibility both in terms of the timing and location of schooling, these types of costs tend to
be lower. Compare the costs for a student currently earning $30,000 per year who could continue to work
while completing a 2-year for-profit program, but who would have to stop working to attend a community
college program because it conflicts with his work schedule. Even if the tuition at the community college
were significantly less, the total cost to the student (tuition plus any foregone earnings) is likely lower at
the for-profit program.
It may be useful to discuss why policy makers should think of high default or debt levels as being
something students should be protected from. Consider high default rates. The negative effects of loan
default may include future difficulty securing loans. Without these costs, a defaulted loan is similar, from
the student's standpoint, to a grant. If there were no long-term penalty from defaulting it would be in the
student's interest to borrow monies he will not pay back. Thus, to the extent that the regulation's intent is
to protect students, it should be made clear that it should protect students from the penalties associated
with default, not from the funding stream that makes an education possible. One might imagine an
alternative regulation that was aimed at reducing the number of students who strategically take out loans
with no intention of repaying. But, it would seem that such a regulation would focus on the process by
which students are approved for loans, and on lenders rather than schools.
Next consider the concept of "too much debt". It is important to consider the purpose of the debt before
deeming it excessive. As we will discuss later in this document, the standard economic analysis of higher
education treats it as an investment. Since education so consistently yields high returns in the form of
increased earnings, lower unemployment rates, lower crime rates and even better health and longevity, it
can be a smart worthwhile investment to borrow even large amounts to be educated.
The question of how much debt is too much can be answered in different ways. The Department of
Education proposal focuses on the ability to make the associated loan payments relative to annual
income. Another way to view the decision is to ask whether taking on the debt and getting the education
increases the present value of a student's lifetime earnings. Those with more education tend to earn
more per year. This is of course a benefit.2 This benefit should be weighed against the costs. A
significant cost of education typically is to forego earnings while in school. The other main cost is tuition.
In Table 1 below, we calculate how much debt a student can take on such that comparing all of the costs
and benefits getting more education leads to an increase in lifetime earnings net of the debt costs.
Before we turn to these calculations, it may be helpful to consider both the benefits of education and the 8
percent proposed debt limit together. It is widely accepted among academic economists that each
2 There are other benefits of education that will be discussed later. The following calculation is conservative in that it ignores non-income
benefits of education.
Page2
Report on Gainful Employment
April 2, 2010
additional year of education increases earnings by some percentage. This percentage has varied from
about 7 to 15 percent over the past 40 years. Recent estimates peg this number currently near 15
percent. This means each year of schooling causes a student's annual earnings to be higher by 15
percent every year that she works. In the first year after school, her earnings are 15 percent higher than
they would be if she had not gone to school.
It cannot make sense to have a limit on debt payments that is less than the earnings return to
education. It does not make logical sense to say that she cannot afford to spend more than 8 percent of
her earnings to have that 15 percent bump in pay. If the government "protects" her from making these
debt payments, she loses the 15 percent annual bonus, and her expenditures are 8 percent lower. It
does students no favors to decrease their earnings by more than you decrease their required
expenditures. If she could not afford to make the debt payments, she certainly cannot afford not to make
them. She has less discretionary income in the world without the debt payments than she does in the
world with the debt payments.
This argument ignores the other costs of education, namely the foregone earnings while in school. Thus,
this argument does not imply that it is always a good investment to pay up to 15 percent of your income
per year to finance the tuition for a year of schooling. However, it very clearly shows that it is wrong to
say that someone is unable to pay more than 8 percent of her annual income to finance schooling. Thus,
to the extent that the proposal is meant to protect students from taking on debt payments they cannot
afford, it is misguided.
Here we consider not just whether students can afford these levels of debt, but how much one should be
willing to borrow to finance a 2-year program if the goal is to maximize lifetime earnings net of costs.
Each row of Table 1 shows the calculation for slightly different situations. In all cases, we consider a
student deciding whether to get two additional years of schooling at age 18. Consider the first row. Here
we evaluate the choice of a student who would earn $30,000 per year with a High School Degree, and
$34,992 per year if she spends two years earning her Associates Degree. This corresponds to an 8
percent increase in earnings for each year of schooling (i.e. an 8 percent "return to education"). This is a
fairly conservative estimate that may have been appropriate 30 years ago before returns to education
increased so dramatically. Column 2 denotes the rate at which the student discounts earnings and costs
that will come in the future. Much of the benefits of education come far in the future so how much these
are discounted are important. Column 3 shows the increase in lifetime earnings associated with the extra
education, in net present value. In other words, this is how much more someone with an Associate
Degree will earn over her lifetime than someone with a High School Degree, properly discounting to
account for the fact that much of the benefits will come many years in the future. Column 4 shows the
annual loan payments associated with the maximum debt someone could take on to cover tuition costs
and still not erase the amount in column 3. Column 5 shows the ratio of that annual debt payment to
annual earnings with an Associate Degree.
The calculation is repeated for more appropriate 1 o and 15 percent schooling returns, and for a 3 percent
discount rate. A 1 o percent return to schooling is closer to estimates of the return per year to a 2-year
Page 3
Report on Gainful Employment
April 2, 2010
college education (e.g. Kane and Rouse, 1995)3 which are based on the experiences of people who
graduated high school in the 1970s and 1980s. Estimates of contemporaneous returns to schooling tend
to be closer to 15 percent per year of education (e.g. Goldin and Katz, 2008).
Table 1: How much is really too much debt?
(1) (2) (3) (4) (5) (6) (7)
Maximum
annual debt
payments
Net present for 10 years Debt to
Return value of such that income% Annual Debt to
per year lifetime education a with max- payments if income%
of school- Discount earnings good in- imum 10 paid off over if paid in
ing rate difference vestment year loan 20 years 20 years
8% 5% $24,696.74 $3,135.44 9.00% $1,904.92 5.40%
10% 5% $45,783.80 $5,812.60 16.00% $3,531.42 9.70%
15% 5% $100,194.22 $12,720.42 35.00% $7,728.23 21 .30%
8% 3% $57,967.21 $6,786.99 19.40% $4,471.15 12.80%
10% 3% $88,196.71 $10,326.35 28.40% $6,802.83 18.70%
15% 3% $166,197.14 $19,458.89 53.60% $12,819.19 35.30%
Note: The calculations are for an individual who would earn $30,000 per year with a high school degree and $34,992, $36,300, or
$39,675 with a 2-year associate degree and an 8, 10, or 15 percent return to education per year of schooling.
A number of things should be noted from the calculations:
The increase in lifetime earnings associated with two additional years of education can be quite
large, even for someone who would have earned $30,000 per year without that schooling. For a 10
percent return to education, using a 5 percent discount rate, the lifetime earnings benefit is more
than $45,000. This number is net of the cost of foregone earnings during the two years while she is
in school.
4
This calculation implies that someone who is trying to maximize her lifetime earnings should be
willing to pay $45,000 for those two years of education. Paying anything less than that in tuition, the
schooling will benefit her over the course of her life.
3 This paper, co-authored by current member of the President's Council of Economic Advisors (CEA) Cecilia Rouse, shows that community
college and other 2-year programs yield approximately the same returns per credit hour as 4-year colleges.
4 If an individual is able to continue working while completing her two year education this benefit increases by as much as $60,000, the current
cost of foregone earnings.
Page 4
Report on Gainful Employment
April 2, 2010
The net present value of the increase in lifetime earnings depends significantly on how future
earnings streams are discounted. The more the future is discounted, the lower the benefits of
education are. Using a 3 percent discount rate, the lifetime earnings benefit of a 2-year degree is
almost $90,000 if the return is 1 o percent per year of schooling.
The return to education matters enormously in determining how much individuals should be willing to
pay for schooling. If the return is 15 percent per year, the present value increase in earnings from a
2-year program is more than $100,000 even using a 5% discount rate.
The Department of Education proposal essentially almost completely discounts (i.e. ignores) all
future benefits of education by focusing on the ability to pay in the years immediately following the
finish of school. This is present in the choice of estimated starting salary (i.e. the 25
1
h percentile
annual earnings) as the ability to pay, in the focus on repayment rates among recent graduates, and
in the use of the 10-year repayment schedule to calculate loan payments. The proposed regulation
is misguided in that it is not a function of the benefits of education. If the returns to education were to
continue to rise, as they have for the past 30 years, students would be restricted from borrowing
more to get this valuable training. Only the children of the rich (i.e. those who could afford to pay
tuition without borrowing) would be able to get this valuable education.
Using a 5 percent discount rate, and assuming a 10 percent return to schooling, a student who would
earn $30,000 per year with a high school education would earn $45,784 more over her lifetime if she
gets an Associate Degree. This calculation accounts both for the fact that she would spend two
years early in her life earning nothing while she is in school, and that the higher earnings associated
with education will come in the future. If she is able to work while in school then the increase in
lifetime earnings is even greater.
The student described above could pay close to $45,000 in tuition for the two years of schooling and
still end up ahead. If she borrowed to cover all of these tuition costs, her annual loan payments
would be $5,813 for the ten years she spends repaying, and in this time her loan payments would be
16% of her annual earnings (double the Department of Education limit).
While it would surely be difficult to make the payments during the 1 0-year repayment period, the
calculation shows that even taking on this high level of debt is a good investment for the student.
Any restriction on borrowing that is more stringent than the levels shown in Table 1 will lead the
student described to earn less over her lifetime.
One reason the loan payments in the table may appear high, even though taking on this much debt
is a good investment for the students, is that the repayment horizon is shorter than the time during
which the benefits of education accrue. The proposed regulation's use of a 1 0-year repayment rate
is another way in which it ignores the future benefits of education. If the student were to pay back
over 20 years instead of 10, the loan payment to income ratio for the student described above would
be 9. 7 percent rather than 16 percent. For the 8 percent return to education calculation, the 20-year
loan payment to income ratio would be 5.4 percent rather than 9.0 percent.
For a 15 percent return to education, the calculations indicate that one should be willing to pay
approximately 20 percent of his income for a 20-year repayment period. The reason this increases
lifetime earnings is that 15 of that 20 percent is accounted for by the earnings increase resulting from
Page 5
Report on Gainful Employment
April 2, 2010
the schooling. For the working years after the loan is repaid, the 15 percent benefit is enjoyed with
no cost.
A major part of the costs of education considered in the calculation is the foregone earnings the
student gives up if she attends school full-time. Schools that allow students to work and earn money
while in school are therefore less costly, even if the tuition charged is the same. The calculations
above would indicate a higher debt ceiling if foregone earnings were not considered as a cost.
The calculation above assumes that the difference in earnings between college and high school
educated individuals is the same at all ages. If instead the earnings of college educated students
start similar to those of high school educated students but grow faster, the role of discounting the
future is even more important.
Illustration: If the 8 percent loan limits were applied to medical school
Doctors spend much of their early years earning pay that is lower than they will earn in the long run, but
continuing to train on the job. Much of the earnings benefits of a medical education come when doctors
are far into their career. For such occupations, restricting debt levels to an amount that can be repaid
given early-career earnings would preclude borrowing for extremely valuable investments. Furthermore,
the lifetime benefits of a medical degree are quite large. Thus, for many it is worth making the investment
of time and large sums of money to obtain the degree. This is the case even though it is typical for
doctors to leave medical school with significant debt.
Here we show that if loan payments for medical school were limited to be 8 percent of the early earnings
of doctors, medical education would be largely restricted to students who could pay tuition costs without
much borrowing. In this case doctors would largely be drawn from wealthy families. The vast majority of
racial minority students and students whose parents have less than a college education would not be
allowed to become doctors. In fact more than half of the households in the U.S. do not have a net worth
high enough to pay the amount of medical school tuition that would not be covered by loans.
To complete the calculation, we use the median medical school tuition for non-resident programs from the
Association of American Medical Colleges web site. We ignore the costs of a medical education incurred
during residency years. The median four-year tuition at medical school is $168,840. If someone financed
80 percent of this, it would require borrowing $135,072. To pay this loan back over the 1 0-year horizon at
the unsubsidized Stafford loan rates would require an annual loan payment of $19,054 (or $1 ,588 per
month). To satisfy the proposed 8 percent criteria, someone would have to earn an annual salary of
$238,173. Payscale.com indicates that the typical annual salary of a doctor with 1-4 years of experience
is, in fact, $140,000. Given these calculations, the median medical school would not meet the proposed
standards.
We can also ask who would be able to afford medical school tuition if borrowing were only allowed up to
the limit implied by the 8 percent standard. Using the $140,000 annual salary, the maximum debt that
leaves annual payments no more than 8 percent of annual earnings is $79,397. Thus, to be able to
attend medical school, students and their families would have to find the funds to cover more than half of
the tuition costs, or $89,443.
Page6
Report on Gainful Employment
April 2, 2010
Table 2: Paying for Medical School
Median non-resident medical school tuition, 2009-10
Total tuition, 4 years non-resident
Debt required to pay 80 percent of tuition
Annual loan payment with 10 year repayment
Annual earnings required to satisfy 8 percent rule
Median salary of Doctors with 1-4 years experience
Maximum allowed debt to satisfy 8 percent rule
Remaining tuition that would have to be paid without borrow-
ing
Source: AAMC.org, Payscale.com.
(http :1/services.aamc.org/tsfreports/report_ median .cfm?yea r _of_ study=201 0)
$42,210
$168,840
$135,072
$19,054
$238,173
$140,000
$79,397
$89,443
How many families in the U.S. have the ability to pay nearly $90,000 without borrowing? The Survey of
Consumer Finances, sponsored by the Federal Reserve Board, indicates that in 2007, about one-third of
U.S. households had total net worth less than $90,000. This means that students living in about this
fraction of households would be precluded from becoming doctors. Many households who have slightly
more than $90,000 in net worth hold a good deal of that wealth as housing equity, meaning that they
would have to choose between owning a home and allowing their child to become a doctor. For minority
students, the problem would be even worse. The median net worth of non-whites and Hispanics in 2007
was $28,200. Thus, the vast majority of non-white and Hispanic students would likely not be able to
become doctors, regardless of their potential for success.
2. ARE THE PROBLEMS AS SEVERE AS ASSUMED?
The Department of Education's proposal presumably aims to address a specific perceived problem. In
this section we explore whether that problem is as severe as might be thought given summary measures
comparing for-profit and not-for-profit schools.
What is the problem that the proposal aims to address? We suspect that the Department of Education
sees the following problem: (a) a significant number of students take on more debt than they can afford to
repay upon entering the labor force and (b) this problem is more severe for students who attend for-profit
schools. We infer the latter because the proposed regulation treats for-profit schools differently than not-
for-profit schools. Here we address whether it is the case that default rates are strongly related to the for-
profit status of schools. The data strongly suggest that a large portion of the difference in default rates
between for-profit and not-for-profit schools is because for-profit and not-for-profit schools serve very
different student populations. For-profit schools are more likely to serve low-income, minority students.
Page?
Report on Gainful Employment
April 2, 2010
without parental support. They are more likely to serve students who are the first in their families to
attend college. Because of their access to outside resources, these groups of students are more likely to
default at both for-profit and not-for-profit schools. We estimate that if for-profit and not-for-profit schools
served the same population of students that the default rates would be significantly closer across the
types of schools. One might effectively argue that it is bad public policy to punish institutes of higher
education for serving students from groups who historically have not had wide access to schooling.
Furthermore. in part because the student populations are different, persistence rates at for-profit and not-
for-profit schools are different. Students who complete higher education programs are more likely to find
the jobs for which those programs prepare students. Some of the difference in default rates across types
of schools is accounted for by the higher completion rates at not-for-profit schools (particularly 4-year
programs). It would also seem to be bad policy to punish schools for the decisions by students not to
complete. At the very least, such a policy punishes the students who work to complete the program by
restricting access for all students, not just those who fail to complete.
Here we lay out the data on which the former conclusions are based. First, consider the difference in
characteristics of students by the type of school attended. Specifically, the data below come from the
Beginning Postsecondary Students Longitudinal Study, 1996 cohort. This is a survey administered by the
U.S. Department of Education that follows a nationally representative sample of students who entered
postsecondary education for the first time in 1996. We focus on this cohort of students rather than more
recent data because a follow-up survey has been done that allows for measurement of default rate by this
cohort.
2.1. COMPARING STUDENT CHARACTERISTICS
A series of figures shows that the demographic and socioeconomic background of students attending for-
profit and not-for-profit are different in systematic ways. All comparisons are based on the fi rst institution
attended, so those who begin at community college and transfer to a 4-year institution are categorized as
community college students. We first examine the characteristics of students who entered postsecondary
education in 1996, because these are the students for whom we can track default rates in the NCES data
(BPS) . For comparison, we also show comparable figures (Appendix A) calculated from the 2008
National Postsecondary Student Aid Study (NPSAS:08).
Figure 1 below shows the average age at which students begin postsecondary schooling. Beginning
students at for-profit schools are on average 25 years old. Students at not-for-profit public and private
programs of 2 years or less are on average 22-23 years old. In contrast, students entering 4-year not-for-
profit colleges are significantly younger, less than 19 years old on average. This difference shows that
the set of students entering these different programs are not the same. Though most of the differences
between for-profit and not-for-profit college students have held up since 1996, this is one that has
changed. While for-profit students continue to be older when they begin, the average age difference is
much closer today than it was in 1996.
Page 8
Report on Gainful Employment
April 2, 2010
Figure 1:
Average age at which students first enroll in postsecondary schooling
26.0
25.0
24.0
(I) 23.0
0)
<U
(I)
0)
22.0
<U
Qj
~
21.0
20.0
19.0
18.0
For-profit Private not-for- Private not-for- Publi c <=2 Publi c 4-year
profit <=2 profit 4-year
Age first enrolled
Notes: Calculated from the Beginning Postsecondary Students Longitudinal Survey, 1996 Cohort.
Next, we tum to the average income of students and/or their parents (Figure 2) . Specifically, we consider
the total income of either parents (for those who are dependents) or students (for those who are
independent at the time they apply) . Students at for-profit schools have the lowest income, prior to
entering college, of any group. VVhereas the average prior family or individual income of students
entering for-profit schools was $22,165 (in 1996 dollars), it was 71 percent higher among public 2-year-or-
less students, 154 percent higher (i.e. more than 2 %times) among public 4-year students, and 178
percent higher (i.e. close to 3 times) among private 4-year students.
Page 9
Report on Gainful Employment
April 2, 2010
Figure 2:
Average income of parents or independent students prior to school entry
$70.000 ~
$60.000
$50.000
Q)
E
0
$40.000
(.)
c
Q)
0'1
!
$30.000
,_
~
<{
$20.000
$10.000
$0
For-profit Private not- Private not- Publi c <=2 Publi c 4-year
for-profit <=2 f or-profit 4-year
Average income. parents or independent. 1994
Notes: Calculated from the Beginning Postsecondary Students Longitudinal Survey, 1996 Cohort.
For-profit students do not just come from families with lower average income, they are also more likely to
come from families that are quite poor. Consider the percent of students whose families collected AFDC
(Aid for Families with Dependent Children, the precursor to TANF, and commonly known as "welfare").
Figure 3 shows a full 16 percent of students who began at for-profit schools in 1996 came from families
collecting AFDC. This was more than double the rate of students attending not-for-profit 2-year-or-less
programs. At publ ic and private not-for-profit 4-year programs the corresponding rates are 2.6 and 1.6
percent.
Page 10
Report on Gainful Employment
April 2, 2010
Figure 3:
Percent of students from families who received AFDC prior to school entry
18
16
14
_,_
12
_,_
- 10 c -
<1>
0
lii
a.. 8
~ -
6
-'-
4
_.___
2 - '- -
0
L::_::_::_::_::_::_:
1;.;.:.:-:-:-:-.' L .I
"
For-profit Pri vate not-for- Privat e not-for- Publi c <=2 Publi c 4-year
profit <=2 profit 4-year
Percent receieved AFDC. 94-95
Notes: Calculated from the Beginning Postsecondary Students Longitudinal Survey, 1996 Cohort.
Students attending for-profit schools are also significantly more likely to be single parents at the time they
begin school. A full 28.5 percent of for-profit students were single parents prior to beginning school in
1996. At not-for-profit 2-year-or-less programs, less than 12 percent were single parents upon entering.
At 4-year programs single parents were extremely rare, less than 2 percent of students (Figure 4 below).
One interesting thing that has changed is that single parents are more likely today to attend college than
they were in 1996, but they are still significantly more likely to attend for-profit schools.
Page 11
Report on Gainful Employment
April 2, 2010
Figure 4:
Percent of students who were single parents prior to school entry
30 ~
25
20
c
<ll
15
0
....
<ll
Q_
10
For-profit Private not-for- Private not-for- Publi c <=2 Publi c 4-yeal'
profit <=2 profit 4-year
Percent single parent. 1995-96
Notes: Calculated from the Beginning Postsecondary Students Longitudinal Survey, 1996 Cohort.
Students at for-profit schools are also the first in their immediate family to attend college. Figure 5 shows
the fraction of students at each school type who have at least one parent with at least some college
education. The figure shows that these rates are lower at for-profit schools, particularly when compared
with 4-year not-for-profit schools.
Page 12
Report on Gainful Employment
April 2, 2010
Figure 5:
Percent of students whose parents attended at least some college
80.0
70 0
For profit
Not for profit
r--
-- --
,__.
r--
600
,.-
500
-
f.--
r---
c
4>
40.0 e
4>
Q.
r----",.-
-- -
r---
r--
300 - - '----- c-

20.0
r - - -- - -- -- r--- --
r--
100
I - -- i- - -- -- -- -
00

r
bJj _
-
til:;.

uf;
Prrvate for Pnvate lor Prrune for Public less- Public 2-year Public 4-year Prrvale not-fo1- Pnvate not-for-
prolit,less-thao- prol>l, 2-yeat profit. 4-year than-2-year 2-year profn 4-year
2-year
Percent whose pare11ts have some college or more
Notes: Calculated from the Beginning Postsecondary Students Longitudinal Survey, 1996 Cohort.
Finally, Figure 6 shows that for-profit schools are much more likely to serve students from racial and
ethnic minority groups. The fraction of students at for-profit schools who are either Black or Hispanic was
43.9, 37.2 and 51.4 percent at less than 2-year, 2-year and 4-year programs respectively. Private not-for-
profit less-than-2-year programs also are likely to serve Black or Hispanic students (35.9 percent) .
However, the share of students who are Black or Hispanic at all other not-for-profit school groups was
less than 25 percent.
Page 13
Report on Gainful Employment
April 2, 2010
Figure 6:
Percent of students who are Black or Hispanic
60 0
500
40.0
c:
30.0
"' 0..
2().0 ...._..
100
00
For profit
Not for profit
Pmatetor
prol'lt, tess4han.
2,yea
Puva!e not-for PriVate noHot Pt<Yale not-tor Pubht tess. Publie2-year Publit .t-year
profit tess-tha.,. profit 2-year profil4-year than 2Yel
2-year

Notes: Calculated from the Beginning Postsecondary Students Longitudinal Survey, 1996 Cohort.
2.2. D EFAULT RATES
Using the same Beginning Postsecondary Students Longitudinal Survey data, we also compared 6-year
default rates of students at different types of schools. The default rate is computed as the fraction of
students in the 1996 entering cohort with any student borrowing who ever default by 2001. This is
calculated in the same way as Mark Kantrowitz did in his recent report.
The light blue bars in Figure 7 below show the raw default rates of students starting in different types of
schools. All for-profit students are considered together. Because sample sizes are small in particular
groups, less-than-2-year and 2-year schools are combined. Without adjusting for the differences in
student background across the different school groups, the 6-year default rate is significantly higher at
for-profit schools than at not-for-profit schools. At for profit schools, almost 25 percent of the 1996 cohort
borrowers defaulted on at least one loan at some point by 2001. The default rate is 17.1 percent at
private not-for-profit 2-year-or-less programs, and 8.5 percent at public not-for-profit 2-year-or-less
programs. The rates at not-for-profit 4-year programs are both around 6.3 percent.
Page 14
Report on Gainful Employment
April 2, 2010
Figure 7:
6-year default rates by type of school, controlling for student characteristics
25.0%
20.0%
15.0%
10.0%
5.0%
0.0%
For-profit Pri vate NFP 2- Pri vate NFP 4- Publi c 2-year or Publi c 4-year
year or less year less
Default rates
Default rates. controlling for race. gender. fami ly income. persi stence. Peii/AFDC receipt. single
parenthood
Notes: Calculations from the Beginning Postsecondary Students Survey: 1996 cohort and 2001 follow-up. The light bars show the
fraction of students beginning in each type of school in 1996 who ever defaulted by 2001 . The dark bars show the default rates after
controlling for race, gender, persistence and completion, Pell grant receipt in 95-96, family AFDC receipt in 94-95, parent or own
income (if dependent), and dependency status.
The dark blue bars show our estimates of what the default rates would be if all schools had similar
entering student bodies. To estimate this, we run a regression of individual students' default status on
controls for their race/ethnicity, gender, family income, dependency status, whether they persisted or
completed their program, and whether they received a Pell grant. The dark blue bars show what we
estimate the default rates would be if all schools served white male students who are dependent, whose
parents earn between $60-75K per year, who completed their programs and who did not receive a Pell
grant.
The estimates show that if all schools served similar students with similar backgrounds and who
completed their programs, the differences in default rates between for-profit and not-for-profit schools
would narrow considerably. Whereas the difference in 6-year default rates between for-profit and public
4-year schools is 18.3 percentage points (24.6- 6.3) without controlling for the differences in student
characteristics, this difference is almost cut in half to 9.6 percentage points (12.2- 2.6) when adjustments
for student characteristics are made. This measure of default also overestimates the difference because
students in shorter programs have more time to default by 2001 . Furthermore, this narrowing is what
Page 15
Report on Gainful Employment
April 2, 2010
happens when adjustments are made for the characteristics that are observed in the BPS data. One
might suspect that there are other dimensions along which for-profit and not-for-profit students are
different. It is possible that if we had data on those characteristics and were able to adjust for them that
the difference in default rates would narrow more.
In addition to narrowing the difference in default rates between for-profrt and not-for-profit students,
adjusting for student characteristics also lowers the default rate at for-profit considerably. These
estimates imply that the default rate at for-profrt schools would be cut in half (from 24.6 to 12.2 percent) if
for-profit schools served wealthier non-minority students, and students who were more likely to complete.
3. WHAT IS THE RATIONALE FOR SUBSIDIZING LOANS FOR
HIGHER EDUCATION?
The standard economic analysis of education considers the choice of an individual whether to get an
additional year of schooling.
5
In this standard way of thinking, individuals weigh the costs and benefits of
schooling. The costs are the earnings foregone if one attends school full time, and tuition/fees. The
benefits include increased earnings in future years. Individuals choose to get more education so long as
the benefits are larger than the costs.
Education is an investment, meaning that the costs are paid up front and the benefits come in the future.
To properly weigh the costs and benefits, one must discount benefits that will not be realized for many
years. To simplify things, use the interest paid on savings accounts or the expected return on personal
investments as the discount rate.
Now consider the education choice of two students: one who has enough personal or family wealth to pay
tuition costs out of savings, the other who must borrow to finance the tuition costs.
For someone who would pay tuition costs out of savings, the decision comes down to comparing the
present value of increased lifetime earnings (the benefits) to the foregone earnings while in school and
the tuition (the costs) . If the benefits are greater than the costs, then the student should continue in her
schooling. If the costs are larger than the benefits, she should end her schooling and begin working.6
Compare this decision with someone who must borrow to pay the tuition costs. This student must
consider as costs the additional interest payments associated with the loan. Those payments must be
paid in the future. If the interest rate on the loan were equal to the interest rate used for discounting (in
this case the interest paid on savings) , then the decision would be the same for both students. Since the
unsubsidized interest rate charged on student loans is typically higher than the interest rate paid on
savings accounts, the cost of furthering education is higher for this student.
5 The standard reference is Human Capital by Gary Becker (University of Chicago), who won the Nobel Prize in Economics for this and other
work.
6 Vllhile it is necessary to consider as a cost the interest she does not earn on the money she takes out of saving to pay tuition, these interest
payments are discounted because they would have happened in the future. If we use the savings account interest rate as the
discount rate, the discounting eliminates this from consideration.
Page 16
Report on Gainful Employment
April 2, 2010
In short, because borrowing interest rates are higher than savings interest rates, the cost of schooling is
higher for those who must borrow to pay for higher education. Because these students almost by
definition come from poorer families, this problem creates access differences that relate to wealth,
socioeconomic status, and race. Subsidies for student loans are meant to narrow the difference between
borrowing and saving interest rates so that the costs of education are less related to family wealth.
Any restriction of access to debt financing for higher education will have the effect of decreasing access
more for poor and minority students. This is completely at odds with the intent and spirit of the Higher
Education Act.
Notice that the economic analysis of the schooling decision does not depend on the level of earnings.
Instead, it focuses on the increase in earnings resulting from the schooling. The proposal's focus on the
ability of students to pay back their loans quickly leads it to focus on the level of earnings.
This will have the effect of differentially punishing students with poor labor market prospects and who
would gain the most from higher education. Students with poor labor market prospects would have low
earnings, and likely high unemployment rates, without any higher education. Among these students, the
ones who would benefit greatly from additional focused schooling may end up in occupations with low
earnings. But, these earnings may be much higher than the student's personal alternative. The proposal
would limit how much this student could borrow based on the low level of earnings, and not based on the
large gains that would be realized from the doors opened by education.
3.1. WHAT ARE THE BENEFITS OF HIGHER EDUCATION?
Education is widely recognized as a source of social mobility. Though the US is regarded as a "land of
opportunity," correlations in earnings between fathers and sons are actually quite high. To understand
how much social mobility there is in the U.S., consider a family of four right at the poverty threshold.
Based on the best current estimates, it would on average take the descendants 5 or 6 generations before
their income is within 5 percent of the national average.?
What's more, studies find less social mobility among families with low net worth, suggesting that the
inability to borrow restricts social mobility. In other words, restrictions on borrowing (coming from poorly
functioning credit markets and high interest rates) makes being bom into wealth or poverty quite
determinative of earnings in adulthood.
One large reason for the effect of net worth and borrowing constraints on intergenerational mobility is
likely access to schooling. As an example consider the economic progress made by African Americans
during the past century. While legislative changes such as the Civil Rights Act of 1964 have led to
significant progress in reducing discrimination, economic research suggest very strongly that
improvements in educational opportunities for blacks have been more important. Until recently
successive cohorts of blacks have obtained progressively more education, and in turn their earnings have
caught up to whites.
7 Mazumder, Bhashkar, "Fortunate Sons: New Estimates of lntergenerational Mobility in the United States Using Social Security Earnings
Data," Review of Economics and Statistics 2005.
Page 17
Report on Gainful Employment
April 2, 2010
One of the most important ways in which the labor market in the U.S. has changed in the past 30 years is
that the benefits of education have increased dramatically. Those with more education have always
earned more on average than those with less. But, the difference in earnings between those with and
without a college education has increased sharply since 1980. In 1980, 4-year college graduates earned
about 40 percent more (about 1 o percent per year of additional schooling) than comparable high school
graduates. By 2005, the benefit of a college education was more than 70 percent (almost 18 percent per
year of schooling) . It is more important than it has been since the 1920s to be educated, and more
important than ever to get education beyond high school.
The changes that have led to this dramatic increase in the monetary benefits to education have also led
to very high levels of inequality. The difference in earnings and economic well-being between the rich
and the poor is also as large as it has been since the early part of the 20
1
h century. And, how much
education you have is significantly determinative of which side of that inequality you lie on. Those with
more education have benefited from the rich getting richer in the past 30 years. Those with less
education have been hurt terribly by the poor getting poorer during that same time.
All of this argues strongly that is as important as it has ever been to assure that all students who will
benefit have access to higher education. The social costs of restricted access are larger than they have
been in almost a century.
3.2. PRESIDENT 0BAMA RECOGNIZES THE IMPORTANCE OF EDUCATION BOTH AS AN ENGINE OF
ECONOMIC GROWTH FOR THE COUNTRY AND AS A SOURCE OF ECONOMIC PROGRESS FOR
INDIVIDUALS FROM ALL CORNERS OF THE U.S. SOCIAL STRUCTURE.
In a Washington Post column, published on July 12, 2009, President Obama called for increase in 5
million students with certificate or associate degree in the next 1 o years. Here are two quotes from what
he wrote:
"In an economy where jobs requiring at least an associate's degree are projected to grow twice as fast as
jobs requiring no college experience, it's never been more essential to continue education and training after
high school. That's why we've set a goal of leading the world in college degrees by 2020. Part of this goal
will be met by helping Americans better afford a college education." Barack Obama, The Washington Post,
July 12, 2009.
we believe it's time to reform our community colleges so that they provide Americans of all ages a chance
to team the skills and knowledge necessary to compete for the jobs of the future. Our community colleges
can serve as 21st-century job training centers, working with local businesses to help workers team the skills
they need to fill the jobs of the future. We can reallocate funding to help them modernize their facilities,
increase the quality of online courses and ultimately meet the goal of graduating 5 million more Americans
from community colleges by 2020." Barack Obama, The Washington Post, July 12, 2009.
What the President calls for cannot be done without increasing the capacity of community colleges and
for profit schools. Even with increased federal support, community colleges will face funding problems as
states continue to deal with severe fiscal problems. By decreasing access to the specific programs the
President wants to increase, the proposal will make the President's goal almost surely unattainable.
The for-profit education sector will be essential in helping the President to achieve this goal. For
example, consider that enrollment at for-profit colleges has grown significantly more than at non-profit
Page 18
Report on Gainful Employment
April 2, 2010
schools. In other words, as the demand for higher education has increased in recent years- likely as a
response to the increased returns described above- the non-profit sector has not been able to meet that
need. The for-profit sector has. Consider the growth rates of enrollment in public, private not-for-profit
and private for-profrt colleges over the past 5, 1 o and 20 years.
Overall enrollment in higher education has grown by more than 2 percent per year during the past 10
years. Capacity at neither public nor private not-for-profit colleges has grown fast enough to keep up with
this increased demand. Note that some of this demand comes from natural population growth, while
some comes from an increased desire to get a college education due to the high returns. Capacity
growth in the not-for-profit (public or private) sectors has not matched the overall increase in demand for
higher education over either the last 5, 10 or 20 years.
Table 3: Five, ten and twenty year enrollment growth by type of
institution, through 2007
Private Private
Not-for- For-
Total Public profit profit
Total percent growth in enroll-
ment:
20 years
39.78% 32.80% 33.60% 438.23%
10 years
25.79%
21 .10% 18.80% 225.60%
5 years
9.85% 5.80% 9.40% 99.60%
Average annual growth rate:
20 years
1.70%
1.40% 1.50% 8.80%
10 years
2.30%
1.90% 1.80% 13.70%
5 years
1.90% 1.10% 1.80% 14.80%
Source: Digest of Education Statistics, 2008, Table 188.
During that same time, the private for-profit sector has grown to meet the needs of students not-for-profit
schools cannot serve. Enrollment growth rates have been significantly higher. Though the private for-
profit sector is smaller than the other two sectors, the significantly higher growth rates have ensured that
overall enrollment could increase.
To meet President Obama's call for 5 million more Associate degree or certificate holders from 2-year
programs, capacity will have to increase in some or all of these sectors. Fiscal difficulties in the states are
likely to restrict community colleges and state colleges from meeting this need, even with increased
funding called for by the President. Given the historical role of for-profit schools in meeting increased
demand, it is likely that these schools will be the most able to expand quickly in response to this need.
Restrictions on student borrowing will curtail the for-profit sector from meeting these needs, and will make
it less likely that the President's goal will be met.
Page 19
Report on Gainful Employment
April 2, 2010
4. WHAT ARE THE BENEFITS OF AN ASSOCIATE DEGREE
EDUCATION IN PARTICULAR?
The Bureau of Labor Statistics administers a survey each month called the Current Population Survey.
This is the survey that is used to calculate the official unemployment rate that is released each month as
a barometer of the health of the economy. In that survey, individuals are asked questions about their
employment status, earnings, educational attainment and demographic information. Since 1996,
individuals have also been asked about their access to health insurance.
The following table presents estimates of the difference in various outcomes between those with an
Associate degree and those with exactly a high school education. We look separately at "academic" and
"vocational/occupational" Associate degrees. All estimates control for individuals' years of labor market
experience, and for changes over time in the outcomes that affect all individuals in the same way.
Positive effects on annual earnings: The way to read the table is the following. The first row shows
how much more those with an Associate degree earn on an annual basis than those with a high school
degree. For example, males with a Vocational/Occupational Associate degree earn 23.2 percent more
each year than males with a high school degree. If the average Associate program were 2 years, this
would correspond to an 11 .6 percent earnings return to each year of schooling.
Table 4: The Benefits of an Associate Degree Education
Benefit of Associate degree or some college,
relative to a high school education for:
Annual earnings
Percent employed full time
Percent employed
Percent with employer sponsored health insurance
Percent with any health insurance
Males
Associate:
Vocational/ Associate:
Occupational Academic
23.20%
3.50%
2.60%
10.10%
9.10%
27.40%
2.10%
2.50%
8.70%
9.30%
Females
Associate:
Vocational/ Associate:
Occupational Academic
27.20% 30.30%
-2.00% -0.70%
0.20% 0.50%
6.40% 8.40%
5.30% 6.70%
Note: Regression estimates from the Current Population Survey. The table shows the difference in various outcomes between those with an
Associate degree and those with exactly a high school education. Two types of Associate degrees are considered, vocational/occupational and
academic. These self-reported by the respondents to the Current Population Survey according to the guidelines described by the Bureau of
Labor Statistics.
Men who go on to get a vocational Associate degree earn 23.2 percent more each year than comparable
high school graduates. The return to an academic Associate degree is slightly larger for men -a
27.4percent increase in annual earnings. For women, the returns are even higher. Women who get a
vocational Associate degree earn 27.2 percent more than high school graduate women, and women who
Page 20
Report on Gainful Employment
April 2, 2010
get an academic Associate degree earn 30.3 percent more than women who stop schooling after
completing high school.
Positive effects on employment for men: The benefits of an Associate degree extend beyond the
earnings of those who work. Men with a vocational Associate degree are 3.5 percent more likely to be
employed than high school graduates. Men with an academic Associate degree are 2.1 percent more
likely to be employed than high school graduates. For women, there is no corresponding effect on
employment. Women with Associate degrees are actually less likely to be employed full-time than high
school graduates, but equally likely to be employed. Though there is no effect on employment for
women, it appears (based on the earnings effects and the health insurance effects discussed below) that
those who work are in better jobs than they would be if they did not get the additional education.
Positive effects on health insurance: An Associate degree education also helps to reduce the number
of uninsured. There is of course great interest currently in access to health insurance. For both men and
women, those with an Associate degree are significantly more likely to have health insurance. As
compared with high school graduates, men with a vocational Associate degree are 10.1 percent more
likely to have employer provided health insurance and 9.1 percent more likely to have health insurance of
any kind. Men with an academic Associate degree are 8.7 percent more likely to have employer provided
health insurance and 9.3 percent more likely to have health insurance of any kind. Women with a
vocational Associate degree are 6.4 percent more likely to have employer provided health insurance and
5.3 percent more likely to have health insurance of any kind. Men with an academic Associate degree
are 8.4 percent more likely to have employer provided health insurance and 6.7 percent more likely to
have health insurance of any kind.
4.1. THE PROPOSAL'S FOCUS ON ANNUAL INCOME IGNORES THE OTHER BENEFITS OF EDUCATION:
Academic research has shown that there are many benefits of education beyond the large increases in
annual income.
Reduced unemployment: The proposed regulation is based on the 25
1
h percentile of earnings among
those working. But, individuals with more education are less likely to be unemployed. It is perfectly
rational for a student to be willing to pay more than 8 percent of her annual income to avoid joblessness.
Insulation from recessions: Recessions typically hit the least educated the most severely. Consider the
current recession and the unemployment rates of people with different levels of education, in January
2008, January 2009 and January 2010, shown below.
Page 21
Report on Gainful Employment
April 2, 2010
Table 5: Unemployment Rates by Schooling Level
Jan- Jan- Jan-
08 09 10
Less than HS 7.7% 12.4% 15.2%
High School 4.6% 8.1% 10.1%
Associate or Some College 3.6% 6.4% 8.5%
Bachelor or more 2.1% 3.9% 4.9%
Source: The Employment Situation, Bureau of Labor Statistics.
First, notice that even in normal economic conditions, such as January 2008, unemployment rates are
strongly related to schooling levels. In addition to earning less, those with fewer years of schooling are
much more likely to experience unemployment. This is true when comparing students with a high school
education to students with some college and/or an Associate degree.
Second, notice that the increase in unemployment rates that has happened during the current severe
recession has impacted all groups, but the less educated more than others. Those with a high school
education saw their unemployment rates increase from 4.6 to 10.1 percent over the last two years, a 5.5
percentage point increase. In comparison, those with some college or an Associate degree saw their
unemployment rate increase by 4.9 percentage points. It is typical that recessions are differentially
burdensome on the least educated.
Furthermore, the most recent employment numbers for February 201 o show that while the overall
unemployment remained steady at 9.7 percent, this obscures very different experienced for more and
less educated Americans. For those with less than a high school degree and high school graduates, the
unemployment rate rose by 0.4 percentage points (15.2 to 15.6 percent for high school dropouts; 10.1 to
10.5 percent for high school graduates) . For those with a Bachelor's degree or more, the unemployment
rate remained essentially unchanged, rising from 4.9 to 5.0 percent. The only education group for which
the unemployment rate fell significantly this month was those with an Associate degree or some college.
For this group, the unemployment rate fell by 0.5 percentage points, from 8.5 to 8.0 percent).
Increased access to health care and health insurance: As shown above, those with more education
are more likely to have both employer-provided health insurance and any health insurance at all. This
relationship is partly explained by the positive effect of schooling on the likelihood of being employed, and
partly explained by the effect of education on income. Those with more income are more likely to be able
to afford health insurance.
For this reason, and possibly because more educated people make more informed decisions regarding
the management of their own health, individuals wffh more education tend to be healthier. Their mortality
rates are lower, they are less likely to smoke, more likely to exercise, more likely to engage in
preventative care, more likely to properly manage chronic conditions such as diabetes.
Page 22
Report on Gainful Employment
April 2, 2010
"In 1999, the age-adjusted mortality rate of high school dropouts ages 28-64 was more than twice as
large as the mortality rate for those with some college (Lyert et al. 2001 , table 26)."8
Comparing across countries, those with higher average education levels have longer life-expectancy
at birth (i.e. people live longer in countries with more educated populations) .
Some argue that this relationship is just an association, that either health causes people to get more
education, or that there is a third factor that causes both health and education to rise together. There are,
however, policy experiments that suggest this relation may be causal -that increasing education may
cause improvements in health.
Various studies show that laws that require children to complete more years of schooling (increases in the
compulsory schooling age) lead to improvements in health when those kids become adults.9
Educated mothers also have healthier babies: One study shows that increases in the number of
colleges nearby increases the likelihood that women attend college, and in turn makes them more likely
to have healthy babies. The study finds that the women were more likely to obtain prenatal care, and less
likely to smoke and drink alcohol during pregnancy. They were also more likely to be married at the time
they gave birth and had fewer children.10
Reduced criminality: There is evidence that obtaining more education makes it less likely that someone
will engage in crime. A study by Lochner and Moretti (2004) finds that compulsory schooling laws reduce
the likelihood that people become incarcerated. The effect is large for whites, and even larger for blacks.
They estimate that there is an additional15-25 percent benefit to each year of education in the form of
reduced crime that is not accounted for by the increased earnings that educated people enjoy. In other
words if we consider the reduced cost of crime imposed on society because of education, the total benefit
of education should be 15-25 percent larger than the increase in earnings that results from the additional
schooling.
General fulfillment: In addition to the benefits of education that are easily measurable, surely education
and learning brings some direct satisfaction. To the extent that students enjoy learning new ideas and
new skills, these are real benefits and they are not accounted for as a benefit of education if we just focus
on monetary earnings. People buy things all the time that bring them pleasure but no monetary return.
For example, consider vacations, televisions, tickets to sporting events, clothing, food. None of these
purchases increase earnings, but no one would dispute that these are reasonable things to buy.
To point out that the proposed regulation is misguided as a way to protect students from borrowing too
much, consider the following. Would there be support for a regulation that restricted individuals from
spending more than 8 percent of their annual earnings on food? This may sound ridiculous, but the logic
is quite similar to the proposal's. Through its effects on schools, the proposed restriction intends to
protect students from spending more than 8 percent of their annual earnings to be educated. Put this
8 Quoted in "Education and Health: Evaluating Theories and Evidence" chapter 2 in Making Americans Healthier: Social and Economic Policy
as Health Policy, ed. Robert F. Schoeni, James S. House and George A. Kaplan, Russell Sage Foundation, 2008.
9 See e.g. Lleras-Muney, 2005; Oreopolous, 2003; Arendt, 2005; Spasojevic, 2003.
10 Currie and Moretti, 2003.
Page 23
Report on Gainful Employment
April 2, 2010
way, the declaration that this is too much to spend on education is not very different logically from a
declaration that it would be too much to spend on any other good that people need or enjoy, such as food
or clothing.
5. WHAT EFFECTS MIGHT THE PROPOSED REGULATION HAVE?
5.1. WHAT ARE THE EFFECTS OF INCREASED ACCESS TO FUNDING FOR HIGHER EDUCATION AND
HOW DO THESE VARY ACROSS DIFFERENT TYPES OF STUDENTS?
There is relatively good scientific evidence of how college costs and the ability to borrow affects access to
higher education. The evidence is divided into two types: (1) estimates of the effect of reducing the price
of higher education, e.g. through grants, on college attendance, and (2) estimates of the effect of
increasing access to borrowing, e.g. through subsidized loans, on college attendance.
While estimates of the effect of eligibility for Pell grants are mixed, various studies of other sources of
grants find a significant effect of reducing the cost of college on college attendance. Studies of the G.l.
Bill and the Social Security student benefit find large effects of these grants on the likelihood that those
who are eligible go to college. Dynarski (2003) for example finds that an extra $1 ,000 grant (i.e. reduction
in tuition) increases college attendance by 4 percentage points.
There are fewer good studies of the effect of access to loans on college attendance. Reyes (1995)
shows that when loan eligibility changed differentially across income groups in the early 1980's, college
enrollment rates increased for the groups for whom loan eligibility increased. Dynarski (2005) finds
positive but smaller effects of loan eligibility on college attendance based on a study of changes in
eligibility induced by the Higher Education Amendments of 1992.
A more recent experimental study may be directly relevant.11 A group of researchers simplified the
FAFSA and worked with H&R Block to automatically fill out the form using information already entered
from individuals' 1040 tax forms. For randomly selected households, H&R Block pre-populated the
FAFSA form and offered to assist the family in filling out the form. Relative to a randomly selected
comparison group, the assistance increased college enrollment significantly both for recent high school
graduates and for older independent students with no college experience. There was no effect on a
second treatment group who were just given information about the FAFSA but no assistance. These
results show that barriers to the availability of financing restrict access to higher education. Based in part
on this research, the President and Secretary of Education recently announced that the FAFSA form will
be drastically simplified.12
5.2. HOW TO MEASURE THE 25TH PERCENTILE OF EARNINGS?
The proposed regulation places a limit on the median debt among students at a program. This limit is
based on the 25
1
h percentile of earnings in occupations for which that program prepares students.
11 Bettinger, Eric, Bridget Terry Long and Philip Oreopolous, "Increasing Postsecondary Enrollment Among Low-Income Families: A Project to
Improve Access to College Information and Financial Aid" (http://gseacademic.harvard.edul-longbr/FAFSA_Project_-
_Bettinger _Long_ Oreopoulos_ -_Description_1-09.pdf)
12 http://www2. ed .gov /news/pressrelea ses/2009/06/06242009 .html
Page 24
Report on Gainful Employment
April 2, 2010
Presumably, the 25th percentile is meant to be an estimate of the typical starting salary of graduates of the
program. However, it is clear that the proposal has the potential to act as a limit on tuitions that for-profit
schools will be able to charge. The extent of these limits will depend on how the 25th percentile of
earnings for a given area of study (CIP code) is determined. In Appendix A of the proposed regulations,
the Department of Education has provided a step-by-step method for calculating what it considers to be
the 25th percentile of earnings for a particular CIP code.
The Department proposes using the wage, earnings and employment data that are regularly collected by
the Bureau of Labor Statistics. However, since the employment information is reported by occupation
based on the Standard Occupational Classification (SOC) system, rather than area of study (CIP codes) ,
areas of study must be linked to one (or more) occupations. According to the Department's methodology
the 25th percentile of earnings for each program (based on the 6-digit CIP code) can be calculated using
the following method:
First, determine all occupations based on the SOC codes available from the 0-Net crosswalk
(http://online.onetcenter.org/crosswalk/CIP/) that are associated with each 6-digit CIP code.
Next, for each soc code determine employment and annual 25th percentile wages using data
from the Bureau of Labor Statistics (BLS) available at http://www.bls.gov/oes/current/oes_stru.htm.
Finally, based on the above values calculate for each CIP code the weighted average of the
annual 25th percentile wages using the total employment of each soc code as the weights.
According to the Department, this weighted average represents the 25th percentile of earnings for
each 6-digit CIP code.
While the Department's calculations are clear and concise, they are neither simple nor correct. The
Department's choice of how to calculate the 25th percentile is also far from innocuous. Below, we
describe three ways in which the calculation of expected earnings can be quite sensitive to choices
concerning the method. All of these choices are made either explicitly or implicitly, and all of them can
have significant effects both on the earnings levels and on the ultimate impact of the proposed regulation.
These examples also point out that future changes in seemingly technical inputs, such as which
occupations are matched in the CIP to soc crosswalk, have the potential to have large impacts on
programs and students.
In calculating the earnings measure the Department makes assumptions regarding the occupations for
which a graduate is likely to enter and the relative importance of each of these occupations in determining
earnings. In addition, the calculated earnings measure is not the 25th percentile across the SOC codes.
Modifying either the assumptions or the method for calculating earnings can have substantial impacts on
whether a program meets the 25th percentile/8 percent rule. Even without explicit changes from the
Department, programs may change from meeting the proposed regulation to not meeting the proposed
regulation because of future changes in BLS coding or employment patterns.
The correspondence between CIP codes and BLS occupation codes is important:
Table 6 below shows the weighted average for the Culinary Arts/Chef Training area of study based on the
25th percentiles of the occupations that are assigned to that CIP code. According to the CIP to SOC
crosswalk that is used by the Department to determine the "25th Percentile" this area of study includes
four occupations: Chefs and Head Cooks; Cooks, Private Household; Cooks, Restaurant; and Cooks, All
Page 25
Report on Gainful Employment
April 2, 2010
Other. While Culinary Arts programs are designed to train Chefs as defined in the first occupational
category, the Department's definition appears to include those individuals working as cooks at fast food
restaurants and cafeterias, and short order cooks. While the majority of students who complete Culinary
Arts/Chef Training programs do not work at fast food restaurants, these workers' low earnings would be
used to estimate graduates' ability to afford student loan payments.
Table 6: Department of Education
Calculation of the 25th Percentile
Weighted 25th Number
Area of Study Average Percentile Employed
12.0503 Culinary Arts/Chef Training $19,278
35-1011.00 Chefs and Head Cooks $29,050 98,040
35-2013.00 Cooks, Private Household $19,030 960
35-2014.00 Cooks, Restaurant $18,230 899,620
35-2019.00 Cooks, All Other $18,390 17,340
52.0201 Business Administration/Management $62,379
11-1011 .00 Chief Executives $102,080 301 ,930
11-1021.00 General and Operations Managers $62,900 1,697,690
11-2022.00 Sales Managers $65,350 333,910
11-3011.00 Administrative Services Managers $52,240 246,930
11-3051 .00 Industrial Production Managers $64,390 154,030
11-3071 .01 Transportation Managers $59,830 96,300
11-9021.00 Construction Managers $60,650 220,550
11-9151 .00 Social and Community Service Managers $42,110 117,150
11-9199.00 Managers, All Other $64,440 365,460
13-1051 .00 Cost Estimators $42,720 218,400
13-1111.00 Management Analysts $54,890 535,850
25-1011 .00 Business Teachers, Postsecondary $46,400 69,690
51.3501 Massage Therapy/Therapeutic Massage $45,777
25-1071 .00 Health Specialties Teachers, Postsecondary $54,850 125,100
31-9011.00 Massage Therapists $23,630 51 ,250
The choice to use a weighted average of 2Sh percentiles is important:
More general areas of study are mapped to many occupations. To calculate the 25th percentile of
earnings among graduates of a program, the Department's method takes a weighted average of the 25
1
h
percentiles in each of the assigned occupations. Taking a weighted average of 25th percentiles within
occupations does not, however, give the 25th percentile of earnings among the workers in those
occupations. Take, for example, the case of Business Administration/Management (shown in the table
above). One of the occupations for which that area of study prepares students, according to the
Department of Education, is Chief Executive. Thus, the Department of Education's method bases the
Page 26
Report on Gainful Employment
April 2, 2010
early career earnings of students finishing with Business Management degrees in part on the 25th
percentile of earnings of Chief Executives. We suspect that a recent college graduate has a vanishingly
small chance of earning a Chief Executive salary in the first few years after finishing school , though some
will become Chief Executives later in their careers. An important implication of this example is that the
allowable debt levels would be very sensitive to future decisions concerning which occupations match to
each CIP code. Removing 'Chief Executive' from the set of occupations for which a Business
Administration/Management program prepares students, for example, would significantly lower the
estimated earnings, and in turn the allowable debt.
As another example, consider trying to calculate the 25
1
h percentile of earnings among workers in two
equally large states: a very high-wage state and a very low-wage state. To make the illustration clear,
imagine the extreme situation in which the lowest-paid worker in the high wage state earns twice as much
as the highest-paid worker in the low-wage state. The average of the 25th percentiles will fall somewhere
in the range between the highest-paid worker from the low-wage state and the lowest-paid worker from
the high-wage state. However, the 25
1
h percentile earner among all the workers in both states is
someone in the middle of the pack in the low-wage state.
The choice to take a weighted average of percentiles may be appropriate in some situations, and the
determination depends on the way the relevant occupations are defined. Consider, for example, that a
student leaving a program has a 50 percent chance of entering occupation X and a 50 percent chance of
entering occupation Y. This student will remain in either of these parallel occupations for his career. In
this case, the average of the 25
1
h percentiles in occupations X and Y may be an appropriate estimate of
his early career earnings experience.
In contrast, consider a student leaving a program who will enter the entry-level occupation A after which
he will eventually progress to occupation B. In this case, the weighted average of 25th percentiles does
not tell us about his experience early in his career.
In some cases occupations are defined by the Bureau of Labor Statistics to correspond to a typology
represented by occupations X andY, and in other cases (as illustrated by some of the examples in the
table above) occupations are defined to correspond more closely to occupations A and B.
The estimate of earnings does not distinguish by degree level:
Furthermore, the CIP code is not specific to a level of degree, but rather just to the area of study.
Therefore, an individual with an Associate's degree in Business Administration/Management will have the
same CIP code as an individual with a Master' s degree in Business Administration/Management. Thus,
the Department's assessment of earnings (and measure of affordable debt) will be the same for these two
individuals.
How to weight information from different occupations is important:
Assuming the Department's goal is to generate an estimate of the early earnings of a program's
graduates given that they may be prepared for multiple occupations, a weighted average of percentiles
may make sense. Even when a weighted average may be appropriate, how to weight is an important
question that must be addressed. The Department's current approach of using total employment in the
full labor market (and not specific to either degree earners or for-profit students) is likely inappropriate in
many situations. For example, the table above shows the occupations associated with Massage
Page 27
Report on Gainful Employment
April 2, 2010
Therapy/Therapeutic Massage. In this case the teachers of health specialties (which includes Massage
Therapy and other health specialties) receives more than 70 percent of the weighted average when it
likely represents a much smaller percentage of Massage Therapy graduate placements.
From a mathematical point of view, the problem is that percentiles are not linear. As a result the average
of percentiles within groups is not the percentile of the full population. It is therefore possible that the
Department of Education's method for calculating the 25th percentile of earnings would not survive the
rulemaking process. For this reason, in our calculations of impact below we present estimates that are
based on an alternative method of calculating the 25th percentile among the workers in the occupations
that match to an area of study.
What is the 2Sh Percentile?
As noted above the Department's calculated earnings measure is not the 25
1
h percentile of the
occupations that are assigned to a CIP code. An arguably more appropriate measure of the 25th
percentile can be obtained by sorting the individual earnings information of all individuals in occupations
assigned to a given CIP code and determining the earnings at the 25th percentile of that set of workers.
To do this, we first obtained a crosswalk between CIP codes and BLS occupation codes from the National
Center for Education Statistics (NCES), a division of the U.S. Department of Education. We then merged
this information with earnings data from the Current Population Survey (CPS) March Annual Demographic
File. Each March, the CPS includes more-detailed questions about annual earnings and demographics.
For each CIP code, we sorted the annual earnings of individuals in the occupations that were matched to
that CIP code, and calculated the 25th percentile of annual earnings. The table below compares the
difference in "25th percentiles" based on the alternative methods of selected CIP codes. In addition, for
each CIP code we have computed the implied maximum debt allowed based on an 8 percent limit on
annual loan payments (assuming a 1 0-year repayment schedule at 6.8 percent interest). As is clear from
the table, the maximum debt can vary substantially depending on the calculation of the 25th percentile.
Page 28
Report on Gainful Employment
April 2, 2010
Table 7: Comparison of 25m Percentile Earnings and Maximum Debt Level f or
Selected CIP Codes
Full-Time Earners De artment of Education
CIP 25th Maximum 25th Maximum
Code CIP Description Percentile Debt Percentile Debt
10.0202 Radio & Television Broadcasting $27,000 $15,312 $27,207 $15,430
Technology/Technician
$7.940 1
12.0503 Culinary Arts/Chef Training $14,000 $19,278 $10,933
14.0901 Computer Engineering, General $48,000 $27,222 1 $73,752 $41 ,826
14.0903 Computer Software Engineering $47,000 $26,655 ' $73,791 $41 ,848
(New)
14.1001 Electrical, Electronics & Communi- $45,000 $25,520 $75,437 $42,782
cations Engineering
15.1301 Drafting & Design Technolo- $35,000 $19,849 $35,266 $20,000
gy!Technician, General
15.1399 Drafting/ Design Engineering Tech- $35,000 $19,849 $35,130 $19,923
nologies/Technicians, Other (New)
31.0501 Health & Physical Education, Gen- $32,000 $18,148 ! $19,927 $11 ,301
era I
31 .0504 Sport & Fitness Administra- $32,000 $18,148 $18,989 $10,769
lion/Management
43.0203 Fire Science/Firefighting $45,000
$25,520 I
$31 ,532 $17,883
47.0201 Heating, Air Conditioning, Ventilation $29,700 $16,843 $31 ,070 $17,620
& Refrigeration Maintenance Tech-
nology!Technician
51 .2001 Pharmacy (PharmD, BS/BPharm) $41 ,000 $23,252 1 $80,585 $45,701
51.3501 Massage Therapy/Therapeutic Mas- $36,000 $20,416 $45,777 $25,961
sage
52.0201 Business Administra- $45,000 $25,520 l $62,379 $35,376
!ion/Management
I
52.0408 General Office Occupations & Cieri- $23,000 $13,044 $23,239 $13,179
cal Services
52.1902 Fashion Merchandising $35,000 $19,849 I
$36,460 $20,677
52.1904 Apparel and Accessories Marketing $35,000 $19,849 $36,460 $20,677
Operations
As shown above, the particular way to calculate the 25th percentile is not innocuous. Small changes in
the way one calculates this number causes large differences in the estimate of early career earnings.
One concern would be that future changes in the method of calculating this number could have serious
consequences. We estimate that differences in earnings levels resulting from changes in how the 25th
percentile is calculated would lead to large differences in the number of students impacted by the
proposed regulation. This suggests that further consideration should be given to: (a) whether the 25th
percenti le concept is appropriate, and (b) whether the method of calculating the student's estimated
ability to pay is overly sensitive to small changes in the future and valid from a scientific standpoint.
Page 29
Report on Gainful Employment
April 2, 2010
5.3. PRELIMINARY ANALYSIS BASED ON DATA SUBMITTED BY CCA MEMBER INSTITUTIONS
To estimate the impact of the proposed regulation on the Title IV eligible for-profit postsecondary
institutions, we collected data from Career College Association (CCA) member institutions. Specifically,
we collected student/loan level data from each institution based on the population included in their 2006,
2007 and 2008 Cohort Default Rate calculation. These data include information on student loans and
default status on all students entering repayment during a given cohort year for 3 years after entering
repayment, and are the actual data that the institution's cohort default rate is based upon. We also
received individual level demographic data from each institution including race, gender, program of study
(CIP code) , OPEID, campus information, total loan amounts (both public and private) , and length of
program. In all, we received data from 17 different institutions, representing approximately 450
campuses, 640,000 students and over 10,000 separate programs from institutions ranging from very
small to very large.
In order to determine the impact of the proposed regulation on the CCA schools for which we received
data, we calculated both the 25th percentile of earnings based on the methodology used by the
Department of Education, and the median debt of graduating students from each of the schools and
programs with available data. In addition, we calculated the 25th percentile of earnings for each CIP
code based on full-time earners in the CPS data. This alternative 25
1
h percentile of earnings calculation
was done according to the method described in the previous section.
The median total loan amounts (public and private) accrued by graduating students were calculated for
each school, OPEID, campus, program length, and 6-digit CIP code. The Department of Education's
methodology requires that students who do not take any loans (public or private) should be included in
the median calculation as having accrued o loans. Since the data we have available for the CCA schools
only include students who have taken some form of government loan, we needed to impute the number of
students not taking any loans. Also, some schools did not provide data on the private loans taken by
students so we needed to impute the value of private loans in these instances. We do not have data on
students who do not take any public loans, but take private loans. We have not included any adjustment
for these individuals.1 3
In order to account for students not taking any loans, we used I PEDS data to calculate the average
percent of students in private, for-profit institutions that do not take any loans (approximately 20 percent).
Since the population that we observe in the data are only 80 percent of the total population that should
be included in the calculations we use the 37 .5th percentile of total loans amounts instead of the 50th
percentile as this would impute a total of 20 percent of the total population as having o loans (since they
would all be below the median).
In cases where no data was available to assess the amount of private loans taken, we multiplied the
value of public loans by 1.47 since the average percent of total loans that were public (based on
NPSAS:2008 data from the NCES) was approximately 70 percent.
13 For most schools we do not have information on loans students took at prior postsecondary institutions. As a result, we underestimate
median total debt, thus possibly underestimate the fraction of programs that would be impacted by the proposed debt limit rule.
Page 30
Report on Gainful Employment
April 2, 2010
Based on the 25th percentile of earnings determined above, we calculated the maximum amount of debt
that could be accrued using the 8 percent rule proposed by the Department of Education (assuming a 10-
year repayment schedule at 6.8percent interest). Comparing the maximum debt value with the median
debt actually accrued from the students in each program, we determined the programs which would
currently be impacted by the proposed regulation (i.e. the programs whose median debt was higher than
the maximum allowed under the regulation's guidelines).
As shown in Table 8, our analysis implies that approximately 18 percent of the programs we examined
would be impacted by the 8-percent/25
1
h-percentile rule when using the Department of Education's
income calculation. Using the CPS full-time earners results in nearly 25 percent of the programs being
impacted. The impact is disproportionately on larger programs as nearly 34 percent of students are
impacted using the Department's approach, and almost 50 percent are impacted using the CPS full-time
earners. In the sample of students analyzed, approximately 29 percent of black students and 35 percent
of Hispanic students would be in programs impacted by the proposed regulation. In addition, 25 percent
of women are in programs that would be affected. All of these percentages are higher when the
alternative measure of the 25
1
h percentile of earnings is used to evaluate programs.
Table 8: Percent of CIPs Impacted by Proposed Regulation
(Assuming 8 Percent Debt Ratio Using the 25th Percentile of Income)
Median Loan Based on Graduates
Total Percent of Percent of Percent of Percent of
Number of Programs Students Females Blacks
Programs Impacted Impacted Impacted Impacted
Department of Education 25
111
Percentile 10,725 18.19% 33.72% 24.79% 28.91%
CPS 25
111
Percentile 10,695 24.58% 49.10% 39.95% 44.91%
Source: Data provided by CCA member institutions.
Percent of Percent of
Hispanics Asians
Impacted Impacted
34.89% 44.26%
47.40% 65.14%
As shown in Table 9 below, if the sample of schools and programs used in the analysis is representative
of the full set of for-profit schools and programs- and we caution that not enough analysis has been done
yet to ascertain whether this is a reasonable assumption- these estimates imply that each year 361 ,000
students, including 68,300 non-Hispanic black students, 78,500 Hispanic, and 179,000 women, would
Page 31
Report on Gainful Employment
April 2, 2010
enter for-profit postsecondary programs that would lose eligibility for participation in the federal Title IV
financial aid programs.14
While some capacity may exist in other sectors of higher education to absorb these students, recent
reports indicate that the most likely alternatives-community colleges-are already oversubscribed in
many cases, and are facing further financial cutbacks as the states that provide much of their funding
face severe financial challenges. Given recent growth rates at for-profit postsecondary institutions, we
estimate that by 2020 approximately 1 million non-Hispanic black students and an additional 1 million
Hispanic students are on track to attend programs that would be adversely affected, and would be denied
access as a result.15
14
The annual flow of students in for-profit programs is estimated from the 12-month enrollment reported in the IPEDS. Since the IPEDS
figures provide the stock of students enrolled at a given point in time we divide the number of students enrolled in a 4-year program
by 4, the number of students enrolled in a 2-year program by 2, and then add those results to the number enrolled in less than 2-
year programs to obtain an estimate of the flow of students Into for-profit schools. This is likely an underestimate of the flow
because all students do not stay enrolled for the full length of the program and institutions are categorized base on the longest
program offered (so, some students recorded in a 4-year program are enrolled in something less than four years).
15 Estimates based on the CPS full-time earners and estimates of impacted students by state are provided in the appendix.
Page 32
Report on Gainful Employment
April 2, 2010
Year
Table 9: Estimated Number of Students Impacted by 2020
Median Loan Based on Graduates
25th Percentile Based on Department of Education Calculation
Number of
Total Number of African- Number of
Number of Female American Hispanic
Students Students Students Students
Impacted Impacted Impacted Impacted
Number of
Asian
Students
Impacted
Using the Department of Education's 25th percentile of annual earnings and 8% Debt-to-Earnings Ratio
2011 361,172 179,149 68,348 78,545
2012 392,955 194,914 74, 363 85,456
2013 427,535 212,066 80,907 92,977
2014 465,158 230,728 88,027 101,159
2015 506,092 251,032 95,773 110,060
2016 550,628 273,123 104,201 119,746
2017 599,084 297,158 113,371 130,283
2018 651,803 323,307 123,347 141,748
2019 709,162 351,759 134,202 154,222
2020 771,568 382,713 146,012 167,794
Total Students Impacted 5,435,157 2,695,948 1,028,550 1,181,990
Note: The number of impacted students assumes that the CCA data is representative of all for-profit
schools, that for-profit schools will cont inue to grow at 8.8% per year (the growth rate over the last five
years), and the relat ive student composition does not change during this period.
15,875
17,272
18,792
20,445
22,245
24,202
26,332
28,649
31,170
33,913
238,895
Based on our estimates, the impact of the regulation would vary across types of programs. Because the
limits on borrowing do not vary with the length of program, longer programs would be more severely
impacted. Whereas approximately 18 percent of students in less than 2 year programs would be
impacted, we estimate that approximately 40 percent of students in 2- and 4-year programs would be
impacted. Table 10 below shows the percent of programs and students impacted by program length. the
results based on the CPS 25
1
h percentile are also provided in Table 10 below.
Page 33
Report on Gainful Employment
April 2, 2010
Table 10: Percent of CIPs Impacted by Proposed Regulation
(Assuming 8 Percent Debt Ratio Using the 25th Percentile of Income)
Median Loan Based on Graduates
Total Percent Percent Percent Percent Percent Percent
Number of of of of of of
of Programs Students Females Blacks Hispanics Asians
Program Type Programs Impacted Impacted Impacted Impacted Impacted Impacted
Department of Education 25
1
h Percentile
Less than 2
Years 2,335 10.66% 18.79% 7.94% 17.06% 13.43% 18.08%
2 Year 4,493 18.52% 39.04% 27.86% 35.71% 52.76% 56.35%
4 Year 2,892 22.23% 40.93% 39.19% 34.98% 53.71% 51 .43%
CPS 25
1
h Percentile
Less than 2
Years 2,335 15.59% 25.88% 16.22% 24.84% 17.65% 31 .86%
2 Year 4,494 24.81% 50.03% 37.95% 49.88% 68.54% 74.62%
4 Year 2,853 28.74% 68.79% 68.47% 65.58% 76.91% 78.01%
Source: Data provided by CCA member institutions.
We also estimate that the impact would not be limited to a few areas of study, but would impact a wide
variety of programs. Table 11 below reports the results aggregated to general CIP categories for
categories for which we have data on at least 1 oo programs. For example, we estimate that nearly 14
percent of Health Professional and Related Clinical Sciences, including Nursing, programs and more than
46 percent of Engineering Related Technologies/Technicians programs would not currently satisfy the
proposed debt limit rule as defined by the Department. If the alternative measure of 25
1
h percentile
earnings were to be adopted, the percent of impacted programs and students substantially would be
higher.
Page 34
Report on Gainful Employment
April 2, 2010
Table 11 : Percent of CIPs Impacted by Proposed Regulation
(Assuming 8 Percent Debt Ratio Using the 25th Percentile of Income)
Median Loan Based on Graduates
Total Percent Percent Percent Percent Percent Percent
Number of of of of of of
of Programs Students Females Blacks Hispanics Asians
CIP CIP Description Programs Impacted Impacted Impacted Impacted Impacted Impacted
Department of Education Income Calculation
9 Communications, Journalism, & Related Fields 111 32.43% 70.89% 77.00% 68.40% 68.05% 86.67%
10 Graphic Communications 219 28.31% 51.27% 45.19% 53.47% 57.24% 55.81%
11 Computer & Information Sciences & Support 1,390 19.35% 32.80% 25.12% 35.15% 42.05% 41 .66%
Services
12 Personal & Culinary Services 542 27.31% 88.82% 85.02% 80.35% 88.81% 94.00%
13 Education 192 26.04% 51 .56% 55.46% 43.28% 26.43% 30.10%
15 Engineering Related Technologies/Technicians 535 46.73% 81 .70% 66.91% 80.37% 75.63% 91.35%
22 Law, Legal Services, & Legal Studies 331 9.97% 21 .91% 22.36% 8.37% 5.28% 2.56%
42 Psychology 185 33.51% 70.71% 69.90% 75.13% 54.15% 78.81%
43 Protective Services 806 9.43% 13.97% 13.79% 11.48% 26.57% 18.02%
47 Mechanic & Repair Technology 160 39.38% 80.84% 73.46% 77.70% 77.64% 87.86%
50 Visual & Performing Arts 1,342 22.35% 56.10% 57.61% 50.60% 58.39% 63.86%
51 Health Professions & Related Clinical Sciences 2,322 13.48% 15.31% 15.57% 14.54% 7.43% 17.53%
52 Business, Management, Marketing, & Related 2,356 11.50% 9.31% 9.24% 9.83% 12.59% 13.52%
Support Services
CPS Full-Time Earners
9 Communications, Journalism, & Related Fields 111 31.53% 71.84% 76.45% 69.71% 71.25% 86.67%
10 Graphic Communications 219 39.27% 84.21% 84.07% 80.68% 91 .13% 91 .73%
11 Computer & Information Sciences & Support 1,390 29.57% 62.37% 57.27% 64.20% 74.15% 75.99%
Services
12 Personal & Culinary Services 537 32.03% 92.81% 89.96% 88.14% 94.45% 96.90%
13 Education 192 32.29% 60.30% 55.85% 95.85% 93.21% 90.29%
15 Engineering Related Technologies/Technicians 535 48.79% 84.17% 68.27% 82.38% 78.42% 94.01%
22 Law, Legal Services, & Legal Studies 356 14.89% 28.66% 28.87% 22.11% 14.76% 7.69%
42 Psychology 185 36.22% 73.48% 72.81% 77.02% 56.45% 81 .78%
43 Protective Services 806 18.24% 39.90% 40.76% 38.17% 54.61% 70.43%
47 Mechanic & Repair Technology 160 41.88% 81 .05% 73.72% 77.85% 77.82% 88.04%
50 Visual & Performing Arts 1,342 26.01% 66.07% 68.47% 68.60% 66.59% 75.50%
51 Health Professions & Related Clinical Sciences 2,322 20.24% 26.89% 26.89% 24.38% 13.56% 38.07%
52 Business, Management, Marketing, & Related 2,298 18.41% 32.04% 30.91% 36.13% 38.00% 44.06%
Support Services
Source: Data provided by CCA member institutions.
Page 35
Report on Gainful Employment
April 2, 2010
6. WHAT ALTERNATIVE REGULATIONS OR POLICIES MIGHT BE
SUGGESTED TO ADDRESS THE PROBLEM AT HAND?
6.1 . FURTHER CRITICISMS OF THE PROPOSED REGULATION
Thus far this report has focused primarily on the first provision of the Department of Education's proposal,
which would limit median debt to be no more than 8 percent of the 25th percentile of earnings in specified
occupations. Programs that fail this test could retain gainful employment status by meeting alternative
tests.
Schools would be allowed to show that the graduates of the program at their school in particular earn
more than the 25th percentile upon entering the workforce. This provision would seem to address some
of the concerns raised above. However, to properly conduct a survey of graduates would be costly, and
some of these costs would be passed on to students. Furthermore, there is no guidance as to how such
a survey would need to be conducted. Would schools be required to show that the respondents were a
representative sample of all graduates? How would this be determined? How large a sample would the
estimated earnings need to be based on? Should the survey focus on the earnings in the appropriate
occupations, as specified by the Department of Education, or would earnings in other occupations count?
How would students who chose to take jobs in other higher-paying occupations be treated? How would
students who chose to take jobs in other occupations because they were unable to find work in the
specified occupation be treated? These and other questions would need to be answered. Many of these
questions highlight that the implementation of this part of the proposal would be messy at best, and quite
possibly arbitrary.
Programs that failed the 8 percent test could also retain gainful employment status by showing that they
maintained 90 percent repayment rates. As others have noted, this would not be based on default
behavior as defined in the Cohort Default Rate calculation. Students who are not current in their
payments, even though they have not yet reached the point of default, would count against a school 's
clean record. Students in deferment or forbearance would also apparently count against a school's
repayment rate. It is difficult to know how many programs would satisfy this standard. Most problematic,
the data necessary to calculate this rate is not readily available to schools. It is therefore almost
impossible to analyze whether the 90 percent standard is appropriate. Furthermore, without the data
underlying this calculation, it is not possible for schools to monitor problems, or to affect the behavior that
leads to low repayment rates. It is also not clear whether the Department of Education based the
standard on any analysis of data.
One argument described above is that it cannot be good policy to have limits on student loan payments
that are less than the benefits to earnings from schooling. The argument is the following. Standard
estimates of the return to education suggest that a student who completes a 2-year program earns 20
percent more per year, every year she works. If the debt payment limit were less than 20 percent, she
could make the loan payments out of her 20 percent schooling bonus, and still have money left over.
Page 36
Report on Gainful Employment
April 2, 2010
If the Department of Education were to consider a debt limit that is approximately equal to the return to
education, several additional factors would need to be considered. For example:
Programs are different lengths. Longer programs have larger returns. Would the department
institute different loan limitations for 1-, 2- and 4-year programs? How would the limit be set if
there were variation within a program in how long students attended (or how many credits
students earned)?
The return to education changes over time. It has risen dramatically in the past 30 years. How
would the Department of Education decide what the return is in each year?
6.2. To DIRECTLY ADDRESS THE PROBLEM, REGULATION SHOULD FOCUS ON DISCLOSURE AND
ENSURING THAT STUDENTS MAKE INFORMED DECISIONS
A comparison of expected returns to education with the costs of education is what students are doing
when they decide whether to get a higher education, and whether to take on loans to finance that
education. If the problem the policy is trying to solve is that students are not doing this well- that they
are not making informed, considered decisions based on comparisons of expected benefits and costs-
then the regulation should address this problem.
To make this case, it may be necessary first to refocus the discussion on which problem is in need of
solving. Whereas the current proposal appears to be based on the perception that students take on too
much debt, we bel ieve this is misguided. As argued above, standard economic analysis clearly indicates
that the amount of debt should be dictated by the benefits of the investment, not by the level of income. It
may not be in the students' interest to be restricted from taking on large amounts of debt; that debt may
be the key to a better future. The important thing is to make sure that students make informed and
intelligent decisions about whether loans are right for them, and whether the benefits of the schooling
they wish to finance are large enough to repay the debt they take on.
This focus on making smart informed decisions leads directly to a policy based on provision of
information, and assistance analyzing the consequences of borrowing.
One way that this problem could be addressed directly is through different forms of disclosure and
education:
Increased scrutiny could be placed on lenders to ensure that every student who takes on a loan is
made aware ofthe costs associated with the loan, the magnitude of the annual or monthly payment,
and the length of the payback period.
Disclosure could also include mandated information regarding typical earnings of workers in the
occupation for which the student is preparing. For example, lenders could be required to show
students the 25
1
h percentile or median of annual earnings in the appropriate occupation.
This could be extended further so that students would be shown a mock budget based on an
estimate of their earnings in the appropriate occupation, their loan payments, and a standardized set
of necessary expenses. This could be done either in a standardized paper form, or an online
application could be developed to allow students to enter various earnings and expense values to
see how they fit into the budget.
Page 37
Report on Gainful Employment
April 2, 2010
7. CONCLUSION
In summary, the proposed regulation is not currently formulated to address a specific problem effectively.
Secretary of Education Arne Duncan has stated publicly that he wants to ensure that the Department
understands its proposal thoroughly so that it prevents any "unintended consequences." Our analysis
suggests that the "unintended consequences" -cutting off access to hundreds of thousands of students
who want postsecondary education-will be much more substantial than the intended consequence,
which we believe to be-though we are not certain-reducing the number of students who over borrow.
To start, the Department of Education has not clearly defined what the problem is that the regulation aims
to address. As discussed above, some perceived problems the regulation may intend to address are not
problems at all but rather a reflection of the fact that for-profit postsecondary schools serve a very
different population than not-for-profit postsecondary schools. If the Department of Education wishes to
address the problem that some students take on excessive debt, the proposed regulation is not well
designed to do so. By applying a rule at the school or program level, many other students would be
negatively affected. Our analysis suggests that 33 percent of students currently in for-profit
postsecondary schooling would be denied access. Many more students would be denied access to
postsecondary schooling than would be protected from excessive borrowing.
Furthermore, it should not be assumed that public postsecondary institutions, particularly community
colleges, would absorb these students. Given the fiscal conditions of the states, it is not obvious that
community colleges will be able to increase capacity to meet the increasing demand for postsecondary
schooling.
Finally, because for-profit schools disproportionately serve racial and ethnic minority students and
students from low-income family backgrounds, the regulation would have the effect of reducing access to
higher education to groups of students that have historically had the lowest levels of access.
Page 38
Report on Gainful Employment
April 2, 2010
8. APPENDIX A
Appendix Figure 1 :
Average age at which students first enroll in postsecondary schooling, 2008
23
22.5
22
21.5
Q)
21 0)
ro
Q)
0)
20.5
ro
.....
Q)
>
20
<(
19.5
19
18.5
18
For-profit Pri vat e not-for- Privat e not-for- Public <=2 Publi c 4-year
profit <=2 profit 4-year
Age first enroll ed
Notes: Calculated from the National Postsecondary Student Aid Study, 2008.
Page 39
Report on Gainful Employment
April 2, 2010
Appendix Figure 2:
Average income of parents or independent students prior to school entry, 2008
90,000.00
80.000.00
70.000.00
<I>
60.000.00
E
0
(.)
50.000.00
c
<I>
Ol
40.000.00
ttl
...
~
<(
30.000.00
20.000.00
10.000.00
0.00
For-profit Private not-
for-profit <=2
Private not-
for-profit 4-
year
Pllblic <=2
Average income, parents or independent. 2007
Notes: Calculated from the National Postsecondary Student Aid Study, 2008.
Pllblic 4-year
Page 40
Report on Gainful Employment
April 2, 2010
Appendix Figure 3:
Percent of students from families who received AFDC prior to school entry, 2008
18 ~ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
16
14
12
c 10
Q)
0
.....
f 8
6
4
2
0
For-profit Pri vate not-for- Private not-for- Publi c <=2
profit <=2 profit 4-year
Percent receieved Food Stamps. 2007-08
Notes: Calculated from the National Postsecondary Student Aid Study, 2008.
Publi c 4-year
Page 41
Report on Gainful Employment
April 2, 2010
Appendix Figure 4:
Percent of students who were single parents prior to school entry, 2008
35 ~ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
30
25
c 20
Q)
()
....
Q)
a. 15
10
5
0
For-profit Pri vate not-for- Private not-for- Publi c <=2
profit <=2 profit 4-year
Percent single parent. 2007-08
Notes: Calculated from the National Postsecondary Student Aid Study, 2008.
Publi c 4-year
Page 42
Report on Gainful Employment
April 2, 201 0
Appendix Figure 5:
Percent of students whose parents attended at least some college, 2008
000
For profit
BOO
700
60.0
50.0 ..
E
..
.,
..
0.
400
300
200
0.0
r:>rrvale for-protot r:>uvate lo .. prolrl
2 yea.s or more 2yt&r
Not for profit
Pubhc 4-year
ooMoctorate
4-)'l)at PrrYal e not-for
doctorate prole 4
year
Notes: Calculated from the National Postsecondary Student Aid Study, 2008.
Prwa1e not-fo.-.
protrt 4-yr

Ptl'l$18 not-lor
pro(J1 4year
doctorate
Page 43
Report on Gainful Employment
April 2, 201 0
Appendix Figure 6:
Percent of students who are Black or Hispanic, 2008
60.0 ~
50.0
40.0
c
~ 30.0
"' 0.
10.0
0.0
For profit
Not for profit
Private for- Private for- PriVate not-for- Private not-for- Private not-for- Public less- Public 2-year Public 4-year Public 4-year
profitless-than profrt 2years or profrt less than proflt4Yr pro1it4-year thar)-2-year nondoctol'31e doctorate
2 - ~ a r more 4-year nondoctorale doctorate
Percent Black or Hispanic
Notes: Calculated from the National Postsecondary Student Aid Study, 2008
Page 44
Report on Gainful Employment
April 2, 201 0
Appendix Table 1
Estimated Number of Students Impacted by 2020
Median Loan Based on Graduates
25th Percentile Based on Full-Time Earners (CPS)
Number of
Total Number of African- Number of Number of
Number of Female American Hispanic Asian
Students Students Students Students Students
Year Impacted Impacted Impacted Impacted Impacted
Using the 25th percentile of annual earnings and 8% Debt-to-Earnings Ratio
2011 525,924 288,663 106,169 106,729 23,365
2012 572,206 314,065 115,511 116,121 25,421
2013 622,560 341,703 125,676 126,340 27,658
2014 677,345 371,773 136,736 137,458 30,092
2015 736,952 404,489 148,769 149,554 32,740
2016 801,803 440,084 161,860 162,715 35,621
2017 872,362 478,811 176,104 177,034 38,756
2018 949,130 520,946 191,601 192,613 42,167
2019 1,032,653 566,790 208,462 209,563 45,877
2020 1,123,527 616,667 226,807 228,004 49,914
Total Students
Impacted 7,914,462 4,343,989 1,597,695 1,606,132 351 ,613
Note: The number of impacted students assumes that the CCA data is representative of all for-profit
schools, that for-profit schools wi ll continue to grow at 8.8% per year (the growth rate over the last five
years), and the relative student composition does not change during this period.
Page 45
Report on Gainful Employment
April 2, 201 0
Appendix Table 2
Estimated Annual Number of Students Impacted by State
Median Loan Based on Graduates
25th Percentile Based on Department of Education Calculation
Number of
Total Number of African- Number of Number of
Number of Female American Hispanic Asian
Students Students Students Students Students
State lmeacted lmeacted lmeacted lmeacted lmeacted
Using the 25th percentile of annual earnings and 8% Debt-to-Earnings Ratio
AL 1,134 630 470 13 11
AR 1,204 718 297 22 19
AZ 33,998 15,928 3,377 3,160 924
CA 44,910 22,958 3,935 16,101 6,713
co 7,593 3,363 981 1,004 248
CT 5,648 2,169 870 1,015 120
DC 4,681 2,266 2,150 368 172
FL 25,202 12,307 5,601 7,738 487
GA 10,324 5,727 4,979 491 242
lA 6,684 3,705 292 117 50
IL 18,988 9,294 4,290 2,372 677
IN 4,978 2,732 1,115 195 37
KS 1,681 893 270 108 58
KY 3,292 1,865 574 41 41
LA 3,885 2,242 1,581 57 42
MA 5,765 2,893 520 742 240
MD 4,297 1,960 2,155 188 94
Ml 8,949 5,099 3,043 232 109
MN 9,406 5,038 1,309 325 365
MO 5,510 2,953 1,349 106 80
MS 1,111 667 576 7 23
NC 2,307 1,116 863 69 44
NH 1,284 760 17 55 18
NJ 9,118 4,366 1,969 2,294 430
NM 1,612 926 102 650 27
NV 2,722 1,344 411 472 265
NY 18,845 8,797 4,928 4,439 1,129
OH 11,686 6,147 3,401 291 129
OK 2,456 1,273 417 163 49
OR 2,901 1,579 107 183 148
PA 16,909 7,653 3,334 896 299
Rl 1,839 936 189 324 38
sc 1,523 878 655 46 19
TN 5,682 2,911 1,848 124 67
TX 29,176 14,832 5,846 12,457 718
UT 2,675 1,380 33 231 83
VA 7,032 3,720 2,854 328 250
WA 3,964 2,107 269 261 490
WI 2,004 1,116 707 100 46
wv 3,226 941 321 215 80
Total Students Impacted 361 ,172 179,149 68,348 78,545 15,875
Note: The number of impacted students assumes that the CCA data is representative of all for-profit schools. States
with less than 1,000 students who are impacted and Puerto Rico are not reported above.
Page 46
Report on Gainful Employment
April 2, 201 0
Appendix Table 3
Estimated Annual Number of Students Impacted by State
Median Loan Based on Graduates
25th Percentile Based on Full-Time Earners (CPS)
Number of
Total Number of African- Number of Number of
Number of Female American Hispanic Asian
Students Students Students Students Students
State lmeacted lmeacted lmeacted lmeacted lmeacted
Using the 25th percentile of annual earnings and 8% Debt-to-Earnings Ratio
AL 1,652 1,016 730 18 17
AR 1,753 1,157 461 29 28
AZ 49,506 25,665 5,246 4,294 1,360
CA 65,396 36,992 6,113 21,879 9,881
co 11 ,056 5,419 1,525 1,364 365
CT 8,224 3,495 1,351 1,379 177
DC 6,816 3,650 3,340 500 253
FL 36,698 19,830 8,700 10,515 717
GA 15,034 9,228 7,733 667 355
HI 1,029 548 23 27 1,002
lA 9,733 5,970 454 158 74
IL 27,650 14,976 6,664 3,223 996
IN 7,249 4,402 1,732 265 54
KS 2,449 1,438 420 146 86
KY 4,794 3,005 892 56 61
LA 5,657 3,613 2,455 78 62
MA 8,395 4,662 808 1,008 353
MD 6,256 3,159 3,348 256 139
ME 1,279 924 26 13 19
Ml 13,032 8,217 4,727 315 161
MN 13,697 8,118 2,033 441 538
MO 8,024 4,759 2,095 144 117
MS 1,618 1,075 895 10 34
NC 3,359 1,798 1,341 94 64
NE 1,181 739 149 41 15
NH 1,870 1,224 26 74 26
NJ 13,277 7,036 3,058 3,117 633
NM 2,347 1,492 158 884 39
NV 3,964 2,166 638 641 391
NY 27,441 14,174 7,654 6,031 1,662
OH 17,017 9,905 5,283 395 189
OK 3,576 2,051 647 222 72
OR 4,225 2,544 166 249 218
PA 24,622 12,331 5,179 1,217 441
Rl 2,679 1,507 293 440 55
sc 2,217 1,415 1,018 62 28
TN 8,274 4,691 2,871 168 99
TX 42,485 23,899 9,081 16,926 1,057
UT 3,896 2,223 52 314 123
VA 10,239 5,994 4,434 446 369
WA 5,773 3,395 418 354 722
WI 2,918 1,798 1,098 135 68
vw 4,697 1,517 499 292 118
Total Students Impacted 525,924 288,663 106,169 106,729 23,365
Note: The number of impacted students assumes that the CCA data is representative of all for-profit schools. States
with less than 1,000 students who are impacted and Puerto Rico are not reported above.
Page 47
(b)(S)
-----Original Message-----
From: Yuan Georgia
To: Jenkins. HarQld
Finley, Steve
Sann, Ronald
CC: Wie,gner Ashley
Rose, Charlie
Miceli Julie
Date: 2/25/2010 1:08:28 PM
Subject: FW: request for meeting
From: Broff, Nancy [ mailto:BroflN@DicksteinShapiro. COM]
Sent Thursday, February 25, 201010:25 AM
To: Rose, Charlie
Cc: Yuan, Georgia; Miceli, Julie; Wiegner, Ashley
Subject RE: request for meeting
H.i Charlie-
I am writing to renew my request for a meeting. The meeting would be
with the CEOs of three higher education companies:
Daniel Hamburger ofDe V ry
Clark Elwood of ITT Educational Services, Inc.
Andrew Clark ofBridgepoint Education
They would like to meet on March 11 if you have an availability that
day. The agenda would be to discuss some of the issues from the
recently completed negotiated rulemaking.
Thanks so much,
Nancy
IRS Circular 230 Disclosure: To ensure compliance with requirements imposed by the IRS, we inform you that any tax
advice contained in this communication (including any attachments) was not intended or written to be used, and cannot
be used, for the purpose of (i) avoiding penalties under the Internal Revenue Code or (ii) promoting, marketing or
recommending to another party any transaction or matter addressed herein.
From: Rose, Charlie [mailto:Charlie.Rose@ed.gov]
Sent Monday, February 15, 2010 10:52 PM
To: Broff, Nancy
Cc: Yuan, Georgia; Mceli, Julie; Wiegner, Ashley
Subject: RE: request for meeting
Nancy: Thanks. Let me check and I will follow-up with you regarding
the meeting. By the way, even by Chicago standards, the DC snow is
tough to deal with. Charlie
-----Original Message-----
From: Broff, Nancy [mailto:BroffN@DicksteinShapiro.COM]
Sent: Tuesday, February 09, 2010 11:54 AM
To: Rose, Charlie
Subject: request for meeting
Hello Charlie-
I hope you are doing well in all this snow. I guess you Chicagoans have
an advantage in the area of coping with real winter.
I am writing to request a meeting with you for the CEO of my client,
Bridgepoint Education, Inc. Andrew Clark will be in town on the morning
ofFebruary 25th and would very much appreciate an opportunity to
discuss some concerns arising out of the recent negotiated rulemaking
sessions. Bridgepoint owns two institutions: Ashford University and
University of the Rockies. I will be happy to provide some background
information on both universities in advance of the meeting. Those of us
in the higher education bar very much appreciated your openness to
hearing from members of the community, and this is an important topic on
which I think it is important for you to hear directly from a school
official.
In the interest of full disclosure, I want to mention that Ashford is in
the midst of an OIG audit (not an investigation), but we have not yet
received a draft report and would not plan to discuss that at this
meeting.
I hope you can make time in your schedule for this meeting.
Best regards,
Nancy
IRS Circular 230 Disclosure: To ensure compliance with requirements
imposed by the IRS, we inform you that any tax advice contained in this
communication (including any attachments) was not intended or written to
be used, and cannot be used, for the purpose of (i) avoiding penalties
under the Internal Revenue Code or (ii) promoting, marketing or
recommending to another party any transaction or matter addressed
herein.
This e-mail message and any attached files are confidential and are
intended solely for the use of the addressee(s) named above. This
communication may contain material protected by attorney-client, work
product, or other privileges. lfyou are not the intended recipient or
person responsible for delivering this confidential communication to the
intended recipient, you have received this communication in error, and
any review, use, dissemination, forwarding, printing, copying, or other
distribution of this e-mail message and any attached files is strictly
prohibited. Dickstein Shapiro reserves the right to monitor any
communication that is created, received, or sent on its network. If you
have received this confidential communication in error, please notify
the sender immediately by reply e-mail message and permanently delete
the original message.
To reply to our email administrator directly, send an email to
postmaster@di cksteinshapi ro.com
Dickstein Shapiro LLP
http:/ /www.Di cksteinShapiro. com
This e-mail message and any attached files are confidential and are intended solely for the use of the addressee(s)
named above. This communication may contain material protected by attorney-client, work product, or other
privileges. If you are not the intended recipient or person responsible for delivering this confidential
communication to the intended recipient, you have received this communication in error, and any review, use,
dissemination, forwarding, printing, copying, or other distribution of this e-mail message and any attached files
is strictly prohibited. Dickstein Shapiro reserves the right to monitor any communication that is created,
received, or sent on its network. If you have received this confidential communication in error, please notify the
sender immediately by reply e-mail message and permanently delete the original message.
To reply to our email administrator directly, send an email to postmaster@dicksteinshapiro.com
Dickstein Shapiro LLP
http:/ /www.DicksteinShapiro.com
(b)(5)
Georgia
From: Fine, Stephanie
Sent: Friday, March 26, 2010 7:14PM
To: Yuan, Georgia
Subject: Response needed
(b)(5)
Stephanie
From: Yuan, Georgia
To: Phillips, Nia
CC: Jenkins. Harold
Rose. Charlie
Date: 3/27/2010 10:08:54 AM
Subject: FW: Response needed
Attachments: attach 0
attach 1
attach 2
From: Andy Rosen <arosen@kaplan.edu>
To: Private- Miller
9
Anthony
CC: Fine Stephanie
Rebecca Campoverde
Date: 3/25/2010 11:29:52 PM
Subject: Thanks and next steps
Tony: Thank you for making time to meet with me on Tuesday regarding gainful employment and other issues of
importance to Kaplan' s students and postsecondary schools. I am encouraged that you are engaged in the important
policy discussions taking place within the Department and the Administration. As I indicated, your proposal to create a
small sector group to work with you and the Department on gainful employment makes a great deal of sense. I am glad
to help organize such a group and suggest that our staff talk regarding the number of participants, time fiame, and
specific dates on which you would be available for such a meeting. I look forward to hearing back from you regarding
proposed next steps.
Again, many thanks for your interest and attention.
Best regards,
Andy
Andrew S. Rosen
Chairman and CEO
Kaplan, Inc.
Tony,
From: McKernan, John <jmckeman@edmc.edu>
To: Private -Miller, Anthony
CC: Guida, Anthony
Rothkopf, Arthur J
Date: 3/25/2010 9:02:34 PM
Subject: Thank you
Thank you for meeting with Arthur and the gang yesterday! I hope it was clear from our meeting that we atEDMC are
supportive of the Department's goals of giving more higher education opportunities for more of our citizens.
We especially appreciated your openness and candor, and we look forward to being constructive participants in our future
discussions. I, in particular, believe there is a need for appropriate regulation of our industry given the amount of tax dollars
supporting students in higher education.
We want to make sure, however, that the proposed regulations address identified problem areas in a way that is both
consistent with the authority delegated to the department and narrow enough to avoid unintended consequences which coul
reduce student opportunities and access to legitimate and beneficial programs.
Our SVP for Regulatory Affairs, Tony Guida, and Mark Pelesh from Corinthian will be following up with your office as we
discussed. I will be monitoring the various proposals being considered and am willing to be as involved as you would like.
Please feel free to include me in the discussions to the extent you feel I can be helpful.
Again, thanks for your involvement. We hope this wiJl be the beginning of a very important dialogue about how best to
regulate an industry that needs to be a significant provider of post secondary education services if, as the President envision
millions of additional students in America are to have the educational opportunities they need and deserve.
With best wishes.
Jock McKernan
CONFIDENTIALITY NOTICE: This email and any files transmitted with it are confidential and intended solely for the ust
of the individual or entity to which they are addressed. If you are not the intended recipient, you may not review, copy or
distribute this message. If you have received this email in error, please notify the sender immediately and delete the original
message. Neither the sender nor the company for which he or she works accepts any liability for any damage caused by an
virus transmitted by this email.
From: Pelesh, Mark <M Pel esh@cci edu>
To: Private- 1\.ililler. Anthony
CC: jmckeman@ecimc edu
1
Massimino. Jack
1
Guida. Anthoni
Rothkopf Arthur J
Date: 3/25/2010 7:26:04 PM
Subject: Follow-up Meeting-- Gainful Employment
Thank you again for meeting with us, our colleagues from EDMC, and Arthur Rothkopf yesterday. We are anxious to
follow-up on our constructive discussion regarding the gainful employment proposal under consideration at the
Department. We very much want to take you up on your suggestion for a follow-up meeting in the near future to discuss
other approaches to address the concerns that may have led to the development of the gainful employment proposal, as
well as a more comprehensive long range approach to accountability.
As we discussed, it would also be very helpful if we could get a full statement of the problems the Department is seeking
to solve. This will allow us to develop better approaches to address those problems than the gainful employment
proposal, which we continue to believe is deeply flawed, and to make the meeting as productive as possible.
We have reviewed our calendars and suggest the following dates for the follow-up meeting: (April6, 12, 13, 14, 15, or
16). I will ask Stephanie Johnson at the Chamber to be in touch with your scheduler to make the arrangements. We look
forward to continuing our dialogue.
MarkPelesh
Executive Vice President, Corinthian Colleges
Please print
From: Fine, Stephanie
Sent: Friday, March 26, 2010 7:14PM
To: Yuan, Georgia
Subject: Response needed
(b)(5)
From: Yuan, Georgia
To: Canada, June
CC:
Date: 3/29/2010 11:23:28 AM
Subject: FW: Response needed
Attachments: attach 0
attach 1
attach 2
From: Andy Rosen <arosen@kaplan edu>
To: Private- Miller, Anthony
CC: Fine, Stephanie
Rebecca Campoverde
Date: 3/25/2010 11 :29:52PM
Subject: Thanks and next steps
Tony: Thank you for making time to meet with me on Tuesday regarding gainful employment and other issues of
importance to Kaplan' s students and postsecondary schools. I am encouraged that you are engaged in the important
policy discussions taking place within the Department and the Administration. As I indicated, your proposal to create a
small sector group to work with you and the Department on gainful employment makes a great deal of sense. I am glad
to help organize such a group and suggest that our staff talk regarding the number of participants, time frame, and
specific dates on which you would be available for such a meeting. I look forward to heating back from you regarding
proposed next steps.
Again, many thanks for your interest and attention.
Best regards,
Andy
Andrew S. Rosen
Chairman and CEO
Kaplan, Inc.
Tony,
From: McKernan, John <jmckeman@edmc edu>
To: Private- Miller Anthony
CC: Guida Anthony
Rothkopf, Arthur J
Date: 3/25/2010 9:02:34 PM
Subject: Thank you
Thank you for meeting with Arthur and the gang yesterday! 1 hope it was clear from our meeting that we at EDMC are
supportive of the Department's goals of giving more higher education opportunities for more of our citizens.
We especially appreciated your openness and candor, and we look forward to being constructive participants in our future
discussions. I, in particular, believe there is a need for appropriate regulation of our industry given the amount of tax dollars
supporting students in higher education.
We want to make sure, however, that the proposed regulations address identified problem areas in a way that is both
consistent with the authority delegated to the department and narrow enough to avoid unintended consequences which coul
reduce student opportunities and access to legitimate and beneficial programs.
Our SVP for Regulatory Affairs, Tony Guida, and Mark Pelesh from Corinthian will be following up with your office as we
discussed. I will be monitoring the various proposals being considered and am wiJling to be as involved as you would like.
Please feel free to include me in the discussions to the extent you feel I can be helpful .
Again, thanks for your involvement. We hope this will be the beginning of a very important dialogue about how best to
regulate an industry that needs to be a significant provider of post secondary education services if, as the President envision
millions of additional students in America are to have the educational opportunities they need and deserve.
With best wishes.
Jock McKernan
CONFIDENTIALITY NOTICE: This email and any files transmitted with it are confidential and intended solely for the use
of the individual or entity to which they are addressed. If you are not the intended recipient, you may not review, copy or
distribute this message. If you have received this email in error, please notify the sender immediately and delete the original
message. Neither the sender nor the company for which he or she works accepts any liability for any damage caused by an
virus transmitted by this email.
From: Pelesh, Mark <M Pel esh@cci edu>
To: Private- 1\.ililler. Anthony
CC: jmckeman@ecimc edu
1
Massimino. Jack
1
Guida. Anthoni
Rothkopf Arthur J
Date: 3/25/2010 7:26:04 PM
Subject: Follow-up Meeting-- Gainful Employment
Thank you again for meeting with us, our colleagues from EDMC, and Arthur Rothkopf yesterday. We are anxious to
follow-up on our constructive discussion regarding the gainful employment proposal under consideration at the
Department. We very much want to take you up on your suggestion for a follow-up meeting in the near future to discuss
other approaches to address the concerns that may have led to the development of the gainful employment proposal, as
well as a more comprehensive long range approach to accountability.
As we discussed, it would also be very helpful if we could get a full statement of the problems the Department is seeking
to solve. This will allow us to develop better approaches to address those problems than the gainful employment
proposal, which we continue to believe is deeply flawed, and to make the meeting as productive as possible.
We have reviewed our calendars and suggest the following dates for the follow-up meeting: (April6, 12, 13, 14, 15, or
16). I will ask Stephanie Johnson at the Chamber to be in touch with your scheduler to make the arrangements. We look
forward to continuing our dialogue.
MarkPelesh
Executive Vice President, Corinthian Colleges
From: Minor Robin
To: Finley, Steve
CC:
Date: 4/30/2010 2:12:40 PM
Subject: FW: Shireman Transcript at NASASPS
You may be interested. This was Bob's testimony at NASASPS this week.
-----Original Message-----
From: Minor, Robin
Sent: Friday, April30, 2010 2:05PM
To: Shireman, Bob; Manheimer, Ann
Subject: FW: Shireman Transcript atNASASPS
I Nonresponsive
-----Original Message-----
From: LoBosco, RaJ ph
Sent: Friday, April 30, 2010 11:43 AM
To: Leon, Geneva; Frola, Michael; Henderson, Linda; Parrott, Douglas; Dragoo, Janet
Cc: White, Carolyn; Minor, Robin
Subject: Shireman Transcript at NASASPS
!Nonresponsive
RaJ ph A LoBosco, Area Director
Federal Student Aid
United States Department of Education
8930 Ward Parkway Suite 2028
Kansas City, MO. 64114-3392
816-268-0440 Direct
816-823-1402 Fax
Ralph.LoBosco@ed.gov
Robert Shireman
Speech to NASASPS
ApriJ 28, 2010
Transcript provided by the Career Education Review
Michael J. Cooney, Editor
mcooney@workforce-com.com
. .. And I had forgotten to put the FAFSA simplification applause line into my actual
remarks so I have applaud when he said it so it never fails and of course- it's not often-
sometimes it's the thing I get questions about but more often its about things like two Pell
awards, two programs in an award year, things more detailed like that. But thank you so
much for that introduction.
Two and a half or three years ago, we started to see a serious economic slide downward
in this country, credit markets had seized up, the sub prime mortgage issue was a major
cause of that and we stmied seeing ppllosing their jobs. We saw ppl in their jobs feeling
much more insecure, much less secure about their ability to invest in higher education,
their ability to buy a home w the collapse of the credit markets and the way to solve that
- long term- is to invest in improving our nations economy, to invest in the kind of
innovation that comes from education, the productivity increases that come from job
training.
Jn order to follow up on that President Obama laid out a bold goal for he country, he said
that by 2002 we want to regain our place as the number one country in the world in terms
of adults with post sec credentials, college degrees, certificates and other job training
programs. In the recovery legislation, now about a yr and a half ago, that included an
expansion of the tax credits that [x] hoping to create in the 90s, an expansion of that tax
credit to $2500, making it for 4 yrs and actually covering more of the types of expenses
that students and families have for higher education. Increases in Pell grants- the usual
approach and what you have seen in your own states, are in an econ downtum, more ppl
are poor, more ppl want to go to school , but instead of following up on that need by
putting more money onto the grant and scholarship programs actually less money goes
into the grant and scholarship programs be of the state budgets.
Fed govt took the opposite approach really, what needs to be countercyclical spending
that helps- like unemployment insurance- spending that needs to follow up on and help
to address the new gaps that families are seeing - so the tax credits were part of that, the
increases in Pell grants are not only meeting the new demand for Pell grant dollars but
actually increasing he size of the Pell grants and proving those increases into the future
with the follow up legislation passed a few weeks ago. Also restoring some certainty to
the student loan program, and making sure that no one has any reason to doubt whether
they will be able to get the federal student loans that they need, again with the refmms
that were implemented a few weeks ago.
I mentioned that when people are losing their jobs, when ppl become insecure in their
jobs, they look for higher education, they look to find what kind of job training can I get,
what kind of skills can I add to my repertoire, what are the skills that I have, how can I
make them better so that ill keep my job, so that ifl lose my job I'll have options. And at
the same t ime, while we saw this increase in demand, which is helpful and useful given
what the President had to say about the need to train our population, we saw state tax
revenue declining in all but a few states, we saw resulting cuts in the budgets of state
colleges and universities and community colleges, resulting in a combination of very
large increases in tuition in some cases and reduced enrollments, fewer seats. So
increased demand - ppl wanting more education and training, and public institution either
had fewer seats and charging more tuition or might not declare going to be enrolling
fewer ppl but their course offerings are cut, fewer kinds of course offerings, so the result
is they are not able to demand for higher education.
Tuition-driven institutions didn't react that way because they're tuition-driven institutions,
and the non-profit institutions have done pretty well despite significant declines in their
endowments because there continued to be significant demand for higher education. The
public., the non profit private colleges did well and in particular the for profit institutions
have come in with investors making sure that there was capacity to be able to serve
additional students, and they knew that those students would come with those federal
dollars, Pell grants, student loans, tax credits and that hat would help them to not only be
consumers who want higher ed but consumers who can pay for that higher education with
that federal support so the for-profit industry, more than any other in this economic
difficult times, has responded.
I want to give you some specific numbers, we now post now on one of the [X] websites,
the quarterly numbers ofPell grants by different kinds of schools s I looked at what the
first 3 quarters- the total of the first 3 quarters of this award year compared to the last
award year for some of the schools that I knew would be here today. So for example,
Corinthian Colleges- 38% increase for first 3 quarters this year compared to last year for
a total of $800M
DeVry- a couple people here from DeVry?- 42% increase up to $1.78
ITT- you guys here? A 44% increase up to $623M
Strayer- still here? Is that you? Well this one- 95% increase, may be something about
the quarters, but up to $414M
APE!- Wally here? And Russell? 94% increase up to $44M
Kaplan- they here? So this total is actually all the Washington Post owned entities, 33%
increase up to $909M, and again this is the first 3 quarters of the year so the totals for the
year are obviously more than that
Career Education Corporation- 29% increase up to $1B this first three quarters
EDMC- several folks here; a 16% increase, $1.1B
Capella- over there? 40% increase to $378M
And I think I've just got a couple of others:
Grand Canyon- 55% increase to $260M
And University of phoenix- you there?- 9% increase but obviously that's on a larger
base. So probably that increase is as much as a lot of others' total dollars, and that
increase is $2.7b total
And Bridgepoint- you guys here? - 6 l% increase, $393M
I think those were all that I had numbers for, obviously I know that there's a few others
here as well.
So I wanted to begin just by thanking the for-profit industry for responding to the critical
demands from ppl out there who need higher education. I'd like everybody to give them a
hand. Now, others of us in the room have the responsibility for making sure those federal
funds I just listed, for education and training, that it's all totally above board. That those
significant increases in fed spending for higher education- loans, grants- are serving
students and tax payers as well as they possibly can. and that is what the Triad is about-
and I know I can say triad in front of this audience because I heard somebody say it
earlier. I want to talk for a second about some things going on in Washington right now,
and I don't mean negotiated rulemaking- I will get to that in a few minutes, but there is a
wall street reform debate going on right now in Washington. What happened in that
credit crisis a couple of years ago had something to do with credit rating agencies-
agencies like S&P, Fitch, other agencies that were responsible for rating instruments-
:financial instruments, looking at what is the quality of these things that have names that
cause people's eyes to roll over- things like collateralized debt obligations, and other
kinds of securitizations- so what is the quality of the loans, mortgages, are they going to
be repaid, how likely are theses loans to be repaid so that an investor purchasing this,
how confident can they be that when they purchase, when they invest in this particular
instrument, that they will get the money back that they are expecting.
The business model for these rating agencies has come under fire in these meetings in
Washington, part of this has to do with the business model of the rating agencies, on the
one hand, their responsibility, their job, the core of their business was to make sure they
did a good job providing an honest rating tor the instrument that they were analyzing. On
the other hand, they relied on the income from the companies who asked them to rate the
instrument, and ill read to you from- a NYTimes- some of the emails that have been
coming out recently.
In 2004, well before wall street's bets on subprime mortgages became widely known,
employees at Standard & Poor's credit rating agency were feeling pressure to expand the
business. One employee warned in an internal email that the company would lose
business if it failed to give high enough ratings to collateralized debt obligations, the
investments that later emerged at the heart of the financial crisis.
Quote, " we are meeting with your group this week to discuss adjusting criteria for rating
CDOs of Real estate assets be of the ongoing threat of losing deals. Lose the CDO and
lose the base business. A self reinforcing loop"
In other words, if we don't loosen up, we don't loosen up in our assessment of these
instruments, nobody is going to come have their instruments assessed by us anymore.
And this created a conflict which led to instruments that should have been questioned not
being questioned, and [leading] over to the financial crisis that we have been suffering
from for the past couple of years.
The other issue besides the business model was the complexity and fast growth of diff
kinds of instruments and ill read from another of the recent articles. "Email. documents
and other messages suggested that executives and analysts at ratings agencies embraced
new business from Wall Street even though they recognized that they couldn't properly
analyze all of the banks' products. And one of the other quotes ends with, "we were so
overwhelmed."
So I want to actually ask, on that issue, the complexity and growth, and I know we're
feeling this with publicly traded corporations and purchases going this way and that way,
and we're trying to figure out what's going on . Are there regulators in the room who feel
like you DO have the analytical firepower you need to assess what is going on with the
entities you regulate in higher education? Those who do feel you have the firepower you
need? I don't think we feel we have the firepower we need.
So the reform back on the financial instrument side of the equation, what they're really
talking about now in Washington on financial reform, one analyst- an academic looking
at what's going on, said it only tinkers with the workings of the ratings agency, it doesn't
end the inherent conflicts of interest, those cont1icts of interest where the ppl who do the
rating are paid for by those who do the ratings. This whole situation w credit agencies,
credit rating agencies, is, as I see it, very similar to the way accrediting agencies work in
this country. The same kind of inherent conflict of interest. Albeit accrediting agencies
are nonprofit and on top of that, what would this crisis look like ifthe banks had actually
been the ones running the credit agencies and were doing a peer review kind of model,
which is the model we have in accreditation, where it is the regulated who are really
looking at each other rather than an outside entity.
So to borrow from Winston Churchill, accreditation, as a part of that triad, in terms of a
way of assessing quality in higher education, is the worst form of accountability except
for all of the others. What Winston Churchill actually said was, democracy is the worst
form of government, except for all the others. So I am bringing up this issue of
accreditation not to say that we should back away from it or change it, 1 actually don't
have a better system for us for assessing quality in higher education. But it is problematic
and we need to remember that as the other two pieces of the triad, as we figure out how
we can do the best job possible in our responsibilities. Federal and state governments
cannot rely on accreditation to insure that consumers and taxpayers are protected t the full
extent that they need to be- all three legs of that three legged stool need to be working
and working well.
There are a number ofthings that we're doing, you've heard about some ofthem-
elevating, monitoring and enforcement, we're working with the inspector general at the
department of education, taking a much closer look at data than ever before to help guide
our selection for program reviews and investigations when necessary by the inspector
general, working with he federal trade commission to join their consumer complaint
system so complaints they get and other agencies that are on their consumer sentinel,
working on the issue of how we can look more at issues of misrepresentation as we do
program reviews and other kinds of monitoring.
A second area besides the monitoring and enforcement is improving consumer
information. We have put graduation rates, retention rates and transfer rates right on the
F AFSA form when students are choosing colleges, the rates are right there as a reminder
to students that they should do some good shopping, look at various kinds of data that
might help them to compare schools. We're also providing them with a more detailed
financial aid estimate in terms of the financial aid that they can get and this is partly to
make sure that people know they can get that aid wherever they go. Sometimes students
think, oh I can get that $12K because the school costs $12K, and I would only get $3k at
a community college that only cost $3K, not realizing hat in fact if they wanted to get
more than what tuition costs at that community college so that they can focus on their
studies instead of working excessive hours, that that is something that they can have
available to them.
And starting this summer, as a result of a regulatory process that has already completed,
schools will have to begin providing placement information and where the have
placement rates, they actually will need to make students aware on their websites of
placement rates they have for programs that they are offering.
Uh coordination and sharing, I head some of the discussion in prior sessions and I look
forward to this af-temoons discussion. Within the federal government, we are working
with the Fed trade commission, the veterans administration around the GI bill , the SEC
because of the involvement of publicly traded institutions, states we have encouraged
involvement in this group and are looking for other ways that we can help. Happy to
discuss that in q&a here as well as this afternoon because we really need to become good
partners if we' re going to do best by taxpayers and students.
And accreditors, there are some new requirements, we're working on sharing some draft
guidance related to all of the requirements for accreditors and again building that triad
and all working together. In fact, I was actually on the internet looking for a three legged
stool to see if [ could bring one for this, and [ noticed one of the three legged stools had a,
not only the three legs, but it had this connecting piece of wood that held the three legs
together, and I thought, well that would be the perfect prop, because that would
demonstrate it's a strong stool if that connecting- that connector is there, making it as
strong as possible.
Also, may of you have heard, reviewing the rules and regulations and where appropriate
revising, in the process of revising those rules. Let me take a I ittle bit of time to tell you
about some of those. We started about a year ago, doing public hearings where we
basically said, we want to know whether we need to in1prove program integrity, are thee
things we need to be doing? Here's a list of some areas, misrepresentation- definition of
credit hour, state authorization, other kinds of things, and we saw input. We did 3 public
hearings, ppl were able to submit items over the internet, through email, and we got a lot
of input about great schools out there, students who were having a good experience,
people who attended the schools, got a job, had a great experience. We also heard from
forn1er students who felt that they were misled, legal aid attorneys who had clients whose
stories were cause for concern.
That was followed up by- we asked for nominations for ppl to serve on committees- the
way this whole process works is that we do our best to work through possible rule
changes with a committee of stakeholders, recommended, nominated by interested parties,
states, various institutions, student organizations, legal aid.tr 3 weeklong sessions,
December, January and February went through each of 14 issues talking about, changes
that might make sense.
One of them, misrepresentation, clarification really against misrepresentation by schools.
High school diploma- one of those things that you take to somebody and think, how
hard can it be to know if a high school dimple is valid?? As you know, not that easy and
many of you are at the state level so you know that the state isn't necessarily declaring
who is good or bad. And the issue of the federal government declaring what is a valid HS
education, for example, gets into areas where the def government isn't supposed to be
declaring such things, so I think a more complicated issue than I think a lot of ppl
expected. We are making - at least in NegReg session - reached some tentative
agreement around the definition.
I would say the most significant thing we are doing is looking at- and I think this is now
a likelihood, when people apply on the F AFSA and it asks for a HS diploma, a list will
actually pop up and they can enter what that high school is, the name of that high school
based on some feral lists we have. It wont necessarily mean that it will be a valid high
school, but it does give us and you the ability to, if for example a suspected diploma mill,
we would be able to see who are, where are the students going who are using this
particular high school as the place they say they get their diploma from. And if we find
that its some particular colleges, that means that it might b encouraging ppl to go and use
and diploma mill. So it will be a useful took for us and you as well. , and that's the most
important change we'll be making there.
Incentive compensation was a major issue, the issue of paid recruiters. A number of
years ago, a number of safe harbors were created and there was a lot of indication that
they were wider loopholes than are appropriate given the wording of the actual law that
prohibits payment of actual compensation based on enrollment. So that's another one that
we are working on.
State authorization, I heard California mentioned and it was a surprise to me when I came
to Washington and asked about California to discover that a legal interpretation of the
Dept of Ed, well if the school is not not authorized, then it is authorized. So this a raised
question that came up in NegReg about what is at least some minimum standard about
what kind of authorization should count in terms of the state role in that Triad.
Satisfactory academic progress is another area taking attendance. What] used to call
R2D2, return to tile 4. and I'm not mentioning all of the issues, but the final one I will
talk some about is Gainful Employment, and this is the one that's been in the news a lot.
It seems that every time I speak somewhere, something thinks I said something new and
calls a stock analyst who then reports it causing the stocks to go up or down or whatever,
and I assure you I am not going to say anything new. If you are a stock analyst or you
know a stock analyst, the answer when they ask you, What did Shireman say? You say
nothing new.
So the statute, the federal law requires that in order for some programs to be eligible for
federal financial aid, they have to lead to gainful employment in a recognized occupation.
This applies to non degree programs at any type of school and it applies to most programs
at for-profit schools, really all except some BA, liberal arts programs through an
exception, a recently enacted exception, that actually begins this July 1
51
but for the most
part, a for-profit institution, in order to be eligible for federal financial aid, has to show
that the program leads to gainful employment or prepares the student for gainful
employment in a recognized occupation.
So a year ago, we began asking the question, what is the definition, what should db e the
definition of gainful employment in a recognized occupation. We had hoped that perhaps
some schools would come forward and say, well when we start a program, here's how we
determine whether or not it complies .. we didn't get that kind of information.
We brought it up in NegReg and made some suggestions for discussion. We suggested,
maybe there should be some relationship to the debt levels that students are taking on and
the expected earnings that they may have from the occupations that you have identified
that you are preparing people for. We also suggested that perhaps a loan repayment rate
approach could be devised where we would be able to see that federal loans are actually
being repaid at a rate that makes sense if ppl were actually gainfully employed.
We looked at the provision and current regulation that currently applies to very short
programs, the 70/70 mle. So 70% completion rate, 70% placement rate, and asked should
something like that be part of the definition of gainful employment? And then for new
programs, we suggested maybe there should be something from an employer, who
employs ppl in the occupations that the program is preparing people for, that at least
asserts that yes, the cuiTiculum, the program that ive seen at this school is designed in a
way where it would prepare people for the jobs that I have in my particular business, so
we suggested that for new programs.
Now in evety other issue in negreg, we got pretty good discussion at the table, sometimes
we actually got consensus from the group on what we should actually, how regulation
should actually be worked out. But for some reason on the gainful employment issue, we
didn't get the kind of discussion that would at least help to guide in a very constructive
way the direction, and to know well, this would be okay with certain kinds of schools but
wouldn't be okay with other kinds of schools.
Instead the reaction from, in particular those who were representing the for-profit
colleges, was you cant do this, you cant define this term, why are you doing this, and that
continued even after the NegReg sessions. We continued to meet, we have gotten
improved input, improved feedback. And where things stand now with whole regulatory
packed, so everything I've just discussed now including the gainful employment, is that
in the next few weeks there will be proposed rule published in the federal register. There
will be a comment period after that proposed rule is published. That will be the
appropriate time to suggest changes or express support for provisions, suggest
alternatives, and then a final mle, our goal would be to publish a final rule by November
I
51
For rules to take effect in general. next year from this July, they need to be published
by November 1
51
So that's where we will be, that's the tin1eline for the rule going
forward.
I wanted to conclude my remarks before going to some Q&A and some discussion with a
piece that Thomas Frank wrote in the WSJ. The title ofthe article is, "Obama and the
Regulat01y Capture," and it is again, back about the financial regulation.
"It was not merely stmctural problems that led certain regulators to nap through the crisis.
The people who filled regulatory jobs in the past administration were asleep at the switch
because they were supposed to be. It was as though they had been hired for their
extraordinary powers of drowsiness.
The reason for that is simple: There are powerful institutions that don't like being
regulated. Regulation sometimes cuts into their profits and interferes with their business.
So they have used the political process to sabotage, redirect, defund, undo or hijack the
regulatory state since the regulatory state was first invented."
So, he follows that up with one more line here, "And it created a situation where banking
regulators posed for pictures with banking lobbyists while putting a chainsaw to a pile of
regulations. Smiles all around. Let the fellows at IndyMac do whatever they want."
So my closing words, is, we should take the photos, we should smile, but lets not shirk
our responsibility for regulating the industry. The schools will make plenty of money and
students and taxpayers will be better off if we do our jobs as best as we can. Thank you
very much.
Host - Some time for questions and answers, and I know gainful employment will come
up, but first I want to just make a comment. I think most of the states, clearly the states in
the room, because the ones that aren't in the room, there might not be a cognizant agency
that actually is involved in the regulation, if it's a registration with the secretary of state
or what-have-you, but fortunately, there is another state within our state that someone can
go get a better rate at or what have you. So I think that's clearly one thing that I was
thinking about. There is an alternative on per se that the school has if they want to be
serving your students, they want to be in your state. They have a set of laws, a little bit
different, but 1 certainly appreciate the parallel, and I think important to add-
Shireman- I was really more talking about the parallel with accrediting and it's not as
much of an issue with states
Host- right so you know, 1 sympathize with that, good parallel. Did the Dept, was there
any discussion when gainful employment issue was being looked at, to say, can somehow
the Dept tap into infom1ation at the Dept of Labor or at the prior [X]. I understand there
may be some challenges there, but we're trying to measure whether or not this individual
is 1, either employed or perhaps employed and had benefited because of the education or
training that they had received and somehow reached their salary or what have you. Was
there any discussion around that area, because it seems like there may be some ways of
doing that?
Shireman -Yes, one of the areas of discussion, so part of the areas we are looking at.;
some of the criticism that has been out there has been about the use of averages or 25lh
percentile of BLS occupational data. And the reason we suggested use of BLS was not as
- it was actually as a way to reduce the work that would need to be done by whoever is
coming up with the income information. So for example, it's going to be the schools
having to figure out what their graduates eamed, we figured why not just have a level
that's an average based on the industry that will eliminate a lot of the school's need to
follow up and look at their own graduates. In other words, if it meets- if the debt I
income ratio is fine given the earnings in the occupation generally, you don't need to
worry about it as a school. Just look at your debt and if it's okay, flne. You only need to
look more detailed if you don't, if it doesn't beat it that way, because maybe your
graduates are earning something more. It was really to try to reduce the number that
would have to go through a process, either using IRS data or a state data system. States
like FL are in a much better position to be able to look individual programs and have this
kind of information using their U[, unemployment insurance, database. So there's a lot
that can be done there, and the discussion at and since negotiated rulemaking has helped
us to think through those issues.
Host- know there's going to be some questions ...
Michael Cooney- Bob, [patt] of your presentation is, do you believe that for profit
institutions are serving their students well?
Shireman -1 think they have to be given their demand, given the number of people that
are going to for-profit institutions, I think it's, I think they are, absolutely. All of them for
all students, I think that's where our role comes in
Cooney- following up on that, then your gainful employment provisions as we
understand it, according to CCA would approximate 300K students from the possibilities
of receiving an education. Is that an unintended consequence or are you all cognizant of
that?
Shireman - there isn't a proposal, and my argument with the- and ive had these
conversations directly with CCA, is work with us about what the definition should be. So
we can get in- what they've done is say, 8% pure labor statistics, these folks are crazy!
But the reality is, thee only way to get to a rule that makes sense, and we need to have a
definition, is to get down in the weeds and start working out the details. And they and
others have come back with, id say some more constructive thoughts in recent weeks, and
those will be considered in the proposed rule, and then they' ll be further input that I hope
we' ll get in a final rule.
[Ken Miller] -Mr. Shireman, im concerned about national averages. Some states are in
much better shape employment-wise than others. I'm fiom Ohio, a county in Ohio that
has 17% unemployment. Question - how can you adjust this rule, this definition, for
those kinds of anomalies between states, and should you?
Shireman -So using actual data from students as was suggested would deal with that
completely. The BLS data actually is available on a state basis so that could potentially
be another option. So this is where we are, we're open to suggestion and ideas about what
would be the best way, or whether it should be various possibilities that a school might
have an option to use. This is that input-seeking process for what would be the right kind
of approach. Again, part ofthe reason we just used the national average was because, to
eliminate a whole lot of work so that fewer needed to look in a more detailed way, and in
some ways I suppose if we had not suggested an approach to reducing work, we wouldn't
have had the BLS stuff on the table and wouldn't have the criticism, but I think it does
make sense, because it was a reasonable, work-reducing item that we put on the table
during negotiated rulemaking.
[Tom CasseJJ(?)] -have you looked at the idea of a value added concept in the
definition?
Gainful to me means you've gained, improved. So if we looked at our students and
where they came to us and what their earni ngs were afterwards, a measurement of change
of gain as the definition?
Shireman - so specific suggestions about that proposal and how to do it, we welcome
that, I mean that's very useful. So in the proposed rule period, that's especially timely.
Make sure, Especially from this point forward, make sure that ideas are considered, that
comment period after we publish the proposed rules. That's the kind of thing that would
be great to work out the details and suggest it during that comment period.
[Keith Boss (?), Northwest Technical Institute) - layers of control can sometimes
create layers of bureaucracy as you well know, you've got to struggle with that. There's
al ready of default rate, and we're very proud of our very low default rate, was it ever
considered that that 's enough control to see that students have the ability to pay offtheir
student loans and why would there need to be more layers of control, more bureaucracy
than that?
Shireman- the discussion at the table around default rate had to do with the fact that the
federal government has made sure that ppl who take out at least federal student loans
aren't put in a situation where it's a choice between food on the table and paying rent and
paying off your student loan. So we have things like forbearance, and if you ask for
forbearance, you could make no payments at all and you're not in default, there's income
based prepayment where you may be making nothing or next to nothing, but if you're on
income-based prepayment, there's very low repayment or maybe nothing at all and
you're not in default. So default rate doesn't measure whether or not people are actually
earning any money.
[Ray]- one of the problems ppl have come to wrestle with this concept is that any
fommla that is derived that results in the elimination of eligibility for a program has the
potential to have a school on the NE corner vs. a school n the SE corner delivering the
same program at the same level of efficiency and quality, but merely because of the
borrowing characteristics of their student population, one school's program could be
declared ineligible while the other school's program continues ... further, as you go
through this and eliminate program eligibility, the students don't go away, they seek out,
in many cases, the same education, in another institution. You could end up with a deluge
in the public institutions because they're not subject to the sanctions of a loss of
eligibility and yet, so it's okay to continue to train that person and for them to continue to
have debt and for state dollars to continue to subsidize their education at a public
institution but meanwhile the resulting product is the same?
Shireman- so what's your suggestion for a definition of gainful employment?
[Ray]- well I don't know- I always thought it was defined in the IRS code, the Social
Security admin ....
Shireman- I need a citation, tell me what we can do, as the US Dept of Ed, to detennine
whether a program is in or out.
[Ray]- I honestly don't know
Shireman- this is the problem. If you cant give me a definition, we're not making any
progress. So anything- I can respond to the details of your question, things like the 90/10
rule - any time you draw a line, there will be people right on one side and people on the
other side of the line that are very similarly situated. But that's what lines are about, so
your suggestions about what would be, and I think asking your colleagues, your own
institution, how do we decide, how do we know our programs are eligible for finical aid,
would help for you to develop some suggestions for us.
Host- see I knew we should have had a workshop on gainful employment.. we've all sat
around trying to figure it out. .. we have some good thinkers in the room ..
Russell - APEI -I have sort of a detailed question im going to pose for you. You
mentioned earlier [xx] with regard to title IV funds, but those fund are not necessatily
earmarked specifically for education costs, or direct education costs, and I think we all
recognize the reality is that many student do in fact realize substantial amount of money
for impotent, incidental spending, a new car, walking around, whatever. My question is,
lets take a scenario where an individual sent to community college for a couple years
borrows $35-40k, then that graduate comes to APEI system, where it will cost you $30k
to get a BA degree from start to finish, we inherit $35-40K worth of debt and add it to the
$30k the student incurs with us, we've got $75-80k of debt. Is that going to count against
us for the purposes of calculating gainful employment?
Shireman -we haven't proposed a rule yet but what we discussed a the table in
negotiated rulemaking, the answer would be no. That's because we proposed using the
median, and using the median means that all those outliers don't pull up the average. It's
the median, schools don't generally have more than half of their students coming in with
huge amounts of debt from someplace else. Now there's also the option in the rule of
saying, it's only debt you incur at the institution, so in the proposed rule in the comment
period will be the time to look back to a particular issue. There's debt/ income ratio in
the rule, and you can provide commentary. Again this is the kind ofthing, the kind of
discussion we needed a year ago and have begun to get recently and I really do appreciate
because they're useful and constructive discussions to have.
Russell - well the reason, ifl can just add another comment to that, I think it's important
to have [other debt] considered because if you take that approach that im only
accountable for my own debt-
Sh.ireman- Right, so to be clear, it's just a proposed rule. Comment period, final rule.
So we've gotten this kind of input, they'll be a proposed rule. If we haven't addressed
this issue and it needs to be addressed further in your view, there will be another
opportunity for you beyond the opportunities you've had in the last several weeks.
Russell - The one thing I would like not to see happen is that for admissions staff to say
we don't want to admit you because you've got debt.
Shireman - understood.
[NA] - can you tell me how this would apply to a public university or state college?
Shireman- well what we discussed at the table, in negotiated rulemaking- and I keep
repeating that because the "this"- we haven't proposed anything, we brought discussion
items to the table during Negotiated rulemaking, but it would mean that for shorter tem1
certificate programs, if more than half of the completers had debt, then the median debt
level of all completers, in other words not just the debt level of those who borrowed, but
the median debt level of all the people who complete the program, that would be the debt
level that would be compared against the expected earnings in the occupation the
program is preparing people for or the set of occupations. And then the discussion earlier
was about well, do you use actual earnings for people who are in the program or do you
use averages or 25th percentile with BLS so there's that question about what the measures
of income should be, so that would be - the concept was that as a standard. And again, at
the table we also had other suggestions for an institution if they didn't meet that
particular measure like graduation placement rates and loan prepayment rate.
LNAJ -so are you going to be asking them to post their placement rates as well/
Shireman- well placement rates have to be posted starting this July anyway, or
placement information. if you have rates, rates need to be posted.
LNA ) -so if they don't do it now, if they don't have placement info now . ..
Shireman - placement information was done as part of a rule done in - Congress
required all schools to post placement information in the Higher Education Opportunity
Act, and that goes into effect under the rule that was adopted last year so that goes into
effect this July.
[John Weir(?)] -from a state regulatory perspective, appreciate the Dept's effort to try
to address the cost issue, and one of the things that we've found at the state level is that
students don't seem to be price conscious when it comes to education. And what do you
think we can do at the state level to help make students better consumers, particularly as
to the cost issue?
Shireman- I think we've seen some improvement in that because of the economic
downturn, are seeing families chop more and really look at, does it really make sense for
me to spend $30-40K a year, what about the state university ... what about other options?
So I think it has improved somewhat, but more needs to be done. There will be, again
because of the Higher Education Opportunity Act, starting I believe this summer, schools
will now post net tuition - I cant remember if it's net tuition or net cost of attendance-
anyway some net figures by income bands so people will actually be able to go and say,
for someone in my fami ly income background, it's more common to pay this price rather
than the other price. And starting in, I think it' s a year and a half, all colleges will have to
have a net cost calculator, where a person can put in their income and other basic
information and get a basic sense of what their net cost would be. So that helps to deal
with some of the issues, we've provided some more tools to help people get more
information and to help them also deal with reality that sometimes a college that looks
really expensive based on its sticker price for a particular family, especially lower income,
may not actually be that expensive. So there's some new tools coming forward. I do
worry that just more information on the internet doesn't really help that much, and that
figuring out who we can provide tools more directly to teachers, counselors, people at
one stop job training centers. You know, our website, when people are actually applying,
finding those moments, those teachable moments when people are actually thinking
through these issues, so suggestions you have about the ways that we can do that, Pell
programs, state programs- we're certainly interested in ideas. It's important and not
easy to figure out how you teach the vast public things that aren't simple.
[Julie]- I guess this is more of a comment, but im somewhat concerned about the
potential for consumer confusion with the net cost calculator and different types of
disclosers about tuition. I know at the state, I'm in Tennessee, we've required that our
enrollment agreements contain the total cost of the program for a very long time. We now
require that the tuition rate be posted on the internet, we've required disclosure of
completion, referral rate, placement rates for a good while. We now do in depth audits of
placement data, so if student looks at our website, it's likely the numbers that they'll see
are going to be very difierent from maybe what they see on the F AFSA or some other
disclosure on the federal side. So how do we address the very likely possibility of
confusion, not knowing different methodologies and standards?
Shireman- I think you're right, there is going to be confusion. And even me, when I
:first started talking about net tuition, I couldn't remember, is it net tuition, is it net cost?
What's included in the definition? And I think there will be confusion, I guess the
positive side of that is, usually when you take a step and it creates those problems, it then
forces us all to sit down and say, how can we improve on what we had come up with
originally? Sort of like, data that's not perfect. Like, we cant post that data, it's not
totally clean. But the way to get people to clean their data is, you post it. And when we
started posting the student debt data that universities had reported to US News & World
Report, and universities would call and say, well that's not our figure and we'd say, well
that's what you reported to US News. And they'd go back, and sure enough, somebody
had reported something wrong, s I think that you're right, there will be some confusion
but that will probably force the issue to some degree. But id also welcome- I mean I'd
love to look at what Tennessee does and take it to some of our folks and ask about what
we can do, A lot of- for some reason Congress in the education space, especially the
higher education space, gets very detailed in the legislation's actual statutory language,
which then makes it very difficult for us in the implementation to do things differently
that what they had prescribed. So we may not have a lot of flexibility in terms of our
implementation but maybe there's some suggestions we could make back to Congress by
pointing out to them these issues for particular states.
[Ray]- there seemed to be a reluctance on the part of Dept officials as well as most
people in the room during the NegReg process to talk about the troublesome aspect of
schools not being able to limit student borrowing,. It seems that we have a statute, a rule
that leaves the amount that students borrow entirely up to them. It's based on fom1ulas,
numbers across[] program lengths are mandated by state laws in many fields of training,
[licensure], all these things are out ofthe control of the institution, they cannot seem to
limit the amount that students borrow except in very specific case-by-case basis, and then
they subject themselves to lawsuits by students claiming they were discriminated against.
Might there be any opportunity to open up and look at some of those rules so that we
might be able to avoid some serious unintended consequences of a formula that could
cause havoc?
Shireman- student loan program and the Pell grant program are entitlements. They are
entitlements for covering tuition, fees, books, supplies, room & board, and to restrict an
entitlement program is a pretty big deal, and that's why it's difficult for us to do in tem1s
of regulations. Now you do have the ability, if you believe someone would default on
their loan, I think that's the ability that the fin aid administrators have ... you're also able
to establish what you feel to be reasonable expenses including in the room & board space.
But restricting loans beyond that would likely be something Congress would need to
consider, and I think part of the reason Congress has had a hard t ime thinking this one
through is higher education has benefitted enormously from Pell grants and student loans,
Pell grants working like an entitlement program on an annual basis, and student loan
programs being an entitlement program. And going down the road of starting to restrict
the entitlement, I think is something to do very cautiously if we want to make sure people
get the need they are eligible for.
END.
l(b)(S)
From: Bergeron, David
Sent: Friday, April 30, 2010 2:37PM
To: Kanter, Martha
Subject: FW: Tony Miller briefing
From: Weko, Tom
Sent: Friday, April 30, 2010 2:35 PM
To: Bergeron, David
Cc: Miller, Elise; Hunt-White, Tracy
Subject: Tony Miller briefing
(b)(5)
From: Kanter Martha
To.: Yuan. Georgia
CC:
Date: 4/30/2010 2:49:24PM
Subject: FW: Tony MiiJer briefing
(b)(5)
(b)(S)
(b)(5)
(b)(S)
1 The "New Majority" of Undergraduates:
2 Students in Community Colleges and
3 For-Profit Institutions
4 Introduction
5 Depictions of postsecondaty education in the news, films, and oth-
6 er media often focus on undergraduates who study at traditional
7 4-year public universities and nonprofit colleges (Swidey 2009;
8 Russo 2006; U.S. Department of Education 2002). Yet even two
9 decades ago these students accounted for just 55 percent of stu-
l 0 dents enrolled in U.S. postsecondaty education (Choy and Gifford
1990).
1
Now a majority of undergraduates in U.S. postsecondaty
education attend community colleges and for-profit institutions.
These two sectors together enrolled 54 percent of the approximate-
ly 21 million undergraduates who attended U.S. postsecondary
institutions in 2007-08. Some 44 percent of undergraduates at-
l6 tended community colleges, the single largest sector of U.S. postse-
condary education, and another 10 percent attended for-profit
institutions (figure 1). While the for-profit sector accounts for a
relatively small percentage of undergraduates, its enrollment is
growing at a rapid pace relative to other postsecondaty sectors
(Horn and Li 2009; Wilson 2010).
Who are the undergraduate students served by community colleges
and for-profit institutions, and how do they compare to students
who are enrolled at traditional 4-year institutions? Drawing on data
from a nationally representative sample of undergraduates enrolled
L Based on fall enrollment in the 1986-87 National Postsecondary Stu-
dent Aid Study (NPSAS:87).
THE "NEW MAJORITY" OF UNDERGRADUATES:
STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS
1 in U.S. postseconda1y institutions, this Statistics in Brief27 enrolled in bachelor's degree programs, while 43 percent
2 addresses the following questions. 28 were pursuing associate's degrees, and the remaining stu-
29 dents were either in certificate programs or not in a de-
3 How do students enrolled in for-profit institutions and 30 gree program (table 1).2 The average 2007-08 tuition
4 community colleges differ from each other and from 31 and fees for full-time, full-year students was $11,900,
5 those enrolled in traditional4-year institutions in terms 32 and for all students, $10,200 (Wei forthcoming).
6 of:
7
8
9
10
11
12
13
33 Community colleges: Also known as public 2-year institu-
1) demographic and socioeconomic characreristics?34 tions, community colleges offer associate's degrees in
2) indicators of high school academic preparation 35 occupational and academic fields, as well as occupational
and being the first in their family to attend col- 36 certificates. Unlike for-profit institutions, which offer
lege? 37 primarily occupational programs, community colleges
38 offer a variety of progran1s with different goals. These
3) enrollment patterns and major fields of study?
39 include academic transfer to 4-year institutions, voca-
4) the extent to which they combine work and
study?
40 tionalltechnical education, continuing education, and
41 developmental education (Cohen and Brawer 2003).
14 The three groups of postsecondary institutions comparcJ2 Conu11w1ity colleges, therefore, serve a range of students,
15 in this Brief are briefly defined below. 43 from those needing basic remedial courses to very high
44 achieving students meeting lower-division bachelor's
16 For-profit institutions: A recent report by the U.S. Gov- 45 degree requirements before transferring to a selective 4-
17 ernment Accountability Office defined propriety institu-46 year college or university (Provasnik and Planty 2008).
18 tions as "Institutions that are privately-owned whose net47 The average 2007-08 tuition and fees for full-time, full-
19 earnings can benefit a shareholder or individual; that is, 48 year students was $2,400, and for all students, $1,200
20 for-profit institutions." (U.S. GovernmentAccountabili-49 (WeiforthcominiJ.
21 ty Office 2009). Among undergraduates enrolled in for-
22 profit institutions in 2007-08, about half attended insti-50
23 tutions authorized to award 4-year (bachelor) degrees, 27
24 percent attended 2-year institutions, and 22 percent at-
25 tended less-than-2-year institutions (figure 2). Half of
26 those in for-profit 4-year institutions were
2
At community colleges, 79 percent of undergraduates were in
associate's degree programs, 7 percent in cerrillcate programs,
and 11 percent were not in a degree program. In traditional 4-
year institutions, 93 percent of lmdergraduates were enrolled
in bachelor's degree programs.
THE "NEW MAJORITY" OF UNDERGRADUATES: I I
STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS 3
1 Public and private nonprofit 4-year institutions: Public 28
2 and private nonprofit institutions serve as the main 29
3 comparison group in this analysis. Consistent with pre- 30
4 vious research, they are referred to as "traditional4-year 31
5 institutions" (Tierney and Hentschke 2007). Together, 32
6 students in these two sectors accounted for 46 percent of
33
7 all undergraduates in 2007-08-32 percent attending
34
8 public 4-year colleges and universities, and 14 percent
35
9 attending private nonprofit 4-year institutions (figure 1)
36
ic), and income (larger proportion oflow-
income students).
3
Community college students
also differ from traditional 4-year college stu-
dents on most of these characteristics, but to a
lesser extent.
4
The part-time status of the majority of commu-
nity coll ege students sets them apart from their
peers in for-profit and traditional 4-year institu-
tions, the majority of whom attend full time.
10 The average 2007-08 tuition and fees for full-time, full-
11 year students in 4-year public institutions was $5,700
37
12 and $6,300 and for all students, $4,100 and $6,300, at 38
Health care fields are the most common fields of
study among students in for-profit institutions,
and arean1ong the most common fields for
community college students.
13 non-doctorate- and doctorate-granting institutions, re- 39
14 spectively. Tuition and fees in private nonprofit institu- 40
15 tions totaled $20, 800 and $25,800 for full-time, full- 41
Regardless of where undergraduates enroll, a
majority work while enrolled.
16 year students and $15,200 and $20,700 for all students, 42
17 at non-doctorate- and doctorate-granting institutions,
18 respectively (Wei forthcoming).
19
20
UNDERGRADUATE AWARDS CONFERRED
Table 2 shows the distribution of undergraduate
awards for the institution comparison groups
(shaded) presented in this report. These comple-
tions data are reported by all institutions that partic-
ipate in Title IV student aid programs. Based on
these data, community colleges awarded 46 per-
cent of all certificates and 69 percent of associate's
degrees in 2007. For-profit institutions awarded 42
percent of all certificates and 16 percent of asso-
ciate's degrees. For-profit 4-year institutions
awarded about 5 percent of bachelor's degrees
and traditional 4-year colleges and universities
awarded the remaining 95 percent.
21 Key findings
43 Demographic and Socioeconomic
44 Characteristics
5 Many demographic characteristics, including gender,
6 race/ethnicity, family status, income and age distinguish
7 students enrolled in for-profit institutions and commu-
8 nity colleges from those enrolled in traditional 4-ycar
9 colleges. Students in the for-profit sector, for example,
0 were the most likely to be female, Black or Hispanic, to
1 have children or other dependents, and to be single with
2 dependents in 2007-08 (figure 3). Women constituted a
3 majority of undergraduates in all three sectors, but com-
4 prised 69 percent of students in for-profit institutions. fn
5 contrast, 57 percent of community college students and
6 55 percent of traditional4-year institution students were
7 women.
22 Students enrolled in for-profit institutions differ
3
All comparisons of estimates were tested for statistical signi-
ficance using the Student's !-statistic, and all differences cited
are statistically significant at the p < .05 level. No adjustments
for mltltiple comparisons were made. The standard errors for
the estimates can be found at
http://nces.ed.goy/das/library/rs:ports.asp.
23 from those enrolled in traditional 4-year institu-
24 tions across a wide range of demographic and
25 socioeconomic characteristics, including gender
26 (larger proportion of women), age (older),
27 race/ethnicity (nearly half are Black or Hispan-
4
Determined by a Student' s t-srarisric that rests the signillc-
ance of the difference of differences.
THE "NEW MAJORITY" OF UNDERGRADUATES:
STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS
1 Nearly half of the students in for-profit institutions were35 students were 31 percent for dependent students and 22
2 Black (25 percent) or Hispanic (21 percent). In commu-36 percent for independent students.
3 nity colleges, the percentages of Black (14 percent) and
4 Hispanic students (15 percent) were lower than those in 37 The income status of community college students rela-
5 for-profit institutions but higher than those in tradition-38 tive to their peers in traditional 4-year institutions dif-
6 al4-year institutions (1 1 and 12 percent, respectively, 39 fered for dependent and independent students. A larger
7 were Black or Hispanic).
40 percentage of dependent community college students
4 1 were from low-income families (31 vs. 20 percent), while
8 Community college students also fel l between for-profit 42 a smaller percentage of independent students earned low
9 and traditional 4-year institutions in terms of their like- 43 incomes (22 vs. 28 percent) .
10 lihood of having dependents and being single with de-
11 pendents. Some 32 percent of communi ty college
44 The average age of students enrolled at conummity col-
12 students had children or other dependents, versus 49 anJ5 leges and for-profit institutions was older than the aver-
13 14 percent in for-profit and traditional 4-year institu-
14
15
tions, respectively. The same pattern was found for the
percentage of undergraduates who were single with de-
16 pendents (16 vs. 31 and 7 percent) .
46 age age of students at traditional 4-year institutions (28
47 vs. 24) (figure 5) . Although the average age of students
48 in both for-profit institutions and communi ty coll eges
49 was 28, the distri bution of students by age group in each
50 type of institution differed (figure 5) . Undergraduates
17 When income is considered for postsecondaty students, 51 enroll ed in for-profit institutions were least likely to be
18 family income is reported for students considered finan- 52 traditional-aged students-23 or younger.
19 cially dependent on their parents, and student (and
20 spousal) income is reported for chose considered finan- 53
2 1 cially independent. Dependency status is based largely 54
22 on age: all students 24 or older are considered indepen- 55
23 dent. Married students and students who are parents are
6
24 also considered independent. Reflecting these characte-
5
25 ristics, just over three-quarters (76 percent) of students
57
26 in for-profit institutions were independent (figure 4). I n ~ ~
27 contrast, 57 percent of community college students and
60
28 32 percent of students at traditional4-year institutions
61
29 were independent.
62
30 Students in the for-profit sector were the most likely to
31
32
63
64
be low income; roughly half (51 percent) of dependent
students had low family incomes, and roughJy one-third 65
33 of independent students earned low incomes (32 per-
66
Indicators of High School Academic
Preparation and First in Family to Attend
College
Students in for-profit institutions appeared to be the
least academically prepared based on indicators of high
school completion and academic preparation available in
the data. Indicators include high school completion sta-
tus (standard diploma), average grades, and the number
of years of mathematics taken. As illustrated in figure 6,
compared with their peers in community colleges and
traditional 4-year institutions, students in for-profit in-
stitutions were the most likely co have enrolled in postse-
condaty education without a standard hi gh school
diploma (19 percent, vs. 14 percent and 7 percent, re-
34 cent) .s Comparable percentages for conununity college 67 spectively, of community college and traditional4-year
5
Low income is defmed as a parent (dependent wldergra-
duates) or student (including spouse's income, for indepen-
dent wldergraduates) income that is at or below the 25th
percentile of the income distribution for each group. In 2006
(the year used for 2007-08 financial ajd applications) the in-
come levels were at or below $36,000 for dependent students
and at or below $11,000 for independent students.
THE "NEW MAJORITY" OF UNDERGRADUATES: I
STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS 5
1 college students) .
6
They were also least likely to report 30 in for-profit institutions attended full time for the full
2 earning grades ofB or higher (54 vs. 6 1 and 81 percent),31 year and 33 percent attended full time for part of the
3 or completing 4 or more years of high school mathemat-32 year. A higher percentage of undergraduates attended
4 ics (48 vs. 54 and 76 percent) ? At the same time, about 33 full time, full year in traditional4-year institutions (59
5 half of for-profi t students were the first in their famil y to34 percent) than in for-profit institutions.
6 enroll in postsecondary education, as measured by the
7 level of education achieved by either parent (figure 6). 35 Not ali students attend classes on campuses, but instead
8 SpecificaHy, among for-profi t students, 51 percent were 36 participate in distance education. Students in communi-
9 from families in which neither parent had an education 37 ty coll eges and for-profit institutions di d so more often
10 beyond high school, compared with 39 and 25 percent, 38 chan their peers at traditional4-year colleges. About one-
11 respectively, of community coll ege and tradi tional 4-yea
1
39 quarter of community college students and about one-
12 college students. 40 fifth of students in for-profi t institutions reported taking
41 distance education courses during the 2007-08 academic
13 Enrollment Patterns and Major Fields of
14 Study
42 year (figure 8) . In contrast, 17 percent of students at tra-
43 di tional 4-year institutions reported the same. In addi-
44 cion, a larger percentage offor-proflt enrollees (12
15 The part-time enrollment patterns of conumrnity collegC45 percent) than those at commttnity colleges (3 percent)
16 students distinguish them from their peers in both for- 46 reported that their entire program was offered through
17 profi t and traditiona14-year institutions (figure 7) . Se- 47 distance education.
18 venty percent attended part time, while the same percen-
19 cage of students in traditional 4-year and for-profit 4 8 With respect to fields of study, community college stu-
20 institutions attended full time. 49 dents sought both academic and occupational prepara-
50 cion (figure 9) . For example, 7 percent of community
2 1
22
23
While part-time attendance offers a flexibility often
5 1 college students and 22 percent of students in traditional
needed by studenrs attending community colleges (Co- 52 4-year institutions majored in humanities and social
wen and Brower 2003) , it is also associated with lower 53 sciences, compared with 2 percent of students in for-
24 rates of degree completion relative to full-time atten-
25 dance (e.g., Berkner, He, and Cataldi 2002).
54 profit institutions. Another 20 percent at community
55 coll eges and 6 percent at tradi tional 4-year institutions
56 majored in general studies and other fields, compared
26 Looking at attendance patterns over the year, in
57
27 2007-08, some 38 percent of community college stu-
58
28 dents attended part time for part of the year and 32 per-
59
29 cent for the full year.
8
By contrast, 37 percent of
with less chan 1 percent at for-profit institutions. Com-
munity coll ege students majoring in humanities, social
sciences, and general studies are largely those who seek to
transfer to a 4-year college (Berkner, Horn, and Chrne
6 1 2000).
6
These students may have dropped out of high school, ob-
tained aGED or high school equivalency certificate, been 62 Health care fields accounted for one-third of undergra-
home schooled, or attended a foreign high school. 63 duace majors in for-profit institutions and about one-
7 Data on high school academic preparation (GPA and
courses) is available only for students younger than 30, which 64 fifth (21 percent) in community coll eges. In conuast, 9
accowued for 77 percent of alltmdergraduares (figure 5). 65 percent of students in traditional 4-year institutions ma-
8 A full year is defined as 9 or more months. Those attending 66 jored in health care fields. l n addition to health care
part year may have completed a program of study or left with-67 fields, students in for-profit institutions were the most
out completing, either temporari ly or permanently.
68 likely to major in fields that are grouped under the head-
I I
THE "NEW MAJORITY" OF UNDERGRADUATES:
6 STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS
1 ing "other applied" (26 vs. 16 and 14 percent at com- 16
2 munity coll eges and tradi t ional 17
3 4-year institutions, respectively). These fields include 18
4 criminal justice, communications, and design and ap- 19
5 plied arts, among others. 20
1
HEALTH MAJORS AT COMMUNITY COLLEGES AND
2
FORPROFIT LESS THAN4YEAR INSTITUTIONS 3
4
Although health care fields were among the most
5
common majors at less-than-2-year and 2-year for-
profit institutions and community colleges, the mix
6
of health majors at each institution type varied.
Figure 10 shows the percentage of health majors
7
at community colleges and 2-year and less-than-2- 8
year for-profit institutions in five fields of study. To-
9
gether, these fields account for about 43 and 46
0
percent of majors at less than-2-year and 2-year
for profit institutions, respectively, and 20 percent
1
at community colleges. A higher percentage of stu-
2
dents majoring in health care fields majored in
3
nursing at community colleges (59 percent) than at
less-than-2-year (12 percent) and 2-year (1 0 per-
4
cent) for-profit institutions. By contrast, higher per-
centages of students at for-profit i nstitutions
5
majored in health and medical administrative ser-
6
vices (21 and 29 percent for less-than-2-year and
2-year institutions, respectively, vs. 5 percent at
7
community colleges) and allied health and medical 8
assisting services (21 and 27 percent for less-than-
9
2-year and 2-year institutions, respectively, vs. 4
0
percent at community colleges) .
1
6
2
7 Combining Work and Study
43
44
8 More than two-thirds of students attending for-profit
9 institutions below the bachelor's degree level and about 45
10 one-half of community college students indicated that 4
6
11 th
fi 11 b
47
. .e1r pnmaty reason or enro mg was to acqUire JO re-
12 lated skills or credentials (figure 11). Many of these stu-
13 dents work while attending classes. l n fact, regardless of
14 where undergraduates enroll , a large majority of students
15 work while enrolled (figure 12).
Community college students were the most likely to
work while enrolled (81 percent) and to work full time
(41 percent of all community college students). Howev-
er, even though a smaller percentage of students in for-
profit institutions attended part time (figure 7), nearly
three-quarters (73 percent) worked whil e enrolled and
39 percent of all for-profit students worked full time
(figure 12). Students in traditional 4-year institutions
were the least likely to work while enrolled, yet 69 per-
cent did so.
When working students were asked if they considered
employment or going to school as their primaty activity,
a majority reported the latter.
9
This was the case for stu-
dents enrolled in all three institution types. About 60
percent of students in both community colleges and for-
profit institutions reported that their primaty role was as
a student, as did 79 percent of students in 4-year institu-
tions.
Find Out More
See Related NCES Products
More detailed information on 2007-08 undergraduates
enroll ed in U.S. postsecondary institutions can be found
in Web Tables produced by NCES using the NPSAS:08
data. These web Tables are a comprehensive source of
information on undergraduate students during the
2007- 08 academic year. Included are estimates of de-
mographics, enrollment, and employment characteris-
tics. In addition, Web Tables documenting how students
pay for their undergraduate education are also avail able.
Web Tables-Profile of Undergraduate Students in U.S Postse-
condmy Institutions: 2007-08 (NCES 2010-205).
[link will be added]
9
Students who worked while enrolled were asked the foliow-
tng question: "Would you say you were primarily a student
working to meer expenses or an employee who decided ro
enroll in school. "
1 Web Tables-Student Financing of Undergraduate Education:
2 2007-08 (NCES 2010-162). [link wil l be added]
3 Readers may also be interested in the following NCES
4 prod ucts related to the topic of this Statistics in Brief:
5 Changes in Awards Below the Bachelors Degree (NCES 201 0-
6 167). 1 OL2Ql QlGZ.pdf
THE "NEW MAJORITY" OF UNDERGRADUATES: I I
STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS 7
33
34
35
36
37
38
39
pondents. Estimates were weighted to adj ust for the un-
equal probabili ty of selection into the sample and for
non response. For an overview of the survey methodolo-
gy, see appendix B of the report 2007-08 National Post-
secondary Student Aid Study (NPSAS:08): Student
FinanciaL Aid Estimates for 2007- 08: First Look
(http://nces.ed.gov /pubs2009/2009166.pdf).
7 The Price of Undergraduate Education: 2007-08 (NCES 201 0-
8 175). [link to be added] 40
T he fi ndings presented here are purely descri ptive in
nature. Al l estimates presented in this Statistics in Brief
were produced using the Data Analysis System (DAS), a
web-based software application that enables users to gen-
9 Student hnancing of Undergraduate Education: 2007-08
10 (N CES 2009-175). [li11 k to be added]
4 1
42
43
1 1 Undergraduate Financial Aid Estimates by Type oflmtitution in 44 erate tables for most of the postsecondary surveys con-
12 2007-08 (NCES 2009-201).
13 http:Unces.ed.gQvlpubs2Q09/2Q092Ql .pdf
14 2007-08 National Postsecondary Student Aid Study
15 (NPSAS:08): Student Finan,ial Aid Estimates: First Look
16 (NCES 2009-166).
17 http://nces.ed.gov/pubs2009/2009166.pdf
4 5 ducted by the National Center for Education Statistics
46 (NCES). T he DAS also contains a detailed description
47 of how each variable was created and incl udes the word-
4 8 ing of questions for variables coming d irectly from the
49 interview. With the DAS, users can repli cate or expand
50 upon the figures and tables presented in this report. For
18 Run Your Own Analysis 51 a description of al l available options, users should access
19 Direct access to the data used in this Statistics in Brief is 52 the DAS website. If users are new to the DAS, the DAS
20 avail able through NCES' s DataLab site 53 User Help Center provides online tutorials on how to
2 1 (http://nces.ed.gov/datalab/). 54 use al l functions of the DAS. The DAS is available at
22 Technical Notes
55 http://nces.ed.gov/DAS.
23 The data in this report come from the National Postse- 56 Variables Used in This Statistics in Brief
4 d d Ps S) dm d
57 Information on al l the variables used in this report can
2 condary Student Ai Stu y (N A a inistere over
25 the 2007- 08 year. T he NPSAS target population in-
58 be obtained from the NPSAS:08 DAS. Users can search
26 II d I
59 the DAS for the label or variable nan1es specified below.
dudes all eligibl e students enro e in T it eN institu-
27 tions in the United States at any time between July 1, 60 T he foll owing variables included in chis Brief had a
28 2007 and June 30, 2008.
10
The sampli ng design for col- 6
1
response race ofless than 85 percent:
29 lecting data consisted of two stages. T he fi rst stage in- 62 ATTEND MR, DEPANY, DISTED UC,
30 volved selecting eligible institutions from which students63 DISTALL, H CYSMATH, H SGPA, JOBENR,
31 would then be sampled in the second stage. Approx- 64 ]OBRO LE, PAREDUC, PCT DEP, PCTlNDEP.
32 imately 114,000 undergraduate students were study res-
10
Title IV institutions are those eligible to participate in the
federal fmancial aid programs incl uded in Title IV of the
Higher Ed ucation Act. These programs include the Pel ! Grant,
federal student loans, work-st udy, and other federal aid.
I I
THE "NEW MAJORITY" OF UNDERGRADUATES:
8 STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS
1 Age (AGE) 40 Choy, S. (200 1). Stur.knts Whose Parmts Did Not Go To Col-
2 Main reasons for attending institution (AITENDMR)
3 Attendance pattern (ATfNSTKD
4 Has dependents (DEPANY)
41
42
43
lege: PostucoruM.ry A cuss, Pnsistmu, and Attainmmt
(NCES 2001-126). Narlonal Center for Education Statis-
tics, U.S. Department of Education. Washington, DC.
5 Dependency status (DEPEND) 44 Cohen, A.M., and Brawer, F.B. (2003). The Ammcan Com-
6 Distance education: took courses in 2007-08
7 (DISTEDUC)
45 munity College {tlth ed.). San Francisco: Jossey-Bass.
8 Distance education: entire program (DIST ALL)
9 Gender (GENDER)
46
47
48
49
Horn, L. and Li, X. Changes in Awards Below the Bachelor's
Degree (NCES 2010-167). National Center for Educa-
tion Statistics, Insti tute of Ed ucation Sciences, U.S. De-
partment of Education. Washington, DC.
10 High school math courses planned/taken
J 1 (HCYSMATH)
50 Russo, A. (2006). Traditional College Students Not So Tradi-
12 High school degree type (HSDEG) 51 tiona!. This Week in Education: Alexander Russo's WeekLy
13 High school grade point average (HSGPA)
14 Work intensity whi le enrolled QOBENR)
52 Roundup of the Best Education News and Analysis. Re-
53 trieved December 16, 2009, from
15 Primary role as student or employee QOBROLE)
16 NPSAS institution level (LEVEL)
54
55
17 Field of study/major (detailed CIP codes) (MAJORS4Y)56
18 Field of study: undergraduate (10 categories) 5?
19 (MAJORS4) ; ~
20 Parent education (PAREOUC) 60
h up://tb jsweeki ned ucarion .blogspor.com/2006/02/
uad jrion a l-eo llegs:-sruden rs-no t -so. b tm I.
Swidey, N. (2009). The Four-Year College Myrh. The Boston
Globe Sunday Magazine. Retrieved November 25, 2009,
from
hnp://www.boston.com/bostonglobs:lmagazine/artkks/2
009/05/31/tbs: four year college myth/.
21 Tncomc percentile for dependent students (PCTDEP)
22
23
24
Income percentile for independent students
(PCTTNDEP)
61 Tierney, W. and Henrschke, G. {2007). New Players, Differmt
62 Game: Understanding the Rise ofFor-Profit Colleges and
63 Universities. Baltimore, MD: The j ohns Hopkins Univer-
Race/ethnicity (RACE)
64 sity Press.
25 I nstitution sector (SECTOR9)
26 Single parent (SJNGLPAR)
65 U.S. Governrnenr Accountability Office (GAO). {2009). Pro-
27 Undergraduate degree (UGDEG)
66
67
68
69
28 References
29 13erkner, L., Horn, L., and McCiw1e, M. (2000). Descriptive 70
30 Summnry of 1995-96 Beginning PostsecoruM.ry Students: 7I
31 Thm Yean Later (NCES 2000-154). NationaJ Center fol
2
32 Education Statistics, U.S. Department of Education. 73
33 Washington, DC. 74
34
35
36
37
38
39
Bradburn, E., Berger, R., Li, X., Peter, K., and Rooney, K. 75
(2003). A Descriptive Summary of1999-2000 Bachelor's 76
Degm Recipients 1 Year Later (NCES 2003-165). Nation77
al Center for Education Statistics, lnstlrute of Education 78
Sciences, U.S. Department of Education. Washington,
DC.
prietary SdJools: Stronger Department of Education Over-
sight Needed to HeLp Ensure OnLy Eligible Students Receive
FederaL Student Aid (GA0-09-600). Washington, DC:
Aurbor.
Wilson, R. (20 I 0). "for-Proflr Colleges Change Higher Edu-
cation's Landscape. Chronicle of Higher Education Febru-
ary 7, 20 I 0. Retrieved February 16, 2010 from
hnp://chronjcls:.com/arrjcls:/For-Profit-Colls:gs:s-
Changs:/640 12/.
Wei, C. (forthcoming) Web Tabks: Student Financing of Unrkr-
graduate Education {NCES 201 0-162). National Cemer
for Education Statistics, U.S. Department of Education.
Washington, DC.
1
www.ed.gov ies.ed.gov
Figure 1. WHERE UNDERGRADUATES ARE ENROLLED:
Percentage distribution of undergraduates, by type of
institution: 2007-08
Private
nonprofit
4-year
Community
colleges
44%
54%
1
Other includes public less-than-2-year and private nonprofit less-
than-4-year institutions.
2
Includes 4-, 2-, and less-than-2-year institutions.
NOTE: Estimates include students enrolled in Title IV eligible
postsecondary institutions in the 50 states, DC, and Puerto Rico.
Detail may not sum to totals because of rounding. Standard error
tables are available at http://nces.ed.gov/dasl librarv/ reports.asp.
SOURCE: U.S. Department of Education, National Center for
Education Statistics, 2007-08 National Postsecondary Student Aid
Study (NPSAS:08).
Figure 2. FOR-PROFIT INSTITUTIONS ARE AT ALL LEVELS:
Percentage distribution of undergraduates who
attended for-profit institutions, by level of institution:
2007-08
4-year
NOTE: Estimates include students enrolled in Title IV eligible post
secondary institutions in the 50 states. DC, and Puerto Rico. Detail
may not sum to totals because of rounding. Standard error tables
are available at http:lfnces.ed.gov/das/library/reports.asp .
SOURCE: U.S. Department of Education, National Center for
Education Statistics, 2007-08 National Postsecondary Student Aid
Study (NPSAS:08).
Figure 3. GENDER, FAMILY STATUS, AND RACE/ETHNCITY: Percentage of undergraduates with selected student characteristics, by type
of institution: 2007-{)8
Percent
100
80
60
40
20
For-profit'
Community
colleges
Public and private
nonprofit 4-year
0
0
69
57
Female
20
Percentage of undergraduates by selected characteristics
55
For-profit'
49
Percentage of
undergraduates with
children or dependents
Community
colleges
16
7
Percentage of
undergraduates who
were single with dependents
C Public and private
nonprofit 4-year
Percentage distribution of undergraduates by race/ethnicity
40 60 80
Percent
White OBiack CHispanic Asian Other or two or more races
1
Includes 4-, 2-, and less-than-2-year institutions.
100
NOTE: Black includes African American, Hispanic includes Latino, and Asian includes Pacific Islander. Other includes American Indian and
Alaska Native, and respondents having origins in a race/ethnicity not listed. Estimates include students enrolled in Tille IV eligible postsecondary
institutions in the 50 states, DC, and Puerto Rico. Detail may not sum to totals because of rounding. Standard error tables are available at
http:/fnces.ed.gov/das!library/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-Q8 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 4.
Percent
100
l
80
J
60
I
40
I
1
20
I
1
0
LOW-INCOME STATUS: Percentage of undergraduates who were independent and percentage in the low-income groups among
dependent and independent students, by type of institution: 2007-08
76
Independent
undergraduates
For-profit'
51
Low-income dependent
undergraduates'
Community
colleges
32
28
Low-income independent
undergraduates
2
oPublic and private
nonprofit 4-year
1
Low-income dependent undergraduates had family incomes of approximately $36.000 or less in 2006 (the year used for 2007-08 financial aid
calculations), which was at or below the 25th percentile of the income distribution for all dependent undergraduates.
2
Low-income independent undergraduates had personal (including spouse's income) incomes of approximately $11,000 or less in 2006 (the year
used for 2007-08 financial aid calculations), which was at or below the 25th percentile of the income distribution for all independent
undergraduates.
3
Includes 4-, 2-. and less-than-2-year institutions.
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Standard error
tables are available at http:/fnces.ed.gov/das/librarv/reports.asp .
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 5. AGE DISTRIBUTION: Average age and percentage distribution of undergraduates by age
and type of institution: 2007-08
Percent
100
..
80
G
60
40 36
20
0
All For-
undergraduates profit'
Average
age 26 28
1
Includes 4-, 2 , and less-than-2-year institutions.
30
Community
colleges
28
40 or older
45 0 24-29
Public and
private
nonprofit
4-year
24
20-23
19 or younger
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states,
DC, and Puerto Rico. Detail may not sum to totals because of rounding. Standard error tables are
available at http:/lnces.ed.qov/ dasllibrarv/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08
National Postsecondary Student Aid Study ( NPSAS:08).
Figure 6. STUDENTS' ACADEMIC BACKGROUND AND WHETHER FIRST IN FAMILY TO ATTEND COLLEGE: Percentage of undergraduates
who did not graduate from high school with a regular high school diploma, who reported having a high school GPA of 8 or
better, who reported taking 4 or more years of mathematics in high school, and who were the first in their family to attend
college, by type of institution: 2007-08
Percent
100
80
60
40
20
0
Students without
a regular high
school diploma'
For-profit'
81
61
54
Students with an
average high school
GPA equivalent
to grade B or better
Community colleges
76
48
54
Took 4 years or more
of mathematics
in high school
2
51
First in family to attend college
(neither parent attended
postsecondary education)
0 Public and private nonprofit 4-year
1
Undergraduates who did not have a regular high school diploma may have attended a foreign high school, been home schooled, or may have
earned a high school completion certificate, graduation equivalency diploma, or other equivalency, or have no high school degree or certificate.
2
Data on academic preparation in high school are available only for undergraduates younger than 30, who account for 77 percent of all
undergraduates.
3
Includes 4-, 2-, and less-than-2-year institutions.
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Standard error
tables are available at http://nces.ed .govfdas/libraryfreports.asp .
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007--08 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 7. FULL- AND PART-TIME ATTENDANCE: Percentage distribution of undergraduates by attendance status and by type of
i nstitution attended: 2007-08
Percent
100
80
~
a Part-time!
part-year
60
12
aPart-time/
33
full-year
14
40
Full-time/
part-year
20
11
a Full-time/
full-year
0
All undergraduates For-profit' Community Public and private
colleges nonprofit 4-year
1
Includes 4-, 2-, and less-than-2-year institutions.
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Detail may not
sum to totals because of rounding. Standard error tables are available at http://nces.ed.gov/das/library/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Postsecondary Student Aid Study
(NPSAS:08) .
Figure 8. DISTANCE EDUCATION: Percentage of all undergraduates who took a distance education course in 2007-08 and percentage of
all undergraduates who attended programs entirely taught through distance education, by institution type: 2007-08
Percent
100
80
60
40
20
0
L
21
24
17
Took a distance
education course in 2007-08
For-profit' Community colleges
1
Includes 4-, 2-, and less-than-2-year institutions.
12
3 2
Percent of undergraduates in programs
entirely taught through distance education
ClPublic and private nonprofit 4-year
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Detail may not
sum to totals because of rounding. Standard error tables are available at http://nces.ed.gov/das/ library/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Postsecondary Student Aid Study
(NPSAS:08) .
Figure 9. MAJOR FIELDS OF STUDY: Percentage distribution of undergraduates by major field of study and type of institution: 2007-08
Percent
100
80
60
40 33
20
22
20
li:n
70
6
9
6
2
..2.__o #
-
_ c:J
Health care fields Other applied' Business STEM
2
Humanities and Education General studies
social sciences and other'
For-profit
4
Community colleges a Public and private nonprofit 4-year
# Rounds to zero.
1
Includes personal and consumer services, manufacturing, construction. repair, and transportation, military technology and protective services,
architecture, communications, public administration and human services, design and applied arts, law and legal studies, and theology
and religious vocations.
2
Science, technology, engineering and math.
3
other includes basic skills, citizenship activities, leisure and recreational activities, personal awareness and self improvement, high school/
secondary diplomas and certificate programs, interpersonal and social skills.
4
Includes 4-. 2-, and less-than-2-year institutions.
NOTE: Excludes undergraduates who were not in a degree program or who had not declared a major field of study. Estimates include students
enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Detail may not sum to totals because of rounding.
Standard error tables are available at http://nces.ed.gov/das/librarv/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 10. MOST COMMON HEALTH CARE FIELDS: Percentage distribution of health majors at community colleges and for-profit
subbaccalaureate institutions by type of institution: 2007-08
Percent
100
80
60
40
20
!
0 '
57
40
11 10 10 10
14
5
26
m:;.,
22
__ _
Nursing Health
professions
and related
clinical sciences
:.J For-profit less-than-2-year
Health and
medical
administrative
services
li!l For-profit 2-year
Allied health
and medical
assisting
services
Other
health care fields'
Community colleges
1
Other health care fields include allied health diagnostic, intervention, and treatment fields, pharmacy and pharmaceutical sciences, and health
aides, among others.
NOTE: Health care fields accounted for 43 percent of majors at for-profit less-than-2-year institutions, 46 percent at for-profit 2-year institutions,
and 20 percent of majors at community colleges. Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states,
DC, and Puerto Rico. Standard error tables are available at http://nces.ed.gov/das/librarv/reports.asp .
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007--Q8 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 11. MAIN REASON FOR ENROLLI NG: Percentage distribution of undergraduates who attended community colleges and
2-year or less-than-2-year for-profit institutions, by their main reason for enrolling and type of institution: 2007-08
Percent
100
80
60
40
20
0
82
51
Acquire job-related
skills or credentials
t Not applicable.
12 12
t
Complete
associate's
degree
n For-profit less-than-2-year
1
From the institution in which they were primarily enrolled.
13
Solely for
personal interest
o For-profit 2-year
25
Prepare
to transfer
Community colleges
5
Earn course
credits at a
different
institution'
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Detail may not sum
to totals because of rounding. Standard error tables are available at http://nces.ed.govfdasllibrary/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-QS National Postsecondary Student Aid Study
(NPSAS:08).
Figure 12. COMBINING WORK AND STUDY: Percent age of all undergraduates who worked and worked full time whil e enrolled and f or
those who worked, reported their main role as being a student, by type of institution: 2007-08
Percent
100
80
60
40
20
0
81
73
69
Among all undergraduates, worked
while enrolled
For-profit
39 41
Among all undergraduates,
worked full time while enrolled
79
58 60
Of those who worked,
primary role as a student'
Community
colleges
o Public and private
nonprofit 4-year
1
Undergraduates who worked while enrolled were asked whether they regarded their primary role as a student or as an employee.
2
1ncludes 4-. 2-. and less-than-2-year institutions.
NOTE: Excludes work-study and assistantships. Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states,
DC, and Puerto Rico. Standard error tables are avai.lable at http://nces.ed.gov/ das/librarv/reports.asp .
SOURCE: U.S. Department of Education. National Center for Education Statistics, 2007--QS National Postsecondary Student Aid Study
(NPSAS:08) .
Table 1. DEGREE OR CERTIFICATE PROGRAM: Percentage distribution of undergraduates attending community colleges, for-profit,
and traditional 4-year institutions by program type: 2007-08
Associate' s Bachelor's No degree or
Program type Certificate degree degree certificate program
Public and private nonprofit 4-year 0.6 4.2 93.1
Community college 6.9 78.9 3.0
1
For-profit 34.4 38.2 25.8
Less-than-2-year 98.4
t t
2-year 35.7 61.8 0.5
4-year 5.6 42.6 50.4
t Not applicable.
1
Students who attended a community college and reported that they were working on a bachelor's degree may have attended a community
college that offers bachelor's degrees or planned to transfer to a 4-year institution.
2.1
11.2
1.6
L6
L9
1.4
NOTE: Detail may not sum to totals because of rounding. Estimates include students enrolled in Title IV eligible postsecondary institutions in the
50 states, DC, and Puerto Rico. Standard error tables are available at http:/lnces.ed.gov/das/librarv/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Postsecondary Student Aid Study
(NPSAS:OS).
Table 2. UNDERGRADUATE AWARDS CONFERRED in Title IV institutions, by type of institution: 2007
Institution type
Total
Type of institution
Total
awards
100.0
Traditional 4-year colleges and universities 53.6
Public 4-year or above 35.6
Private nonprofit 4-year and above 18.0
Public 2-year (community colleges) 28.2
Public less-than-2cyear 1 .0
Private nonprofit 2-year 0.5
Private nonprofit less-than-2-year 0.3
Private for-profit 16.4
4-year or above 4.8
2-year 5.5
Less-than-2-year 6.1
t Not applicable.
NOTE: Detail may not sum to totals because of rounding.
Percentage distribution of
undergraduate awards conferred in 2007
Associate's
Ceritifcates degree
100.0 100.0
5.1 14.2
2.9 9.2
2.2 5.0
46.2 69.1
4.3
t
1.1 0.8
1.3
t
41 .9 15.8
2.1 7.8
14.6 8.0
25.2 t
Bachelor's
degree
100.0
95.4
63.7
31 .7
t
t
t
t
4.6
4.6
t
t
SOURCE: U.S. Department of Education, National Center for Education Statistics, Integrated Postsecondary Education Data System (I PEDS),
"Completions Survey" and "Institutional Characteristics Survey," 2007.
Table A. Number of for-profit Title IV institutions, by level: United States, academic year
2004-05 to 2008-09.
-08
Total institutions, all controls 6383
6,441 6,536 6,551
Total for-profit institutions

Institutional Level
4-year 370 407 453 490 530
793 816 844 857 893
Less-than 2-vear 1318 1,34o 1,385 1,403
Change
3.9%
13.9%

1!l.6%
o.4%
SOURCE: U.S. Department of Education, National Center for Education Statistics, Integrated
Postsecondary Education Data System (!PEDS), Fall 2004to Fall 2008, Institutional
Characteristics Component
Table B. Fall
Enrollment by
institution control :
United States, Fall

Private For Profit
Private Nonprofit
Public
Total
SOURCE: U.S. Department
of Education, National
Center for Education
Statistics, Integrated
Postsecondary Education
Data System, Institutional
Characteristics and Fall
Enrollment Surveys: 2004
2008.
2008
6n,851 1,188,881 1,380,355 1, 797,563
3,137, 108 3,299,094 3,440,559 3,543,455 3,684,n3
11,891,450 12,883,071 13,081,358
15,701,409 17,710,798 19,574,395
Percent
Change

7.1%
7.7%
10. 5%
Table C. Percent
of Fall
Enrollment by
institution
control: United
States, Fall 2004

Private For Profit
Private Nonprofit
Public
SOURCE: Integrated
Postsecondary Education
Data System, Institutional
Characteristics and FaiJ
8nrollment Surveys: 2004
2008.

4.8%

75.7%

5.0%
19.4%
75.6%

6.7%
19.4%
73.9%
Percent
Change
7.6% 87%
19.5% 18.8% 8%
73.0% n.o% 3%
Table D. Number and percentage distribution of awards conferred at for-profit Title IV institutions, by level:
United States, academic year 2003-04 to 2007-08.
2003-04 2004-05 2005-06 2006-07 2007-08 Change
Total
All institutions2
2,998,114 3,085,464 3,165,016 3,232,300 3,314,978
10.6%
For-profit institutions
Number of awards2
356,128 382,146 405,542 425,377 444,307
24.8%
Percent of all awards2
11.9% 12.4% 12.8% 13.2% 13.4% 12.8%
4-year institutions}
All institutions2
2,213,225
2,293,350 2,371,219 2,443,619 2,521,319 13.9%
For-profit institutions
Number of awards2
111,586
134,081 161, 160 182,445 202,410 81.4%
Percent of all awards2 5.0% 5.8% 6.8% 7.5% 8.0% 59.2%
2-year institutionsl
All institutions2
542,177
557,172 564,964 563,875 571,964 5.5%
For-profit institutions
Number of awards2
55,080
55,259 57,461 59,381 58,415 6.1%
Percent of all awards2 10.2% 9.9% 10.2% 10.5% 10.2% 0.5%
Less-than 2-year institutions3
All institutions4
242,712
234,942 228,833 224,806 221,695 -8.7%
For-profit institutions
Number of awards4
189,462
192,806 186,921 183,551 183,482 -3.2%
Percent of all awards4 7R1% 82.1% 81.7% 81.6% 82.8% 6.0%
1 Includes degree-granting
institutions only
2 Excluding certificates
3 Includes all institutions, both
degree and non-degree granting
4 Includes all awards, including
both degrees and certificates
SOURCE: U.S. Department of Education, National Center for Education Statistics, Integrated Postsecondary
Education Data System (!PEDS), Fall 2003 to Fall 2008, Completions component.
1 The Expansion of Private Loans in
2 Postsecondary Education
3 Government officials and higher education associations are con-
4 cerned that, out of confusion or ignorance, some postsecondaty
5 students may not cake full advantage of federal loan programs and
6 turn instead to more costly private loans (e.g., Federal Trade
7 Commission 2008; King 2007). Others argue that federal loan lim-
8 its don't meet some students' education financing needs and that
9 such students, therefore, seek additional funds through private
10 loans (McSwain, Price, and Cunningham 2006).
Most students borrow money for postsecondary education through
federal loan programs, which have eligibility requirements and bor-
rowing limits. Students can also obtain private loans from banks
and other lending institutions. Private (or "alternative") loans are
not guaranteed by the government and can be relatively expensive,
as they generally have higher fees and interest rates than federal
student loans.
Private education loans are estimated to have reached a peak of
about $22 billion in 2007- 08 (College Board 2009). That same
year, many lenders increased their direct marketing to students,
highlighting a quick and easy application and approval process for
private loans; some of these lenders were accused of deceptive mar-
keting practices (U.S. Senate Committee 2007). According to the
U.S. Department of Education (2008), "[p]rivate loans and credit
cards are consumer loans and are vety expensive ways of financing
your education."
Since 2007-08, however, the volume of private loans for postse-
condaty education is thought to have declined substantially due to
a shortage of capital and higher underwriting standards by lenders
(Student Lending Analytics 2009a, 2009b). Recent data released by
I 2 I THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION
1 the College Board (2009) are consistent with this expla-
2 nation.
3 This Statistics in Brief examines how the use of private
4 education loans changed between 2003-04 and 2007-
5 08 by addressing the following study questions:
6 1. How did undergraduate borrowing from private
7 sources change between 2003-04 and 2007-08?
8 2. Who obtained private loans?
9 3. To what extent did undergraduates combine
10 private and public loans?
11 4. Did undergraduates borrow the maximum
12 amount from federal Stafford loans before turn-
13 ing to private loans?
14 5. How did private borrowing change among
15 graduate and professional students?
16 Key Findings
17 The percentage of undergraduates obtaining private
18 loans nearly tripled, from 5 to 14 percent.
19 Undergraduates at for-profit institutions had the
20
21
22
23
24
25
highest rate of borrowing from private sources-42
percent took out private loans in 2007- 08.
Dependent undergraduates from middle-income
families borrowed from private sources at higher
rates than did students from low- or high-income
families.
26 About one-half of full-time, full-year undergraduates
27 who obtained a private loan had borrowed the max-
28 imum federal (Stafford) loan amount.
29
30
MAJOR TYPES OF
HIGHER EDUCATION LOANS
Private loans. Private loans are education loans,
not guaranteed by the federal government, from
commercial lenders, credit unions, or other non-
profit entities. Their terms are determined by the
lender. Private loans carry a market interest rate,
usually variable and based on credit history, and
they generally have higher fees and interest rates
than federal student loans.(See question on p. 8)
Stafford loans. These student loans have fixed
interest rates and various repayment benefits and
are guaranteed by the federal government. They
have eligibility requirements and limits on amounts.
There are two types of federal Stafford Loans: sub-
sidized and unsubsidized. Subsidized Stafford
loans are awarded based on financial need, and
the federal government pays interest on the loan
until the student begins repayment and during au-
thorized periods of deferment thereafter. Unsubsi-
dized Stafford loans are not need based; students
are charged interest for the duration of the loan,
although it can be capitalized.
Parent PLUS loans. These federally guaranteed
loans are available only to the parents of depen-
dent students. The interest rate in 2007- 08 was
fixed at ei ther 7.9 or 8.5 percent. Borrowers cannot
have a negative credit history, and the amount is
limited to the cost of attendance minus other finan-
cial aid. The loans carry the benefits and protec-
tions of all federal loans.
Graduate PLUS loans. These are federally guar-
anteed loans for graduate and professional stu-
dents that became available in 2006. The terms are
the same as for Parent PLUS loans, with the same
interest rates, restrictions, and benefits.
THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION I 3
1 Detailed Findings 28 tutions was $5,500 in 2007-08; at private nonprofit 4-
2 How did undergraduate borrowing from private
29 year institutions, it was $17,800; and at for-profit insti-
3 sources change between 2003-Q4 and 2007-QS?
30 tutions, it was $10,200. Further, the for-profit sector is
3 1 more likely to enroll low-income students. The median
4 T he rate of undergraduate private loan borrowing in- 32 income for dependent students' parents at public 4-year
5 creased from 5 to 14 percent between 2003-04 and 33 institutions was $75,700 in 2007-08.
4
At for-profit in-
6 2007- 08 (figure 1).
1
But the amount borrowed held 34 stitutions, the corresponding median income was
7 steady after adj usting for inflation: the average private 35 $35,700.
8 loan in 2003-04 was $6,600 and $6,500 in 2007-08.
2
36 I n general, the higher the tuition, the higher the rate of
9 Stafford loan borrowing among undergraduates also in- 37 private borrowing. For example, the highest rates of bor-
1 0 creased from 32 percent to 35 percent, and the average 38 rowing (30- 32 percent) occurred among students whose
11 Stafford loan amount rose from $4,900 to $5,000. 39 tuition was more than $10,000 per year (figure 3) . In
12 The rate of any borrowing rose from 34 percent in
13 2003- 04 to 39 percent in 2007- 08.
3
T he average loan
14 amount from all sources, including Parent PLUS loans,
15 increased from $6,900 to $8, 100.
40 comparison, 22 percent of students paying $5,000 to
41 $9,999 in tuition took out private loans, as did 14 per-
42 cent of those paying $3,000 to $4,999 and 9 percent or
43 less of those paying under $3,000 in tuition.
16 Who obtained private loans?
44 Dependent undergraduates from middle-income fami li es
45 attending full-time full-year borrowed from private
17 In 2007-08, private loan borrowing varied by type of
46 sources at higher rates than students from low- or high-
18
. . . . . I 1 d d d d
4
7
tnstttuuon, tuttton eve , stu ent epen ency status, an
19 family income. T he rate of private borrowing was high-
4 8
20 est at for-profit institutions, where the rate about

2 1 from 13 percent to 42 percent, from 2003-04 to 2007-
50
22 08 (figure 2). Private borrowing at private nonprofit 4-
51
23 year institutions was also high and about doubled, from
52
24 11 percent to 25 percent. 53
income famil ies. One-fifth of studenrs in the two mid-
dle-income groups took out private loans, compared
with 15-16 percent of students in the low- and high-
income groups (figure 4). Looking at aU borrowing, in-
cluding Parent PLUS loans, dependent students from
high-income fan1Uies borrowed at the lowest rate (40
percent).
54 Among independent students, upper middle- and high-
25 T he higher rate of borrowing among students at private 55 income students took out private loans relatively more
26 institutions reflects to some extent the higher tuition at 56 often than low-income students. One-fifth of low-
27 these institutions. Average tuition at public 4-year insti- 57
income independent undergraduates took out private
58 loans, compared with 27- 28 percent of those in the up-
1 All comparisons of estimates were rested for statistical signi- 5
9
per middle- and high-income groups.
ftcance usi ng the Student's t-statistic, and all differences cited
are statistically significant at the p < .05 level. No adj ustments
for multiple comparisons were made. The standard errors for
the estimates can be found at
http://nces.ed.gov/das/li brary/reporrs.asp.
2
All dollar amounts for 2003-04 have been adj usted for infla-
tion to 2007 doll ars using the Consumer Price Index.
3
Includes Parent PLUS loans as well as Stafford, Perkins, and
private loans.
4
Independent students are age 24 or older and students under
24 who are married, have dependents, are veterans, or are an
orphan or ward of the courts. Other undergraduates under age
24 are considered to be dependent. Income for dependent
students is the income of their parents. For independent stu-
dents, income includes a spouse' s income if the st udent is mar-
ried.
I 4 I THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION
1 To what extent did undergraduates combine pri-
2 vate and public loans?
37 Did undergraduates borrow the maximum amount
38 from federal Stafford loans before turning to pri-
39 vate loans?
3 Most undergraduates who borrowed did so through a
4 federally guaranteed loan program. Some 63 percent of 40
5 undergraduates who borrowed obtained loans from pubA 1
6 lie,
5
mostly federal, sources exclusively, and another 27 42
7 percent borrowed from both public and private sources 43
8 (figme 5) . The remaining 9 percent borrowed only from44
9 private sources. Students borrowed from public and pri- 45
10 vate sources at different rates, however, depending on 46
In 2007- 08, one-half of full-time, full-year undergra-
duates exhausted their annual Stafford loan eligibility
before taking out private loans (figure 6). Independent
private loan borrowers exhausted Stafford el igibility at a
lower rate than did dependent borrowers.
Policymakers are concerned that some students seek pri-
vate loans because they are unaware of the advantages of
11 the types of institutions attended.
47 federal loan programs. They are also concerned that
48 some students may not borrow the maximum Stafford
12 Undergraduates in public 2-year institutions took out
49 loan amount before turning to private loans (U.S. De-
13 exclusively private loans at higher rates than those in oth-
50 partment of Education 2008). Applying for federal aid is
14 er types of institutions. At community colleges, 21 per-
51 necessaty to obtain federal loans. Consequently, it is use-
15 cent of borrowers took out only private loans, compared 52 ful to examine whether students did apply for federal aid
16 with 9 percent among students in all institutions. This
53 when comparing public and private borrowing.
6
17 higher rate of private borrowing, however, occurred with-
18 in the context of lower rates of borrowing overall. In 54 Among full-time, full-year undergraduates who took out
19 2007-08, some 13 percent of all community college stu-55 private loans, some 19 percent did not obtain a Stafford
20 dents borrowed from any SOLtrce, representing the lowest56 loan, and about half these students (10 percent of all
21 rate among students in all types of institutions (Wei et al57 private borrowers) did not apply for financial aid. 7
22 2009a, table 13). Some policy analysts believe that many53 Another 31 percent took out Stafford loans, but appear
23 community colleges do not encourage borrowing federal 59 to have borrowed less than the ma:ximwn amount. s The
24 loans (Cochrane, Shireman, and Sivashankaran 2008). 60 remaining 50 percent had taken out the maximum
25 The largest proportion of borrowers who took out pri-
26 vate loans either exclusively or in combination with pub-
27 lie loans (42 percent) was found among those enrolled at
28 for-profit institutions (figure 2), and those enrolled at
29 institutions with programs of 2 or more years had the
30 highest percentage who took out both private and public
31 loans (figure 5). Among those enrolled at for-profit insti-
32 tutions with programs taking 2 or more years to com-
33 plete, 45 percent of undergraduate borrowers took out
34 both private and public loans, while the overall percen-
35 cage of undergraduates who cook out both types of loans
36 at all institutions was 27 percent.
5
"Public loans" includes Stafford loans, other federal loans
(e.g., Perkins and PLUS), state, and institutional loans.
6
Foreign students and those attending part time or part yea.r
were excluded from this part of the analysis because they either
are ineligible for federal loans or qualifY for very little. Other
reasons that students cannot borrow from federal loan pro-
grams couJd not be identifled from the NPSAS data.
7
Some may have taken out other loans, such as Perkins, state,
or institutional loans, but those are comparatively rare.
8
The maximum Stafford amomu is limited for those attend-
ing part time or for less than a fttll year, so this analysis of
maximizing Stafford loans was limited to fuJI-time, full-year
students. There are other restrictions on Stafford borrowing
based on program or budget that were not taken into account.
Because of these restrictions, some borrowers who appear to be
borrowing below the maximum amount are in fact borrowing
at their maximum. Therefore, our estimates of the percentage
of borrowers who are borrowing at less than the maximum
couJd be overstated.
THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION I 5 I
1 amount allowed under the Stafford loan program. These24 points,
9
compared with an increase of9 percentage
2 students sought more money than was available from 25 points among undergraduates (figure 1 ).
3 federal programs to pay for their education expenses.
26 Graduate students differed significantly in borrowing
4 Undergraduate students have different annual Stafford 27 rates depending on the degree program in which they
5 loan limits based upon both. dependency status and class28 were enrolled (figure 7). Master's degree students' rate of
6 level. For example, for first-year undergraduates, the lim-29 total borrowing increased from 38 percent to 44 percent
7 its were $3,500 for dependent students and $7,500 for 30 over this period. The rate of private borrowing among
8 independent students. In 2007- 08, about one-third (35 31 master's degree students doubled between 2003- 04 and
9 percent) offull-time, full-year independent undergra- 32 2007-08 (from 6 percent to 12 percent), while their rate
10 duates who took out private loans also borrowed the 33 of Stafford borrowing increased from 36 percent to 39
11 maximum Stafford loan amount, compared with. 56 per-34 percent.
12 cent of dependent undergraduates who took out the
13 maximum Stafford loan amount (figure 6) . 35 Among doctoral students, changes in private, Stafford,
14
Annual Undergraduate
Stafford Loan Limits
(for loans taken out between July 1, 2007
and June 30, 2008)
36 or total borrowing were not evident. Between 5 and 7
7 percent took out private loans in both 2003- 04 and
8 2007-08.
9 Students in first-professional programs had the highest
f0 rate of private borrowing, but the proportion with pri-
Dependent Independent
Academic year student student i 1 vate loans declined over time, while borrowing from fed-
------''----------------- i2 eral sources increased.
10
In both years, students in frrst-
First year $3,500 $7,500 .
Second year 4,500 8,500 !3 professional programs borrowed from all sources at high-
Third and remaining years 5,500 10,500 i4 er rates than students in other degree progran1s. In
NOTE: Aside from dependency status and year in school, the i5 2003-04, about one-fourth (23 percent) of students in
amount a student can borrow under the Stafford loan program
can be further reduced depending on the cost of attendance, the
student's expected family contribution, attendance status,
whether the program is less than a year long, and how much
other financial aid is received.
SOURCE: U.S. Department of Education. (2007). 2007-oB
Federal Student Aid Handbook. Retrieved April 23, 2010, from
http://ifap.ed.gov/ifaplbyAwardYear.jsp?tyoe-fsahandbook&awa
i6 first-professional programs took out private loans, com-
i7 pared with 5 to 6 percent of master's and doctoral stu-
iS dents. By 2007-08, however, the rate of private
i9 borrowing among first-professional students had
;o dropped to 16 percent, while their rate of Stafford bor-
_::rd;:v::ea;:r;:=2::00=:
7
- ::2:;00:;8::,. _______________ , ; 1 rowing increased from 69 to 76 percent and one-fourth
15 How did private borrowing change among gradu-
52 (25 percent) took out Graduate PLUS loans.
16 ate and first-professional students? 53
54
17 ln 2006, graduate students became eligible for the feder-
55
18 al Graduate PLUS loan program. This gave graduate 56
19 students another source ofloans from a government-
20 guaranteed progran1 with competitive, ftxed-interest
21
22
23
rates, eliminating some of the need for private loans.
Between 2003- 04 and 2007- 08, the rate at which grad-
uate students took out private loans rose 4 percentage
Between 2003- 04 and 2007- 08, after adjusting for in-
flation, overall average loan amounts for all graduate and
first-professional students increased from $17,700 to
$18,500, largely due to the newly established Graduate
9
NPSA$:04 and NPSA$:08 Data Analysis System; data not
shown in figures or tables.
1
First professional programs include dentistry, medicine,
optometry, osteopathic medicine, pharmacy, podiatric medi-
cine, veterinaty medicine, chiropractic, law, and theology.
I 6 I THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION
1 PLUS loans. The average private loan amount for gradu-37
2 ate students decreased from $ l 0.500 to $8,400, and the
38
3 average Stafford loan amount decreased from $16, 100 to
39
Technical Notes
Dara in this report come from the National Postsecon-
dary Student Aid Study (NPSAS) administered for the
4 $15,600.
40 years 2003-04 and 2007-08. Conducted every 4 years,
5 Find Out More
4 1 NPSAS is based on a nationally representative sample of
42 al l students enroll ed in Title N -eligible institutions in
6 More detailed information on 2007-08 undergraduates 43 the United States between J uly 1 and J une 30 of a given
7 and graduate students enrolled in U.S. postsecondaty 44 academic year. The sampling design for coll ecting data
8 institutions can be found in Web Tables produced by 45 consists of two stages. The first stage involved selecting
9 the National Center for Education Statistics (NC.ES) 46 eligible institutions from which students would then be
10 using the 2007- 08 National Postsecondaty Student Aid 47 sampled in the second stage. In 2007- 08, approximately
11 Study (NPSAS:08) data. These Web Tables are a com- 48 114,000 undergraduate students were study respondents.
12 prehensive source of information on students during the49 Estimates were weighted to adjust for the unequal prob-
13 2007- 08 academic year. Included are estimates of de- 50 abili ty of selection into the sample and for nonresponse.
14 mographi cs, enroll ment, and employment characteris-
15 tics. Web Tables docw
11
enting how students pay for 51 NCES Statistical Standard 4-4-1 states that "[a]ny survey
16 their undergraduate education are also available. 52 stage of data collection wim a unit or item response rate
53 less than 85 percent must be evaluated for the potential
17 Web Tables-Profile of Undergraduate Students in US. Postse- 54 magnitude of nonresponse bias before the data or any
18 condary !mtitutiom: 2007-08 (NCES 20 I 0-205). [link 55 analysis using me data may be released" (U.S. Depart-
19 will be added)
56 ment of Education 2003). The only variable included in
20 Web Tables-Student Financing of Undergraduate Education: 57 this Bri ef that has a response rate of less than 85 percent
2 1 2007-08 (NCES 2010-162) . [link will be added) 58 is TOTLOAN.
22 Web Tables-Profile of Students in Graduate andFirst-
23 Professional Education: 2007-08 (NCES 20 10-177). [link59
For an overview of the survey memodology and a discus-
24 wil l be added] 60
sion of nonresponse bias, see appendix B of the report
6 1 2007- 08 National Postsecondary Student Aid Study
25 Readers may also be interested in the following NCES
26 products related to me topic of mis Statistics in Brief
62
63
27 2007-08 National Postsecondary Student Aid Study
28 (NPSAS:08): Student Finandal Aid Estimates: First Look 64
29 (NCES 2009-166).
30 hrrp:l/nces.ed .gov/pubs2009/2009l66.pdf
3 1 Undergraduate Financial Aid Estimates by Type of Imtitution in
32 2007-08 (NCES 2009-201).
33
34
35
36
h ttp:l/nces.ed.g,ov/pubs2009/2009201.pdf
Federal Loam in 2007-08: Undergraduate Cumulative Debt
and Students Who Bonow at the Maximum (NCES 2010-
151). [link to be added)
(NPSAS:OB): Student Financial Aid Estimates for 2007-08:
First Look (http:/ / nces.ed. gov/pubs2009/2009166.pdt).
THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION
1 References
39
40
41
2 The College Board. (2009). Trends in Student Aid. Table l.
3
4
5
Retrieved October 28, 2009, from
http://www.rrends-collegeboard.com/student aid/pdf/
2009 Trends Smdent Ald.pdf.
42
43
44
6 Cochrane, D. F., Shireman, R., and Sivashankaran, S. (2008,
7
8
9
10
11
12
13
14
15
16
17
18
19
20
2 1
22
April). Denied: Community College Students Lack Access to 45
A/fordable Loam. The Project on Smdent Debt. Retrieved46
September 9, 2009, from 47
48
http://ticas.org/ flies/pub/ denied .pdf.
Federal Trade Commission. (2008, June). Student Loam:
49
50
Avoiding Deceptive Offers. FTC Facts for Consumers. 51
Federal Trade Commission and U.S. Department of Edu-
cation. Retrieved September 9, 2009, from
h rrp://sruden taid.ed.gov/smden ts/auach menrs/
siteresources/loa nsA voidDecep. pdf.
King, J.. (2007). Who Borrows Private Loam? ACE Issue
52
53
54
55
56
Brief. Washington, DC: American Council on Education.
Retrieved October 28, 2008, fiom 57
http://www.acenet.edu/AJ\If/Template.cfm?Section-Searc58
h&seqion-issue briefs&templare-/CM/ContenrDisplay.59
cfm&ConrenrFileiD- 3445. 60
6 1
23 McSwain, C., Price, D., and Cwutingham, A. (2006). The 62
24 Future of Private Loam: Who Is Bonowing, and Why?
25 Washington, DC: lnstitute for Higher Education Policy. 63
64
26 Student Lending Analytics. (2009a, July 23). The Incredible 65
27 Shrinking Private Student Loan Mmket: Outlook for 2009"'66
28 10. Slog. Retrieved September 9, 2009, from 67
29 h up://studen tlendi nganalytic.qypepad.com/studen t lend68
30 ing analytics/2009/07/the-i ncredible-shrinking-private- 69
31 loan-market.html.
70
32 Student Lending Analytics. (2009b, August 17). Bank Lending
33 Standards May Remain Tight Through Second Half of
34 2010. Blog. Retrieved September 9, 2009, from
3 5 http://srudenrlendinganalytics. typepad.corn/smdent lend
36 ing analytics/2009/08/bank-lending-standards-may-
37 remain-tight-through-second-half-of-20 l O.htrnl.
38
U.S. Department of Education, Federal Student Aid, St udents
Charmel. (2008). Your Federal Student Loans: Learn the
Basics and Manage Your Debt. Washington, DC: Author.
U.S. Department of Education, National Center for Educa-
tion Statistics. (2003). NCES StatisticaL Standards (NCES
2003-601). Washington, DC.
U.S. Senate Committee on Banking, Housing and Urban Af-
fairs. (2007, J tme 6). Hearing on "Paying for College:
The Role of Private Student Lending." Washi ngton, DC.
Retrieved November l, 2009, from
h rtp://banki ng.senate.gov/ pub! ic/i ndex .cfm?Fuseaction-
Hearings.Hearing&Hearing ID-8c46al82-alff-4bc5-
af53-e0982b49527f.
Wei, C. C. (forthcoming). What is the Real Price of College?
Sticker, Net, and Out-of-Pocket Prices in 2007-08
(NCES 2010-175). National Center for Education Statis-
tics, Institute of Education Sciences, U.S. Department of
Education. Washington, DC.
Wei, C.C., Berkner, L. , He, S., ar1d Lew, S. (2009a). Web-
Tables: Undergraduate Financial Aid Estimates by Type of
Institution in 2007-08 (NCES 2009-201). National Cen-
ter for Education Statistics, Institute of Education
Sciences, U.S. Deparunenr of Education. Washington,
DC.
Wei, C. C., Berkner, L., He, S., Lew. S., Cominole, M., and
Siegel, P. (2009b). 2007-08 NationaL Postsecondmy Stu-
dent Aid Study (NPSAS:08): Student Financial Aid Esti-
mates for 2007-08 (flrst Look) (NCES 2009-166).
National Center for Education Statistics, Institute of
Education Sciences, U.S. Department of education.
Washington, DC.
I 8 I THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION
1 Run Your Own Analysis in Datalab 38 "During the 2007-2008 school year, did you take out
39
2 Direct access to the data used in this Statistics in Brief is
40
3 available through the NCES DataLab
any private or alternative loans from a financial institu-
tion? Some examples of commonly used private loans
include: 41
4 (http://nces.ed.gov/datalab/). You can generate your own
5 tables using one of three programs: PowerS tats, Quick- 42 Sallie Mae Signature Student Loan
6 Stats, or the Data Analysis System (DAS). 43 CitiAssist Undergraduate and Graduate Loan
7 All estimates presented in this Brief were produced using44 Chase Education One Private Student Loan
8 the DAS, a web-based software application that enables 45 Nellie Mae EXCEL Loan
9 users to generate tables for most of the postsecondary
46 Access Group Loans
10 surveys conducted by NCES. These figures were gener-
11 ated using the DAS, which contains a detailed descrip- 47 (Keep in mind that many lenders that offer private loans
12 tion of how each variable was created and includes the 48 might also offer federal Stafford, Graduate/Professional
13 wording of questions for variables coming directly from 49 PLUS loans, and Parent PLUS loans. For this question
14 the interview. 50 we want to know about private or alternative loans on-
51 ly.)"
15 You can replicate or expand upon the figures and tables
16 presented in this report by using DataLab. All variables 52 Package of private and nonprivate loans (PRJVPACK):
17 used to generate the figures and ta.bles in this Brief can
18 be found in the NPSAS:04 and NPSAS:08 DAS. More
19 information and descriptions of each variable can be
20 found at http://nces.ed.gov/DAS.
21 The variables used in this report are as follows:
53 Indicates the loan package by whether the loan received
54 was private (alternative) or not during the 2007- 08 aca-
55 demic year. For students with any loan (TOTLOAN2 >
56 0), indicates whether a borrower had only private oral-
57 ternative loans (PRIVLOAN), only nonprivate loans
58 consisting of federal loans (TFEDLN2, including
59 PLUS), state loans (STLNAMT), or institutional loans
60 (INLNAMT), or both private and nonprivate.
22 Private (alternative) loans (PRJVLOAN): Indicates the
23 amount of private or alternative loans received by stu-
24 dents for the 2003- 04 and 2007- 08 academic years.
61 Stafford total maximum (STAFCT3): Classifies the total
25 These are education loans from commercial lenders that 62 Stafford Joan amount received in 2007-08
26 are not guaJanteed by the federal government and that 63 (STAFFAMT) into categories based on the maximum
27 cany market interest rates based on credit scores. Exam- 64 Joan limits for subsidized and unsubsidized Stafford
28 pies of such loans are the Sallie Mae Signature Student 65
29
30
loans combined and includes a category for those who
didn't apply for federal aid. The normal maximum loan
an1ounts in 2007-08 for undergraduates were deter-
Loan, CitiAssist Loan, or Chase Education One Private 66
Student Loan. Private loans differ from Stafford, Parent 67
31 PLUS, Perkins, and Graduate PLUS loans, which are
32
33
68 mined by the student's undergraduate class level and
guaranteed by the federal government; private loans, 69 dependency status. This variable does not take into ac-
however, can be offered by the same lenders or by states 70 count further borrowing limits that might be imposed,
34 or other nonprofit institutions. Institutions do notal- 71 e.g., Stafford loan amounts cannot exceed a borrower's
35 ways have information on students' private or price of attendance, and subsidized Stafford loans cannot
36 loans, so this information came primarily from student 73 exceed a borrower's need amount.
37 interviews, in which students were asked the following:
1 Other Variables
2 Attendance pattern (A ITNST AT)
3 Citizenship status (CITIZEN2)
4 NPSAS institution control (CONTROL)
5 Dependency status (DEPEND)
6 Parents' income (DEPlNC)
7 Independent student's income (INDEPINC)
8 Applied for federal aid (FEDAPP)
9 Graduate PLUS loan total (GPLUSAMT)
10 Graduate degree program (GRADDEG)
11 Total federal Parent PLUS loans (PLUSAMT)
12 Type of institution (AID SECT)
13 Institution sector (SECI'OR4)
14 Federal subsidized loans (except PLUS) (SUBLOAN)
15 Total student loans from all sources (TOTLOAN)
16 Institution tuition and fees (TUITION2)
17
THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION I 9 I
www.ed.gov ies.ed.gov
Figure 1. Percentage of undergraduates who borrowed, by type of loan: 2003-04 and 2007-08
Percent
100
80
60
40
32
35
20
14
5
0
Private Stafford
2003-04 2007-08
34
39
All loans
Average undergraduate loan amounts,
by type of loan:
2003--04 and 2007-QS
2003--04 2007-QS
Private loans $6,600 $6,500
Stafford loans 4,900 5,000
All loans including PLUS 6,900 8,100
NOTE: Amounts for 2003-04 have been adjusted for inflation
using the Consumer Price Index for urban households (CPI
U). Arrounts are averages for those who received the
specified type of aid.
NOTE: Private loans are education loans from commercial lenders which are not guaranteed by t he federal government and carry market
interest rates.
SOURCE: U.S. Department of Education, Nationa.l Center for Education Statistics, 2003-04 and 2007-08 National Postsecondary Student Aid
Studies (NPSAS:04 and NPSAS:OS) .
Figure 2. Percentage of undergraduates who took out private loans, by institution type: 2003-04 and 2007-08
Percent
100
80
60
40
20
0
Public
2-year
14
Public
4-year
2003-04
25
11
Private
nonprofit
4-year
2007-08
42
13
For-profit
Average tuition and fees paid by
undergraduates, by institution type:
2007-QS
(includes full-time and part-time students)
Public 2-year
Public 4-year
Private nonprofit 4-year
For-profit
$1,200
5,500
17,800
10,200
Median income, by institution type: 2007-QB
Dependent lndepen-
students dent
parents students'
Public 2-year $54,200 $29,400
Public 4-year 75,700 22,000
Private nonprom 4-year 84,500 30,300
For-profit 35,700 19,500
1
Independent students are age 24 or older and 1udents under
24 who are married, have dependents, are veterans, or are an
orphan or ward of the oourts. Other undergraduates under age
24 are considered to be dependent. Income for dependent
students is the inoome of their parents. For independent
students. income includes a spouse's income if the student is
married.
NOTE: Private loans are education loans from commercial lenders which are not guaranteed by t he federal government and carry market
interest rates. Students who attended more than one inst itution or any type of institution not listed in the figure were not included due to the small
number of cases.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2003-04 and 2007-08 National Postsecondary Student Aid
St udies (NPSAS:04 and NPSAS:08).
Figure 3. Percentage of undergraduates who took out private loans, by
tuition amount: 2007-os
$15,000 or more
$10,000-$14,999
$5,000-$9,999
$3,000-$4,999 14
$1 ,500-$2,999
$1 ,499 or less
0 20
32
30
22
40 60
Percent
80
NOTE: Private loans are education loans from commercial lenders which
100
are not guaranteed by the federal government and carry market interest rates.
SOURCE: U.S. Department of Education. National Center for Education
Statistics, 2007- 08 National Postsecondary Student Aid St udy (NPSAS:08).
Figure 4. Percentage of fulltime, full-year undergraduates who took out private loans and any loans including Parent PLUS, by
dependency status and family income level: 2007-QS
Percent
100
80
60
54
56
61
66
69
52
40
40
27
28
21
20
0
60
Low-income Lower
middle-
income
Upper High-income
middle-
Low-income Lower
middle-
income
Upper High-income
middle-
income income
Dependent Independent
Private loans All borrowing including PLUS
NOTE: Family income categories were based upon the annual income distribution for 2006, from the lowest to the highest quartile ranges.
Among dependent students, low-income was less than $36,100: lower middle-income was $36.100-$66,600: upper middle-income was
$66,600-$104,600: and high-income was $104,600 or more. Among independent students, low-income was less than $11 ,000; lower middle-
income was $11,000-$26,000; upper middle-income was $26,000- $48,400; and high-income was $48,400 or more. Private loans are education
loans from commercial lenders which are not guaranteed by the federal government and carry market interest rates.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-()8 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 5. Percentage distribution of undergraduate borrowers, by borrowing package and institution type: 2007-QS
Percent
100
..

2
Exclusively
80 20
private
27
35 45 loans
60 Both
programs
40
Exclusively
20
public
loans
0
Total Public Public Private For-profit For-profit
2-year 4-year nonprofit less-than- 2 years
4-year 2-year or more
NOTE: Public loans includes Stafford loans, other federal loans (e.g., Perkins and PLUS), state, and institutional loans. Private loans are
education loans from commercial lenders which are not guaranteed by the federal government and carry market interest rates. Total estimates
also include those attending public less-than2-year institutions, those attending private nonprofit less-than-4-year institutions, and those
attending more than one institution but t hese are not shown separately. Detail may not sum t o totals because of rounding.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-QS National Postsecondary St udent Aid Study
(NPSAS:OS).
Figure 6. Percentage distribution of full-time, full-year undergraduates who took
out private loans, by status of Stafford borrowing, status of federal aid
application, and dependency status: 2007-QB
Percent
100
80
60
40
20
0
Total Dependent Independent
~ M a x i m u m
Stafford
DLessthan
maximum
Applied,
no Stafford
Did not
apply for
aid
NOTE: In this analysis, full-lime undergraduates excludes foreign students and
unclassified undergraduates. Private loans are education loans from commercial
lenders which are not guaranteed by the federal government and carry market
interest rates. Maximum Stafford is a loan in the amount of the annual limit for
Stafford loans based on dependency status and class level. Some restrictions on
Stafford borrowing were not taken into account when computing a student's Stafford
borrowing limit. Because of these restrictions, some borrowers who appear to be
borrowing below the maximum amount are in fact borrowing at their maximum.
Therefore, estimates of the percentage of borrowers who are borrowing at less than
the maximum may be overstated. Detail may not sum to totals because of
rounding.
SOURCE: U.S. Department of Education, National Center for Education Statistics,
2007-08 National Postsecondary Student Aid Study (NPSAS:08).
Figure 7. Percentage of graduate students who took out private, Stafford, Graduate PLUS, and all loans, by type of degree
program: 2003-04 and 2007-08
Private loans
Stafford loans
Graduate PLUS loans
Total loans
L6
~ 1 2
I 3
38
44
p,,. .. '""' ~ 5 7
Stafford loans ~
2 6
2 9
Graduate PLUS loans
Total loans
Private loans
Stafford loans
Graduate PLUS loans
Total loans
0 20
28
32
40
Percent
69
60
Average loan amounts received by graduate
students, by type of loan: 2003--{)4 and 2007-{)8
Private loans
Stafford loans
Graduate PLUS loans
Total loans
2003-Q4 2007--{)8
$10,500
16,100
17,700
$8.400
15,600
15,500
18,500
NOTE: Amounts for 2003-04 have been adjusted for inflation
using the Consumer Price Index for urban households (CPI-U).
Amounts are averages for those who received the specified
type of aid.
2003-04
2007-08
76
75
79
80 100
1
First-professional programs include dentistry, medicine, optometry, osteopat hic medicine, pharmacy, podiatric medicine, veterinary medicine,
chiropractic, law, and theology.
NOTE: Private loans are education loans f rom commercial tenders which are not guaranteed by the federal government and carry market interest
rates. Graduate PLUS loans were not available in 2003-Q4.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2003-Q4 and 2007-QS National Postsecondary Student Aid
Studies (NPSAS:04 and NPSA$:08).
U.S. DEPARTMENT OF EDUCATION APRIL 2010
What Is the Price
of College?
Sticker, Net, and
Out -of-Pocket Prices
in 2007-08
AUTHOR
Christina Chang Wei
MPR Associates, Inc.
PROJECT OFFICER
Tracy Hunt-White
National Center for Education Statistics
Students and parents see
college attendance as a principal avenue
to middle-class life, and, given the rising
price of postsecondary education, they
are apprehensive about their ability t o af-
f ord it.
1
In a recent survey of college
freshmen, about two-thirds (66 percent)
reported having concerns about being
able to fi nance their education.
2
Many U. S.
policymakers and researchers share their
concern, and are exploring ways to make
college more affordable.
3
Legislators have
required colleges and universities t o pro-
vide more extensive information about
tuition and prices, and in the 2008 Higher
Education Opportunity Act mandated a
host of price-related measures, including
institutional disclosure of net prices
charged to students, the reporting of net
price data to the U.S. Department of Edu-
cation, and the creation and disclosure of
"tuition watch lists."
4
This Statistics in Brief shows the amounts
U.S. undergraduates pay on average for
postsecondary education, wi th and wi th-
out financial aid. Drawing upon data from
the National Postsecondary Student Aid
Study, a nationally representative survey
of all postsecondary students, included in
this report are the average prices for pub-
lic 2-year, public 4-year, private nonprofit
4-year, and for-profit institutions.
1
A recent public opinion poll showed that increasing numbers of
Americans view college as a necessity for success (lmmerwahr et al.
2009). The College Board (2009) reports that the published, or stick
er, price of college has increased more rapidly than the price of other
goods and services over the past three decades. See also Marchand
(2010), which reported on these findings.
1
The Higher Education Research Institute at the University of Cali
fornia, Los Angeles, has conducted a survey of freshmen each year
since 1973 (Pryor et al. 2009). See also recent media anent ion to the
issue (Lewin 2009).
3
Both governmental and nongovernmental study commissions have
convened to address this. See, for example, The College Board
(2008) and Harvey et al. (1998).
4
Public Law 110315, title I, Sec. 132. August 14, 2008. Higher Edu
cation Opportunity Act .

1es
NATIONAL CENTER FOR
. . EDUCATION STATISTICS
STUDY QUESTIONS
What are the average
prices paid by full-time
undergraduate students
and how do these prices
vary by the type of insti-
tution attended?
Most of the figures in this Brief display
data only for full-time undergraduates
5
who attended one institution. These
students comprised about one-third
(35 percent) of all undergraduates in
2007-08.
6
Of all full-time undergra-
duates, 18 percent were enrolled in
public 2-year institutions, 43 percent
attended public 4-year institutions, 21
percent were at private nonprofit 4-
year institutions, and 8 percent were
enrolled in for-profit institutions (table
1).
7
Focusing on full-time students who
attended only one institution allows for
comparisons in tuition, price of atten-
dance, and financial aid. Those attend-
ing full time generally have higher
overall expenses than do all students.
They also generally qualify for federal
aid and other assistance not available
to many part-time students (table 2).
5
"full time" status is defined as having been enrolled full time
in one postsecondary institution for 9 months or more during
the academic year.
6
National Postsecondary Student Aid Study (NPSAS:OS) Data
Analysis System.
1
The remaining students were enrolled in other types of insti
tutions or in more than one institution during the academic
year.
What factors are re-
lated to variations in
average sticker and
net prices among
those attending these
institutions?
KEY FINDINGS
There is a wide range of prices for
postsecondary education. Students
enrolled at public 2-year institutions
had the lowest average sticker price
($12,600) while those at private
nonprofit 4-year institutions had the
highest {$35,500).
8
Those at public
4-year institutions had an average
sticker price of $18,900, and those at
for-profit institutions had an aver-
age sticker price of $28,600.
After all financial aid is received, {in-
cluding grants, loans, and work-
study), the average out-of-pocket
net price ranged from $9,100 at
public 2-year institutions to $10,300
at public 4-year institutions, $16,000
at for-profit institutions, and
$16,600 at private nonprofit 4-year
institutions.
3
All comparisons of estimates were tested for statistical signi
ficance using the Student's t-statistic, and all differences cited
are statistically significant at the p < .OS level. No adjustments
for multiple comparisons were made. The standard errors for
the estimates c a n ~ found at http://nces.ed.gov/dasf
librarv/reoorts.asp.
2
How do the net
prices paid by under-
graduates vary by
family income?
Lower sticker prices often mean a
lower need-or eligibility-for fi-
nancial aid. Those attending private
institutions had the highest tuition
but they also received the most fi-
nancial aid. Undergraduates at pri-
vate nonprofit 4-year institutions
received the greatest amount of in-
stitutional grant aid while those at
for-profit institutions had the largest
proportion of borrowers.
Among low-income undergra-
duates enrolled full time, those at-
tending for-profit institutions had
the highest average net price after
grants and average out-of-pocket
net price.
1
What are the average prices paid by full-time
undergraduate students and how do these
prices vary by the type of institution attended?
This Brief discusses three different
measures of the price of an undergra-
duate education: the sticker price, the
net price after grants, and the out-of-
pocket net price.
The Sticker Price
The sticker price is the total price of an
undergraduate education including
tuition
9
and all other nontuition and
living expenses such as books, sup-
plies, and housing. The sticker price va-
ried widely by the type of institution
attended in 2007-08, rangi ng from an
average of $12,600 among undergra-
duates enrolled full time at public 2-
year institutions, to $18,900 at public 4-
year institutions, $28,600 at for-prof it
institutions, and $35,500 at private
nonprofit 4-year institutions (figure 1 ).
9
1n this report, the term "tuition includes both tuition and
fees. Sometimes institutions treat tuition and fees as a single
charge, and sometimes as separate charges. Tuition is defined
as the price ofinstruction and fees as the price of other services
provided by the school. The tuition amounts shown here in
elude those charged to outof-state and out-of-district students
enrolled in public 4-year and public 1-year institutions, respec-
tively.
FIGURE 1.
THE PRICE OF AN UNDERGRADUATE EDUCATION
for full-time, full-year undergraduates,
by type of institution attended: 2007- 08
Price
$40,000
30,000
20,000
10,000
0
Public 2-year Public 4-year For-profit
Type of institution
Private nonprofit
4-year
Average sticker w Average net price ~ A v e r a g e out-of-
price after grants pocket net price
NOTE: The "sticker price" is the total price of attendance which includes tuition and fees, books and supplies, housing,
meals, transportation, and other miscellaneous, or personal, expenses. The "net price is the price that must be paid after
receiving grants. The "out-of-pocket net price subtracts from the sticker price all forms of financial aid, including grants,
student loans, Parent PLUS loans, work-study, employer aid, job training benefits, veterans benefits, and any other finan
cia I aid received. Both the net price after grants and the out-of-pocket net price are calculated for all full-time, full-year
students, regardless of whether or not they received any financial ald. "Full-time, full-year' is defined as having been
enrolled in one postsecondary institution for 9 months or more full time. Estimates include students enrolled in Title IV
eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Standard error tables are available at
hnpl/nces.ed.gov/das/library/reports.asp.
SOURCE: U5. Department of Education, National Center for Education Statistics, 1007-08 National Postsecondary Student
Aid Study (NPSAS:08).
3
For some institutions (private nonprofit known in financial aid parlance as the supporting families (Wei forthcoming),
4-year institutions in particular), the "student budget." Student budgets which raises their nontuition expenses
tuition is a large component of the vary according to students' attendance as compared with students who have
sticker price. In 2007-08, average tui- and dependency status, family respon- no family responsibilities.
tion was $2,400 at public 2-year institu- sibilities, and residence (e.g., living at
tions, $7,100 at public 4-year home with parents, in an on-campus Students attending public 2-year insti-
institutions, $11,900 at for-profit insti- dormitory, or in off-campus housing). tutions-mostly community colleges-
tutions, and $23,400 at private non-
profit 4-year institutions {figure 2).
In 2007-08, students enrolled full time
in for-profit institutions had the high-
Nontuition expenses, which include est average nontuition expenses
had the lowest nontuition expenses. A
larger proportion lived at home with
their parents, which reduces costs for
room and board, as compared with
students at for-profit institutions (many
of whom are supporting themselves or
their own families) and those at 4-year
institutions (where a larger proportion
lived on campus or in off-campus hous-
ing) (Wei forthcoming).
books and supplies, housing and ($16,700), when compared with those
meals, transportation, and personal at other types of institutions {where
{or miscellaneous) expenses, also can average nontuition expenses ranged
vary by institution type. College or uni- from $10,200 to $12,1 00). Many stu-
versity financial aid officers usually de- dents at for-profit institutions are fi-
velop an estimate of the sticker price, nancially independent and are
In-state vs. out-of-state tuition
At most public 4-year
institutions, tuition charges are
generally higher for out-of-state
students than for in-state
residents, reflecting the state
subsidies public institutions
receive. In 2007- 08, the average
in-state tuition was $6,200 and
the average out-of-state tuition
was $15,1 00 for full-time
undergraduates enrolled in
public 4-year institutions.*
NPSAS:OS Data Analysis System (data not shown).
FIGURE 2.
TUITION AND NONTUITION EXPENSES
for full-time, full-year undergraduates,
by type of institution attended: 2007- 08
Price
$40,000
30,000
20,000
10,000
0
Public 2-year Public 4ycar For profit
Type of institution
Private
nonprofit
4-year
tJ Average
nontuition
expenses
Average
tuition
NOTE:"full-time, full year is defined as havlng been enrolled in one postsecondary institution for 9 months or more full
time. "Tuition" includes all tuition and fees. Non tuition expenses" include books, supplies, housing, meals, transporta
tion, and miscellaneous 01 pe1>onal expenses. Estimates include students enrolled in Title IV eligible postsecondary institu-
tions in the 50 states, DC, and Puerto Rico. Standard error tables are available at
htto://nces.ed.gov/das/librarv/reoorts.aso.
SOURCE: U 5. Department of Education, National (enter for Education Statistics, 2007-08 National Postsecondary Student
Aid Study (NPSAS:OS).
4
The Net Price After Grants those at public 4-year institutions,
Policymakers and researchers generally grant aid to full-time undergraduates
subtract grants from the sticker price lowered the average sticker price of
when discussing the net price of at- $18,900 to an average net price after
tending a postsecondary institution. grants of $15,200.
For example, Congress recently re-
quired institutions to make public both For those at private institutions (both
the sticker price and the average net for-profit and nonprofit), the average
price of attendance, calculated as the net price after grants was about
total price of attendance minus all $25,700. For those attending for-profit
grants received. In this Brief, two
measures of net price are discussed:
the "net price after grants" and the
"out-of-pocket net price," which is the
price after all financial aid (i ncluding
loans, which must be repaid} is t aken
into account.
10
Grant aid helped to lower the average
st icker price among full-time under-
graduates at public 2-year institutions
from $12,600 to an average net price
after grants of $10,600 (figure 1 ). For
10
Both net price and out of-pocket net price averages are cal-
culated for all students, regardless of whether they received
any financial aid. This method of calculating net price averages
for all students differs from that used to calculate average aid
amounts in this report (see table 2). Average aid amounts are
calculated only for students receiving a particular type of aid.
Those not receiving a specific type of aid (i.e., zero values) are
not included in the average for that aid. The average grant
therefore, will be greater than the difference between the
sticker price and the net price after grants.
institutions, the difference between
the average sticker price and the net
price af ter grants was about $2,800. At
private nonprofit 4-year instituti ons,
however, that difference was $10,000.
The Out-of-Pocket Net Price
The "out-of-pocket net price," defined
as the sticker price less all fi nancial aid
received, takes into account all forms
of financial aid, including grants, loans,
work-study, and other aid (as well as
Parent PLUS loans). The out-of-pocket
net price represents the amount t hat
must be paid immediately to enroll in a
postsecondary institution f or that aca-
demic year.'' Because the out-of-
pocket net price subtracts loans from
the sticker price, it measures the net
price only in the short t erm. Loans off-
set immediate costs to students and
their families, but they must be repaid
overtime.
Full-time undergraduates enroll ed at
public 2-year institutions had the low-
est average out-of-pocket net price
($9,100}, reflecting the already lower
sticker price at these institutions (figure
1}. Those at public 4-year instituti ons
had a slightly higher average out-of-
pocket net price ($1 0,300}. Students at
private institutions had the highest av-
erage out-of-pocket net prices ($16,000
at for-profit institutions and $16,600 at
private nonprofit 4-year institutions).
11
Since 1998, the federal government has also provided postse-
condary students and their families with various federal tax
benefits. These are not included in the definition of financial aid
and are not used in the calculation of net price in this study.
2
What factors are related to variations
in average sticker and net prices among
those attending these institutions?
Prices vary by institution type for sev-
eral reasons. Institutions charge differ-
ent levels of tuition (based on whether
they are public or private and the
amount of state and local subsidies re-
ceived); the demographic characteris-
tics of students (and thus their
nontuition expenses and eligibility for
federal and state grant aid) vary by
type of institution; and institutional
policies for awarding institutional aid
differ.
As noted previously, the amount of tui-
tion charged by institutions is a prima-
ry factor in the total price of
attendance, but students also incur dif-
ferent nontuition expenses depending
on their family responsibilities and liv-
ing arrangements.
Average net prices are further affected
by differences in the proportion of aid
recipients at each type of institution.
For example, the number of students
eligible for federal Pell Grants or state-
funded grant aid at a particular school
will affect the average net price after
grants, as will a school's policies for
awarding grants from institutional
funds. The out-of-pocket net price fur-
ther depends on the level of borrowing
among students and their parents and
other types of aid received, such as
work-study.
Public 2-Year Institutions
Full-time students attending public 2-
year institutions had the lowest aver-
age sticker price ($12,600), net price af-
ter grants ($1 0,600), and out-of-pocket
net price ($9, 1 00) among all undergra-
duates (figure 1 ).
Because students at public 2-year insti-
tutions had a lower average sticker
price initially, their average net prices
also were lower-even though they
had the smallest proportions of grant
recipients (56 percent) and students
who borrowed (23 percent took out a
student loan) (table 2).
6
Public 4-Year Institutions
Tuition at public 4-year institutions was
higher than at public 2-year institu-
tions, but not as high as at the private
institutions (figure 2). Students at these
institutions also have slightly higher
nontuition expenses than those at
public 2-year institutions, with a larger
proportion living on campus or away
from home {see Wei forthcoming).
Among those enrolled full time at pub-
lic 4-year institutions, 60 percent re-
ceived grant aid, about one-half (53
percent) took out student loans, and 10
percent received work-study support
(table 2). Grant aid helped to lower the
net price after grants to an average of
$15,200, and the addition of loans,
work-study, and other aid resulted in
an average out-of-pocket net price of
$10,300 (figure 1 ). This compares to an
average out-of-pocket net price of
$9,100 at public 2-year institutions-a
difference of $1,200 in the average out-
of-pocket net price, even though the
difference in the average sticker price
was $6,300.
For-Profit Institutions
For-profit institutions are privately
owned and operated and the profits
they generate benefit individual own-
ers and shareholders. The programs
can range from less than 1 year to 4-
year bachelor's and graduate degrees.
Most undergraduates enrolled in less-
than-4-year for-profit institutions are
pursuing certificates or associate's de-
grees in occupational training pro-
grams (Staklis forthcoming).
Undergraduates at for-profit institu-
tions also tend to be older, financially
independent, and have family respon-
sibilities. This increases their nontuition
expenses and hence, their average
sticker price. More undergraduates at
for-profit institutions received federal
grants (62 percent) than did students
in any other type of institution in our
analysis (figure 3). However, a smaller
percentage of for-profit students re-
ceived state, institutional, or private
grants than students in other sectors.
On average, these students had a net
price after grants of $25,80Q-not
measurably different than that of pri-
vate nonprofit 4-year institutions, and
higher than that of public institutions
(figure 1 ).
FIGURE 3.
SOURCES OF GRANT AID
for full-time, full-year undergraduates,
by type of institution attended: 2007- 08
Percent
100
80
60
40
20
0
Public 2year Public 4-year
62
For-profit
Type of institution
67
Private nonprofit
4-year
Federal grants ~ s t a t e grant s Institutional grants n Private source grants
NOTE: "Grants" include S<holarships and tuition waivers. full-time, full-year" is defined as having been enrolled in one
postsecondary institution for 9 months or more full time. Estimates indude students enrolled in Title IV eligible postsecon-
dary institutions ln the 50 states, DC, and Puerto Rico. Standard error tables are available at
hnpJ/nces.ed.gov/dasllibrarv/reports.asp.
SOURCE: U.S. Oepartment of Education, National Center for Education Statistics, 2007- 08 National Postsecondary Student
Aid Study (NPSAS:OS).
7
Student loans were critical to reducing Private Nonprofit 4-Year Institutions next highest sticker price: for-profit in-
the average out-of-pocket net price for Even though they had the highest av- stitutions ($25,800 and $16,000, re-
al I undergraduates, but particularly for erage sticker price ($35,500) of the four spectively) (figure 1 ).
those at for-profit institutions. For-
profit institutions had the largest pro-
portion of full-time undergraduates
with at least one loan in thei r financial
aid package: 9 out of 10 (92 percent)
received an aid package containi ng a
loan (table 2 and figure 4}, compared
to 65 percent of those at private non-
profit 4-year institutions, 53 percent at
public 4-year institutions, and 23 per-
cent at public 2-year institutions. The
high level of student borrowing at for-
profit institutions resulted in an aver-
age out-of-pocket net price of $16,000
(figure 1 ).
FIGURE 4.
TYPE OF AID PACKAGE
major sectors, financial aid recipients at
private nonprofit 4-year institutions al-
so received the largest average amount
of total aid ($21, 100), when compared
with those at for-profit institutions
($13, 1 00), public 4-year institutions
($11 ,000}, and public 2-year institutions
($5,400) (table 2).
In fact, aid received by undergraduates
at private nonprofit 4-year institutions
resulted in an average net price of
$25,500 and an average out-of-pocket
net price of $16,600, both of which
were not measurably different from
undergraduates at schools with the
for full-time, full-year undergraduates, by aid package received
and type of institution attended: 2007- 08
Percent
100
80
60
40
20
0
Percent distribution receiving aid
16
Public 2ycar Public 4-year For-profit
Type of institution
Private
nonprofit
4-year
Aid package
Grants, work-
study, or other
aid, with loans
Loans only
;::: Grants, work-
study, or other
aid, without loans
No aid received
NOTE: "Grants" include scholarships and tuition waivers. "Loans" include federal, state, ins1itutional, or private student
loans, excludiog Parent PLUS loans. "Othe(' aid includes job training, veterans benefits, employer aid, and Parent PLUS
loans. "Full-time, full-year" is defined as having been enrolled In one postse<ondary institution for 9 months or more full
time. Estimates include stlldents enrolled in Title IV eligible postsecondary institutions in the 50 states, OC, and Puerto
Rico. Detail may not sum to totals because of rounding. Standard error tables are available at http://nces.ed.gov/das/
librarv/reports.asp.
SOURCE: U.S. Department of EduCiltion, National Center for Education Statistics, 2007-08 National Postsecondary Stlldent
Aid Study (NPSAS:08).
8
The receipt of institutional grants, in
particular, was critical in lowering the
price for those attending private non-
profit 4-year colleges and universities.
About two-thirds (67 percent) of stu-
dents in private nonprofit +year
schools received institutional grants or
tuition waivers, a larger proportion
than at any other type of institution (30
percent at public 4-year institutions, 17
percent at public 2-year institutions,
and 7 percent at for-profit institutions)
(f igure 3). The average institutional
grant received by those attending pri-
vate nonprofit 4-year institutions was
$10,400,
12
which helped reduce the av-
erage sticker price to an average net
price after grants of $25,500 (figure 1 ).
Work-study was also an important
source of aid to those at private non-
profit 4-year institutions. Nearly one-
third (31 percent) of all full-time un-
dergraduates at private nonprofit 4-
year institutions received work-study
aid, the highest percentage among all
full-time undergraduates (between 2
and 10 percent of undergraduates at
other types of institutions received
work-study aid) (table 2). With the aid
of student loans, work-study, and other
types of support, full-time undergra-
duates at private nonprofit 4-year insti-
tutions had an average out-of-pocket
net price of $16,600-not measurably
different from those attending for-
profit institutions ($16,000) (figure 1 ).
12
NPSAS:OS Data Analysis System (data not shown).
3
How do the net prices
paid by undergraduates
vary by family income?
Figure 5 shows the average net price
after grants among dependent under-
graduates by their family income and
type of institution attended. Among
low-income and low middle-income
dependent students, those with the
highest average net price after grants
were enrolled at for-profit institutions.
In contrast, among high middle-
income and high-income students, the
average net price af ter grants for those
at for-profit institutions was not mea-
surably different from those at private
nonprofit 4-year institutions.
FIGURE 5.
NET PRICE AFTER GRANTS BY INCOME
for full-time, full-year dependent undergraduates, by family income
category and type of institution attended: 2007-08
Price
$40,000
30,000
20,000
10,000
0
Low income
Average net price after grants
Low
middle-income
High
middle-income
Highincome
Public 2year D Public 4-year Forprofit 0 Private nonprofit 4year
NOTE: The "net price after grants" Is the price of attending a postsecondary institution after all grants have been received.
This amount is calculated by subtracting total grant aid from the sticker price. The sticker price is the total price of atten-
dance and includes tuition and fees, books and supplies, housing, meals, transportation, and other miscellaneous, or per-
sonal, expenses. Grant aid includes grants, scholarships, tuition waivers, and other gift aid that does not need to be repaid.
The net price after grants is calculated for all students, regardless of whether or notthey received any grant aid. Family
income categories were based upon parents' annual income in 2006. Dollar cutoffs are based on the distribution among all
dependent undergraduates: "low-income" was the lowest 25th percentile (less than $36, 100); "low middle-income"was
the 26tll to 50th percentile ($36, 100-$66,600); "High middle-income was the 51st to 75th percentile ($66,60Q-
$104,6001; and "High-income" was the highest 25th percentile ($104,600 or more). ' Full-time, full-yea(' is defined as
having been enrolled in one postsecondary institution for9 months or more full time. Estimates Include students enrolled
in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Standard error tables are available at
http://nct>5.ed.gov/dasllibrarv/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics. 2007-08 National Postsecondary Student
Aid Study (NPSAS:OS).
9
The average out-of-pocket net price,
on the other hand, shows a slightly dif-
ferent pattern (figure 6). After borrow-
ing, low-income and low middle-
income undergraduates enrolled at for-
prof it institutions continued to have
the highest average out-of-pocket net
price, when compared with those at
other institutions. However, among
those with incomes above t he median
(i.e., high middle-income and high-
income students), the average out-of-
pocket net price was highest for those
enrolled at private nonprofit 4-year in-
stitutions.
FIGURE 6.
OUT -OF-POCKET NET PRICE BY INCOME
for full-time, full-year dependent undergraduates,
by family il)come: 2007- 08
Price
$40,000
30,000
20,000
10,000
0
Low income
Average out-of-pocket net price
Low
middle-income
High
middle-income
Highincome
Public 2year 0 Public 4year For-profit IJ Private nonprofit 4-year
NOTE: The "out-of-pocket net price is the price of attending a post5econdaryinstitution after all aid has been received.
This amount is calculated by subtracting total financial aid from the sticker price. The sticker price is the total price of at
tendance and includes tuition and fees, books and supplies, housing, meals, transportation, and other miscellaneoos, or
personal, expenses. Financial aid includes grants, student loans, Parent PLUS loans, work-study, employer aid, job training
benefi\5, veterans benefits, and any other type of aid.lhe ootof-pocket net price is calculated for all students, regardless
of whether or not they received any aid. Family income categories were based upon parents' annual income in 2006. Dollar
cutoffs are based on the distribution among all dependent undergraduates: "low-income was the lowest 25th percentile
(less than $3-6,100); "low middle-income was the 26th to 50th percentile ($36, 106-$66,600); "High middle-income"
was the 51st to 75th peJcentile ($66,600-$104,600); and''Higlrincome was the highest 25th percentile ($104,600 or
more). "Full-time, full-year" is defined as having been enrolled in one postsecondary instirution for 9 months or more full
time. Estimates include studen\5 enrolled in Trtle IV eligible postsecondary institutions in the 50 states, OC, and Puerto
Rico. Standard error tables are available at htto:l/nm.ed.gov/das/librarv/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Post5econdary Student
Aid Study (NPSAS:OS).
10
SUPPLEMENTAL TABLES
TABLE 1.
UNDERGRADUATE ENROLLMENT, TUITION, AND STICKER PRICE
for all undergraduates and full-time, full-year undergraduates, by type of
institution attended: 2007- 08
All undergraduates Full-time,
Percent Average Percent Average
distribution Average sticker distribution Average sticker
enrolled tuition' price' enrolled tuition' price
1
Total 100.0 $5,800 $14,000 100.0 $10,300 $22,400
Type of institution
Public 2-year 40.0 1,200 7,000 17.8 2,400 12,600
Public 4-year 29.2 5,500 15,200 42.7 7,100 18,900
Private nonprofit 4-year 13.0 17,800 28,200 20.6 23,400 35,500
For-profit 9.0 10,200 20,600 8.5 11,900 28,600
Other, or more than
one institution 8.8 4,800 12,300 10.5 7,000 18,000
TABLE 2.
FINANCIAL AID
for all undergraduates and full-time, full-year undergraduates,
by type of institution attended: 2007- 08
1
Average tuition and average sticker price estimates
are shown for those attending one lnstiwtlon only.
NOTE: "full-time, full-year is defined as having been
enrolled in a postsecondary institution for 9 months
or more full time. "Tuition includes all tllition and
fees. "Stkker price" is thetotal price of attendance
which includes tllition and fees, books and supplies,
housing, meals, transportation, and other miscella-
neous, or personal, expenses. Estimates include stu-
dents enrolled in Title IV eligible postsecondary
institutions in the 50 states, DC, and Puerto Rico. De-
tail may not sum to totals because of rounding. Stan
dard error tables are available at

SOURCE: U.S. Department of Education, National Cen-
ter for Education Statistics, 2007- 08 National Postse
condary Student Ai d Study (NPSAS:OS).
Any grants Any loans Any work-study Any aid NOTE:"Grants" include scholarships
Average Average Average Average
and tuition waiver;. "Loans" Include
Percent amount Percent amount Percent amount Percent amount
federal, state, institutional, or private
student loans, excluding Parent PLUS
loans. rota!" aid includes grants,
All undergraduates 51.7 $4,900 38.5 $7,100 7.4 $2,400 65.6 $9,100
loans, job training, veterans benefits,
employer aid, and Parent PLUS loans.
"All undergraduates" include those
Type of institution
attending more than one institution.
"Full-time, full-yea(' is defined as
Public 2-year 39.6 2,200 13.2 4,100 3.3 3,000 47.6 3,400 having been enrolled in one postse-
Public4-year 52.9 5,200 46.2 6,600 7.7 2,500 71.3 9.400
condary institution for 9 months or
more full time. Average aid amounts
Private nonprofit 4-year 73.6 10,200 58.9 9,100 23.2 2,100 84.7 17,400 are calculated only for students re-
For-profit 70.4 3,200 91.6 8,100 2.0 3,500 96.3 10,800
ceiving a particular type of aid. Those
not receiving a specific type of aid
(i.e., zero values) are not included in
Full-time, full-year 65.3 7,200 53.1 8,000 13.7 2,300 80.1 12,900
theaverageforthat aid. Estimates
include students enrolled in mle IV
eligible pnstsecondary institutions in
Type of institution
the 50 states, DC, and Puerto Rico.
Standard error tables are available at
Public 2-year 55.7 3.700 22.5 4,900 6.9 2,600 65.7 5,400
Public4-year 60.4 6,100 52.7 7,100 10.5 2,400 78.3 11,000

SOURCE: U.S. Oepartmem ofEducation,
Private nonprofit 4-year 81.2 12,300 65.0 9,800 31.5 2,100 89.4 21,100 National Ceoter for fducation Statistics,
For-profit 71.9 4,000 91.6 9,600 2.0 3,600 96.7 13,100
2007-08 National Postsecondary Stu
dent Aid Study (NPSAS:08).
1l
FIND OUT MORE
For questions about content, ordering additional copies of this Statistics in Brief,
or to view this report online, go to:
http://nces.ed.gov/insert-url/
More detailed information on the price of undergra-
duate education and undergraduate financing can be
found in Web Tables produced by the National Center
for Education Statistics (NCES) using the 2007-08 Na-
tional Postsecondary Student Aid Study (NPSAS:08)
data. These Web Tables are a comprehensive source of
information on financial aid awarded to undergra-
duate students during the 2007-08 academic year. In-
cluded are estimates of tuition, price of attendance,
and financial aid. Additional information on the de-
mographic characteristics of 2007- 08 undergraduates
can be found in a second set of Web Tables.
Web Tables-Student Financing of Undergraduate Edu-
cation: 2007-08 (NCES 2010-162). (link will be added)
Web Tables-Profile of Undergraduate Students in U.S.
Postsecondary Institutions: 2007- 08 (NCES 2010-205).
(link will be added)
Readers may also be interested in the following
NCES products related to the topic of this Statistics in
Brief:
2007-08 National Postsecondary Student Aid Study
(NPSAS:08): Student Financial Aid Estimates: First Look
(NCES 2009-166).
http://nces.ed.gov /pu bs2009 /2009166.pdf
Undergraduate Financial Aid Estimates by Type of Insti-
tution in 2007-08 (NCES 2009-201).
http://nces.ed.gov/pubs2009/2009201.pdf
-------------------------------------- 12 --------------------------------------
TECHNICAL NOTES
Survey Methodology
The statistics provided in this Statistics
in Brief are based on data collected
through the 2007- 08 National Postse-
condary Student Aid Study (NPSAS:08).
NPSAS:08 is the seventh administration
of NPSAS, which has been conducted
every 3 to 4 years since 1986-87. The
NPSAS:08 target population includes
all eligible students enrolled in Title IV
institutions in the United States at any
time between July 1, 2007 and June 30,
2008, representing about 21 million
undergraduates and 3 million graduate
students enrolled in over 6,000 institu-
tions.
The institution sampling frame for
NPSAS:08 was constructed from the
2004-05 and 2005-061nstitutional
Characteristics, Fall Enrollment, and
Completions files of the Integrated
Postsecondary Education Data System
(IPEDS). The sampling design consisted
of first selecting eligible institutions,
from which students were sampled in
the second stage. Institutions were se-
lected with probabilities proportional
to a composite measure of size based
on expected enrollment. Approximate-
ly 1,700 institutions participated in the
study, yielding an unweighted re-
sponse rate of 89 percent. Approx-
imately 114,000 undergraduates and
14,000 graduate students were study
respondents, yielding an unweighted
response rate of 96 percent. Estimates
were weighted to adjust for the un-
equal probability of selection into the
sample and for nonresponse.
NPSAS covers broad topics concerning
student enrollment in postsecondary
education and how students, and their
families, finance their education. Res-
pondents provided data through in-
struments administered over the
Internet, by telephone, or in person. In
addition to respondent-provided data,
data on respondents were collected
from other relevant databases, includ-
ing U.S. Department of Education
records on student loan and grant pro-
grams and student financial aid appli-
cations.
VARIABLES USED
Two broad categories of error occur in
estimates generated from surveys:
sampling and nonsampling errors.
Sampling errors occur when observa-
tions are based on samples rather than
on entire populations. Nonsampling
errors can be attributed to several
sources: incomplete information about
all respondents (e.g., some students or
institutions refused to participate, or
students participated but answered
only certain items); differences among
respondents in question interpretation;
inability or unwillingness to give cor-
All estimates presented in this Statistics in Brief were produced using the
Data Analysis System (DAS). See "Run Your Own Analysis" for more informa-
tion on the DAS and other web-based software applications that enable us-
ers to generate tables for most of the postsecondary surveys conducted by
NCES. The program files that generated the statistics presented here can be
found at [insert-uri-here}. The variables used in these analyses include the
following:
Label
Attendance status
Family income for dependent students
Federal grants
Institutional grants
Net price after grants
Non-tuition expenses
Out-of-pocket net price
Private grants
State grants
Sticker price
Total financial aid
Total grants
Total student loans
Tuition
Type of aid package
Type of institution
Work-study
Name
ATINSTAT
PGDEP
TFEDGRT
INGRTAMT
NETCST3
BUDNONAJ
NETCSTl
PRIVAID
STGTAMT
BUDGETAJ
TOT AID
TOTGRT
TOTLOAN
TUITION2
AIDTYPE
SEGOR4
TOTWKST
-------------------------------------- 13 --------------------------------------
rect information; mistakes in recording
or coding data; and other errors of col-
lecting, processing, sampling, and im-
puting missing data.
For more information on NPSA$:08 me-
thodology, see 2007-08 National Post-
secondary Student Aid Study (NPSAS:08)
Full-scale Methodology Report (NCES
201 0-1 88) (forthcoming).
Item Response Rates
NCES Statistical Standard 4-4-1 states
that " [a)ny survey stage of data collec-
tion with a unit or item response rate
less than 85 percent must be evaluated
for the potential magnitude of nonres-
ponse bias before the data or any anal-
REFERENCES
The College Board (2009). Trends in
College Pricing 2009. Retrieved on
February 11,2010, from
http://www.trends-col legeboard.
com/college pricing/pdf/
2009 Trends College Pricing.pdf.
The College Board (2008). Fulfilling the
Commitment: Recommendations for
Reforming Federal Student Aid: TheRe-
port from the Rethinking Student Aid
Study Group. Retrieved on January 15,
201 0, from .!:!.llQ.;LL
professionals.collegeboard.com/
profdownload/rethinking-stu-aid-
fulfilling-commitment-
recom mendations.pdf.
Harvey, J., Williams, R.M., Kirshstein, RJ.,
O'Malley, A.S., and Wellman, J.V.
( 1998). Straight Talk About College
Costs and Prices. Report of the National
Commission on the Cost of Higher Edu-
cation. Phoenix, AZ: Oryx Press.
ysis using the data may be released"
(U.S. Department of Education 2003).
Several derived variables used in this
report were comprised of variables that
had response rates of less than 85 per-
cent before imputation. EMPLYAM1
and EMPL YAM2 (measures of employer
aid received, and components of
TOTGRT) had response rates of 79 per-
cent and 80 percent, respectively. Non-
response bias analyses showed that for
EMPL YAM 1, there was statistically sig-
nificant bias on 87 percent of the cha-
racteristics analyzed, and for
EMPLYAM2, there was significant bias
on 76 percent of all characteristics.
PRIVLOAN (private loans received, and
lmmerwahr, J., Johnson, J., Gasbarra, P.,
Ott, A., and Rochkind, J. (2009).
Squeeze Play 2009: The Public's Views
on College Costs Today. New York and
San Jose, CA: Public Agenda and The
National Center for Public Policy and
Higher Education.
Lewin, T. (2009, October 20). College
Costs Keep Rising, Report Says. The
New York Times.
Marchand, A. (201 0, January 21 ). Cost of
College Is a Big Worry of Freshmen in
National Survey. The Chronicle of High-
er Education.
Pryor, J.H., Hurtado, S., DeAngelo, L., Pa-
lucki Blake, l., and Tran, S. (2009). The
American Freshman: National Norms
Fall2009. Los Angeles: Higher Educa-
tion Research Instit ute, University of
California, Los Angeles.
a component ofTOTLOAN) had a re-
sponse rate of 67 percent. The nonres-
ponse bias analysis for this variable
showed that respondent s and nonres-
pondents were significantly different
on 80 percent of all analyzed characte-
ristics. PCTDEP had a response rate of
56 percent, and the non response bias
analysis also showed significant bias on
80 percent of all characteristics. For
more detailed information on nonres-
ponse bias analysis and an overview of
the survey methodology, see 2007-08
National Postsecondary Student Aid
Study (NPSAS:08) Full-scale Methodology
Report (NCES 201 0-188) (forthcoming).
Staklis, S. (forthcoming). Web Tables-
Profile of Undergraduate Students:
2007-08 (NCES 2010-205). National
Center for Education Statistics, Insti-
tute of Education Sciences, U.S. De-
partment of Education. Washington,
DC.
U.S. Department of Education, National
Center for Education Statistics. (2003).
NCES Statistical Standards (NCES 2003-
601 ). Washington, DC.
Wei, C. (forthcoming). Web Tables-
Student Financing of Undergraduate
Education: 2007- 08 (NCES 201 0-162).
National Center for Education Statis-
tics, Institute of Education Sciences,
U.S. Department of Education. Wash-
ington, DC.
-------------------------------------- 14 --------------------------------------
RUN YOUR OWN ANALYSIS WITH DATALAB
You can replicate or expand upon the figures and tables in this report, or
even create your own. Data lab has several different tools that allow you to
customize and generate output from a variety of different survey datasets.
Visit Data Lab at:
QuickS tats
Create a simple table
quickly
View )'O'Jr cutp-..rt as a
chart or table
Choose from many data
sets each with about one
hundred Y!lriables
Select from
postsecondary surmes
Cover artwork iStockphoto.<Om!centauria.
http:/ /nces.ed.gov/datalab/
Dotolab
PowerS tats
Produee- complex tables
Run linear and logistic

Choose from many data
sets each with thousands
of variabl&s
library
Search existing tables and
figures to find to your
questions Vistt NCES Tables
sod Figures
Comr119 soon Thousands of
pubhshed tables using
Po,verStats and OuickStats
Help
Need Ccntact
John Vuicka
MPR Assoc.iatas Inc.
d3tala!rhelp@mpnrte com
Aurora D'Amico. NCES
Au.ora gcw
Need access to restncted
data? Learn more on The
Restnctedll se Data
Procedures Manual
-------------------------------------- 15 --------------------------------------
www.ed.gov ies.ed.gov
Please print
From: Kanter, Martha
Sent: Friday, April 30, 2010 2:49PM
To: Yuan, Georgia
Subject: FW: Tony Miller briefing
[(b)(S)
From: Bergeron, David
Sent: Friday, April 30, 2010 2:37PM
To: Kanter, Martha
Subject: FW: Tony Miller briefing
From: Weko, Tom
Sent: Friday, April 30, 2010 2:35PM
To: Bergeron, David
Cc: Miller, Elise; Hunt-White, Tracy
Subject: Tony Mill er briefing
(b)(5)
From: Yuan Georgia
To: Canada, June
CC:
Date: 4/30/2010 3: 11 :22 PM
Subject: FW: Tony Mmerbriefing
(b)(S)
e
n
(b)(5)
(b){5)
(b)(5)
1 The "New Majority" of Undergraduates:
2 Students in Community Colleges and
3 For-Profit Institutions
4 Introduction
5 Depictions of postsecondaty education in the news, films, and oth-
6 er media often focus on undergraduates who study at traditional
7 4-year public universities and nonprofit colleges (Swidey 2009;
8 Russo 2006; U.S. Department of Education 2002). Yet even two
9 decades ago these students accounted for just 55 percent of stu-
l 0 dents enrolled in U.S. postsecondaty education (Choy and Gifford
1990).
1
Now a majority of undergraduates in U.S. postsecondaty
education attend community colleges and for-profit institutions.
These two sectors together enrolled 54 percent of the approximate-
ly 21 million undergraduates who attended U.S. postsecondary
institutions in 2007-08. Some 44 percent of undergraduates at-
l6 tended community colleges, the single largest sector of U.S. postse-
condary education, and another 10 percent attended for-profit
institutions (figure 1). While the for-profit sector accounts for a
relatively small percentage of undergraduates, its enrollment is
growing at a rapid pace relative to other postsecondaty sectors
(Horn and Li 2009; Wilson 2010).
Who are the undergraduate students served by community colleges
and for-profit institutions, and how do they compare to students
who are enrolled at traditional 4-year institutions? Drawing on data
from a nationally representative sample of undergraduates enrolled
L Based on fall enrollment in the 1986-87 National Postsecondary Stu-
dent Aid Study (NPSAS:87).
THE "NEW MAJORITY" OF UNDERGRADUATES:
STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS
1 in U.S. postseconda1y institutions, this Statistics in Brief27 enrolled in bachelor's degree programs, while 43 percent
2 addresses the following questions. 28 were pursuing associate's degrees, and the remaining stu-
29 dents were either in certificate programs or not in a de-
3 How do students enrolled in for-profit institutions and 30 gree program (table 1).2 The average 2007-08 tuition
4 community colleges differ from each other and from 31 and fees for full-time, full-year students was $11,900,
5 those enrolled in traditional4-year institutions in terms 32 and for all students, $10,200 (Wei forthcoming).
6 of:
7
8
9
10
11
12
13
33 Community colleges: Also known as public 2-year institu-
1) demographic and socioeconomic characreristics?34 tions, community colleges offer associate's degrees in
2) indicators of high school academic preparation 35 occupational and academic fields, as well as occupational
and being the first in their family to attend col- 36 certificates. Unlike for-profit institutions, which offer
lege? 37 primarily occupational programs, community colleges
38 offer a variety of progran1s with different goals. These
3) enrollment patterns and major fields of study?
39 include academic transfer to 4-year institutions, voca-
4) the extent to which they combine work and
study?
40 tionalltechnical education, continuing education, and
41 developmental education (Cohen and Brawer 2003).
14 The three groups of postsecondary institutions comparcJ2 Conu11w1ity colleges, therefore, serve a range of students,
15 in this Brief are briefly defined below. 43 from those needing basic remedial courses to very high
44 achieving students meeting lower-division bachelor's
16 For-profit institutions: A recent report by the U.S. Gov- 45 degree requirements before transferring to a selective 4-
17 ernment Accountability Office defined propriety institu-46 year college or university (Provasnik and Planty 2008).
18 tions as "Institutions that are privately-owned whose net47 The average 2007-08 tuition and fees for full-time, full-
19 earnings can benefit a shareholder or individual; that is, 48 year students was $2,400, and for all students, $1,200
20 for-profit institutions." (U.S. GovernmentAccountabili-49 (WeiforthcominiJ.
21 ty Office 2009). Among undergraduates enrolled in for-
22 profit institutions in 2007-08, about half attended insti-50
23 tutions authorized to award 4-year (bachelor) degrees, 27
24 percent attended 2-year institutions, and 22 percent at-
25 tended less-than-2-year institutions (figure 2). Half of
26 those in for-profit 4-year institutions were
2
At community colleges, 79 percent of undergraduates were in
associate's degree programs, 7 percent in cerrillcate programs,
and 11 percent were not in a degree program. In traditional 4-
year institutions, 93 percent of lmdergraduates were enrolled
in bachelor's degree programs.
THE "NEW MAJORITY" OF UNDERGRADUATES: I I
STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS 3
1 Public and private nonprofit 4-year institutions: Public 28
2 and private nonprofit institutions serve as the main 29
3 comparison group in this analysis. Consistent with pre- 30
4 vious research, they are referred to as "traditional4-year 31
5 institutions" (Tierney and Hentschke 2007). Together, 32
6 students in these two sectors accounted for 46 percent of
33
7 all undergraduates in 2007-08-32 percent attending
34
8 public 4-year colleges and universities, and 14 percent
35
9 attending private nonprofit 4-year institutions (figure 1)
36
ic), and income (larger proportion oflow-
income students).
3
Community college students
also differ from traditional 4-year college stu-
dents on most of these characteristics, but to a
lesser extent.
4
The part-time status of the majority of commu-
nity coll ege students sets them apart from their
peers in for-profit and traditional 4-year institu-
tions, the majority of whom attend full time.
10 The average 2007-08 tuition and fees for full-time, full-
11 year students in 4-year public institutions was $5,700
37
12 and $6,300 and for all students, $4,100 and $6,300, at 38
Health care fields are the most common fields of
study among students in for-profit institutions,
and arean1ong the most common fields for
community college students.
13 non-doctorate- and doctorate-granting institutions, re- 39
14 spectively. Tuition and fees in private nonprofit institu- 40
15 tions totaled $20, 800 and $25,800 for full-time, full- 41
Regardless of where undergraduates enroll, a
majority work while enrolled.
16 year students and $15,200 and $20,700 for all students, 42
17 at non-doctorate- and doctorate-granting institutions,
18 respectively (Wei forthcoming).
19
20
UNDERGRADUATE AWARDS CONFERRED
Table 2 shows the distribution of undergraduate
awards for the institution comparison groups
(shaded) presented in this report. These comple-
tions data are reported by all institutions that partic-
ipate in Title IV student aid programs. Based on
these data, community colleges awarded 46 per-
cent of all certificates and 69 percent of associate's
degrees in 2007. For-profit institutions awarded 42
percent of all certificates and 16 percent of asso-
ciate's degrees. For-profit 4-year institutions
awarded about 5 percent of bachelor's degrees
and traditional 4-year colleges and universities
awarded the remaining 95 percent.
21 Key findings
43 Demographic and Socioeconomic
44 Characteristics
5 Many demographic characteristics, including gender,
6 race/ethnicity, family status, income and age distinguish
7 students enrolled in for-profit institutions and commu-
8 nity colleges from those enrolled in traditional 4-ycar
9 colleges. Students in the for-profit sector, for example,
0 were the most likely to be female, Black or Hispanic, to
1 have children or other dependents, and to be single with
2 dependents in 2007-08 (figure 3). Women constituted a
3 majority of undergraduates in all three sectors, but com-
4 prised 69 percent of students in for-profit institutions. fn
5 contrast, 57 percent of community college students and
6 55 percent of traditional4-year institution students were
7 women.
22 Students enrolled in for-profit institutions differ
3
All comparisons of estimates were tested for statistical signi-
ficance using the Student's !-statistic, and all differences cited
are statistically significant at the p < .05 level. No adjustments
for mltltiple comparisons were made. The standard errors for
the estimates can be found at
http://nces.ed.goy/das/library/rs:ports.asp.
23 from those enrolled in traditional 4-year institu-
24 tions across a wide range of demographic and
25 socioeconomic characteristics, including gender
26 (larger proportion of women), age (older),
27 race/ethnicity (nearly half are Black or Hispan-
4
Determined by a Student' s t-srarisric that rests the signillc-
ance of the difference of differences.
THE "NEW MAJORITY" OF UNDERGRADUATES:
STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS
1 Nearly half of the students in for-profit institutions were35 students were 31 percent for dependent students and 22
2 Black (25 percent) or Hispanic (21 percent). In commu-36 percent for independent students.
3 nity colleges, the percentages of Black (14 percent) and
4 Hispanic students (15 percent) were lower than those in 37 The income status of community college students rela-
5 for-profit institutions but higher than those in tradition-38 tive to their peers in traditional 4-year institutions dif-
6 al4-year institutions (1 1 and 12 percent, respectively, 39 fered for dependent and independent students. A larger
7 were Black or Hispanic).
40 percentage of dependent community college students
4 1 were from low-income families (31 vs. 20 percent), while
8 Community college students also fel l between for-profit 42 a smaller percentage of independent students earned low
9 and traditional 4-year institutions in terms of their like- 43 incomes (22 vs. 28 percent) .
10 lihood of having dependents and being single with de-
11 pendents. Some 32 percent of communi ty college
44 The average age of students enrolled at conummity col-
12 students had children or other dependents, versus 49 anJ5 leges and for-profit institutions was older than the aver-
13 14 percent in for-profit and traditional 4-year institu-
14
15
tions, respectively. The same pattern was found for the
percentage of undergraduates who were single with de-
16 pendents (16 vs. 31 and 7 percent) .
46 age age of students at traditional 4-year institutions (28
47 vs. 24) (figure 5) . Although the average age of students
48 in both for-profit institutions and communi ty coll eges
49 was 28, the distri bution of students by age group in each
50 type of institution differed (figure 5) . Undergraduates
17 When income is considered for postsecondaty students, 51 enroll ed in for-profit institutions were least likely to be
18 family income is reported for students considered finan- 52 traditional-aged students-23 or younger.
19 cially dependent on their parents, and student (and
20 spousal) income is reported for chose considered finan- 53
2 1 cially independent. Dependency status is based largely 54
22 on age: all students 24 or older are considered indepen- 55
23 dent. Married students and students who are parents are
6
24 also considered independent. Reflecting these characte-
5
25 ristics, just over three-quarters (76 percent) of students
57
26 in for-profit institutions were independent (figure 4). I n ~ ~
27 contrast, 57 percent of community college students and
60
28 32 percent of students at traditional4-year institutions
61
29 were independent.
62
30 Students in the for-profit sector were the most likely to
31
32
63
64
be low income; roughly half (51 percent) of dependent
students had low family incomes, and roughJy one-third 65
33 of independent students earned low incomes (32 per-
66
Indicators of High School Academic
Preparation and First in Family to Attend
College
Students in for-profit institutions appeared to be the
least academically prepared based on indicators of high
school completion and academic preparation available in
the data. Indicators include high school completion sta-
tus (standard diploma), average grades, and the number
of years of mathematics taken. As illustrated in figure 6,
compared with their peers in community colleges and
traditional 4-year institutions, students in for-profit in-
stitutions were the most likely co have enrolled in postse-
condaty education without a standard hi gh school
diploma (19 percent, vs. 14 percent and 7 percent, re-
34 cent) .s Comparable percentages for conununity college 67 spectively, of community college and traditional4-year
5
Low income is defmed as a parent (dependent wldergra-
duates) or student (including spouse's income, for indepen-
dent wldergraduates) income that is at or below the 25th
percentile of the income distribution for each group. In 2006
(the year used for 2007-08 financial ajd applications) the in-
come levels were at or below $36,000 for dependent students
and at or below $11,000 for independent students.
THE "NEW MAJORITY" OF UNDERGRADUATES: I
STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS 5
1 college students) .
6
They were also least likely to report 30 in for-profit institutions attended full time for the full
2 earning grades ofB or higher (54 vs. 6 1 and 81 percent),31 year and 33 percent attended full time for part of the
3 or completing 4 or more years of high school mathemat-32 year. A higher percentage of undergraduates attended
4 ics (48 vs. 54 and 76 percent) ? At the same time, about 33 full time, full year in traditional4-year institutions (59
5 half of for-profi t students were the first in their famil y to34 percent) than in for-profit institutions.
6 enroll in postsecondary education, as measured by the
7 level of education achieved by either parent (figure 6). 35 Not ali students attend classes on campuses, but instead
8 SpecificaHy, among for-profi t students, 51 percent were 36 participate in distance education. Students in communi-
9 from families in which neither parent had an education 37 ty coll eges and for-profit institutions di d so more often
10 beyond high school, compared with 39 and 25 percent, 38 chan their peers at traditional4-year colleges. About one-
11 respectively, of community coll ege and tradi tional 4-yea
1
39 quarter of community college students and about one-
12 college students. 40 fifth of students in for-profi t institutions reported taking
41 distance education courses during the 2007-08 academic
13 Enrollment Patterns and Major Fields of
14 Study
42 year (figure 8) . In contrast, 17 percent of students at tra-
43 di tional 4-year institutions reported the same. In addi-
44 cion, a larger percentage offor-proflt enrollees (12
15 The part-time enrollment patterns of conumrnity collegC45 percent) than those at commttnity colleges (3 percent)
16 students distinguish them from their peers in both for- 46 reported that their entire program was offered through
17 profi t and traditiona14-year institutions (figure 7) . Se- 47 distance education.
18 venty percent attended part time, while the same percen-
19 cage of students in traditional 4-year and for-profit 4 8 With respect to fields of study, community college stu-
20 institutions attended full time. 49 dents sought both academic and occupational prepara-
50 cion (figure 9) . For example, 7 percent of community
2 1
22
23
While part-time attendance offers a flexibility often
5 1 college students and 22 percent of students in traditional
needed by studenrs attending community colleges (Co- 52 4-year institutions majored in humanities and social
wen and Brower 2003) , it is also associated with lower 53 sciences, compared with 2 percent of students in for-
24 rates of degree completion relative to full-time atten-
25 dance (e.g., Berkner, He, and Cataldi 2002).
54 profit institutions. Another 20 percent at community
55 coll eges and 6 percent at tradi tional 4-year institutions
56 majored in general studies and other fields, compared
26 Looking at attendance patterns over the year, in
57
27 2007-08, some 38 percent of community college stu-
58
28 dents attended part time for part of the year and 32 per-
59
29 cent for the full year.
8
By contrast, 37 percent of
with less chan 1 percent at for-profit institutions. Com-
munity coll ege students majoring in humanities, social
sciences, and general studies are largely those who seek to
transfer to a 4-year college (Berkner, Horn, and Chrne
6 1 2000).
6
These students may have dropped out of high school, ob-
tained aGED or high school equivalency certificate, been 62 Health care fields accounted for one-third of undergra-
home schooled, or attended a foreign high school. 63 duace majors in for-profit institutions and about one-
7 Data on high school academic preparation (GPA and
courses) is available only for students younger than 30, which 64 fifth (21 percent) in community coll eges. In conuast, 9
accowued for 77 percent of alltmdergraduares (figure 5). 65 percent of students in traditional 4-year institutions ma-
8 A full year is defined as 9 or more months. Those attending 66 jored in health care fields. l n addition to health care
part year may have completed a program of study or left with-67 fields, students in for-profit institutions were the most
out completing, either temporari ly or permanently.
68 likely to major in fields that are grouped under the head-
I I
THE "NEW MAJORITY" OF UNDERGRADUATES:
6 STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS
1 ing "other applied" (26 vs. 16 and 14 percent at com- 16
2 munity coll eges and tradi t ional 17
3 4-year institutions, respectively). These fields include 18
4 criminal justice, communications, and design and ap- 19
5 plied arts, among others. 20
1
HEALTH MAJORS AT COMMUNITY COLLEGES AND
2
FORPROFIT LESS THAN4YEAR INSTITUTIONS 3
4
Although health care fields were among the most
5
common majors at less-than-2-year and 2-year for-
profit institutions and community colleges, the mix
6
of health majors at each institution type varied.
Figure 10 shows the percentage of health majors
7
at community colleges and 2-year and less-than-2- 8
year for-profit institutions in five fields of study. To-
9
gether, these fields account for about 43 and 46
0
percent of majors at less than-2-year and 2-year
for profit institutions, respectively, and 20 percent
1
at community colleges. A higher percentage of stu-
2
dents majoring in health care fields majored in
3
nursing at community colleges (59 percent) than at
less-than-2-year (12 percent) and 2-year (1 0 per-
4
cent) for-profit institutions. By contrast, higher per-
centages of students at for-profit i nstitutions
5
majored in health and medical administrative ser-
6
vices (21 and 29 percent for less-than-2-year and
2-year institutions, respectively, vs. 5 percent at
7
community colleges) and allied health and medical 8
assisting services (21 and 27 percent for less-than-
9
2-year and 2-year institutions, respectively, vs. 4
0
percent at community colleges) .
1
6
2
7 Combining Work and Study
43
44
8 More than two-thirds of students attending for-profit
9 institutions below the bachelor's degree level and about 45
10 one-half of community college students indicated that 4
6
11 th
fi 11 b
47
. .e1r pnmaty reason or enro mg was to acqUire JO re-
12 lated skills or credentials (figure 11). Many of these stu-
13 dents work while attending classes. l n fact, regardless of
14 where undergraduates enroll , a large majority of students
15 work while enrolled (figure 12).
Community college students were the most likely to
work while enrolled (81 percent) and to work full time
(41 percent of all community college students). Howev-
er, even though a smaller percentage of students in for-
profit institutions attended part time (figure 7), nearly
three-quarters (73 percent) worked whil e enrolled and
39 percent of all for-profit students worked full time
(figure 12). Students in traditional 4-year institutions
were the least likely to work while enrolled, yet 69 per-
cent did so.
When working students were asked if they considered
employment or going to school as their primaty activity,
a majority reported the latter.
9
This was the case for stu-
dents enrolled in all three institution types. About 60
percent of students in both community colleges and for-
profit institutions reported that their primaty role was as
a student, as did 79 percent of students in 4-year institu-
tions.
Find Out More
See Related NCES Products
More detailed information on 2007-08 undergraduates
enroll ed in U.S. postsecondary institutions can be found
in Web Tables produced by NCES using the NPSAS:08
data. These web Tables are a comprehensive source of
information on undergraduate students during the
2007- 08 academic year. Included are estimates of de-
mographics, enrollment, and employment characteris-
tics. In addition, Web Tables documenting how students
pay for their undergraduate education are also avail able.
Web Tables-Profile of Undergraduate Students in U.S Postse-
condmy Institutions: 2007-08 (NCES 2010-205).
[link will be added]
9
Students who worked while enrolled were asked the foliow-
tng question: "Would you say you were primarily a student
working to meer expenses or an employee who decided ro
enroll in school. "
1 Web Tables-Student Financing of Undergraduate Education:
2 2007-08 (NCES 2010-162). [link wil l be added]
3 Readers may also be interested in the following NCES
4 prod ucts related to the topic of this Statistics in Brief:
5 Changes in Awards Below the Bachelors Degree (NCES 201 0-
6 167). 1 OL2Ql QlGZ.pdf
THE "NEW MAJORITY" OF UNDERGRADUATES: I I
STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS 7
33
34
35
36
37
38
39
pondents. Estimates were weighted to adj ust for the un-
equal probabili ty of selection into the sample and for
non response. For an overview of the survey methodolo-
gy, see appendix B of the report 2007-08 National Post-
secondary Student Aid Study (NPSAS:08): Student
FinanciaL Aid Estimates for 2007- 08: First Look
(http://nces.ed.gov /pubs2009/2009166.pdf).
7 The Price of Undergraduate Education: 2007-08 (NCES 201 0-
8 175). [link to be added] 40
T he fi ndings presented here are purely descri ptive in
nature. Al l estimates presented in this Statistics in Brief
were produced using the Data Analysis System (DAS), a
web-based software application that enables users to gen-
9 Student hnancing of Undergraduate Education: 2007-08
10 (N CES 2009-175). [li11 k to be added]
4 1
42
43
1 1 Undergraduate Financial Aid Estimates by Type oflmtitution in 44 erate tables for most of the postsecondary surveys con-
12 2007-08 (NCES 2009-201).
13 http:Unces.ed.gQvlpubs2Q09/2Q092Ql .pdf
14 2007-08 National Postsecondary Student Aid Study
15 (NPSAS:08): Student Finan,ial Aid Estimates: First Look
16 (NCES 2009-166).
17 http://nces.ed.gov/pubs2009/2009166.pdf
4 5 ducted by the National Center for Education Statistics
46 (NCES). T he DAS also contains a detailed description
47 of how each variable was created and incl udes the word-
4 8 ing of questions for variables coming d irectly from the
49 interview. With the DAS, users can repli cate or expand
50 upon the figures and tables presented in this report. For
18 Run Your Own Analysis 51 a description of al l available options, users should access
19 Direct access to the data used in this Statistics in Brief is 52 the DAS website. If users are new to the DAS, the DAS
20 avail able through NCES' s DataLab site 53 User Help Center provides online tutorials on how to
2 1 (http://nces.ed.gov/datalab/). 54 use al l functions of the DAS. The DAS is available at
22 Technical Notes
55 http://nces.ed.gov/DAS.
23 The data in this report come from the National Postse- 56 Variables Used in This Statistics in Brief
4 d d Ps S) dm d
57 Information on al l the variables used in this report can
2 condary Student Ai Stu y (N A a inistere over
25 the 2007- 08 year. T he NPSAS target population in-
58 be obtained from the NPSAS:08 DAS. Users can search
26 II d I
59 the DAS for the label or variable nan1es specified below.
dudes all eligibl e students enro e in T it eN institu-
27 tions in the United States at any time between July 1, 60 T he foll owing variables included in chis Brief had a
28 2007 and June 30, 2008.
10
The sampli ng design for col- 6
1
response race ofless than 85 percent:
29 lecting data consisted of two stages. T he fi rst stage in- 62 ATTEND MR, DEPANY, DISTED UC,
30 volved selecting eligible institutions from which students63 DISTALL, H CYSMATH, H SGPA, JOBENR,
31 would then be sampled in the second stage. Approx- 64 ]OBRO LE, PAREDUC, PCT DEP, PCTlNDEP.
32 imately 114,000 undergraduate students were study res-
10
Title IV institutions are those eligible to participate in the
federal fmancial aid programs incl uded in Title IV of the
Higher Ed ucation Act. These programs include the Pel ! Grant,
federal student loans, work-st udy, and other federal aid.
I I
THE "NEW MAJORITY" OF UNDERGRADUATES:
8 STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS
1 Age (AGE) 40 Choy, S. (200 1). Stur.knts Whose Parmts Did Not Go To Col-
2 Main reasons for attending institution (AITENDMR)
3 Attendance pattern (ATfNSTKD
4 Has dependents (DEPANY)
41
42
43
lege: PostucoruM.ry A cuss, Pnsistmu, and Attainmmt
(NCES 2001-126). Narlonal Center for Education Statis-
tics, U.S. Department of Education. Washington, DC.
5 Dependency status (DEPEND) 44 Cohen, A.M., and Brawer, F.B. (2003). The Ammcan Com-
6 Distance education: took courses in 2007-08
7 (DISTEDUC)
45 munity College {tlth ed.). San Francisco: Jossey-Bass.
8 Distance education: entire program (DIST ALL)
9 Gender (GENDER)
46
47
48
49
Horn, L. and Li, X. Changes in Awards Below the Bachelor's
Degree (NCES 2010-167). National Center for Educa-
tion Statistics, Insti tute of Ed ucation Sciences, U.S. De-
partment of Education. Washington, DC.
10 High school math courses planned/taken
J 1 (HCYSMATH)
50 Russo, A. (2006). Traditional College Students Not So Tradi-
12 High school degree type (HSDEG) 51 tiona!. This Week in Education: Alexander Russo's WeekLy
13 High school grade point average (HSGPA)
14 Work intensity whi le enrolled QOBENR)
52 Roundup of the Best Education News and Analysis. Re-
53 trieved December 16, 2009, from
15 Primary role as student or employee QOBROLE)
16 NPSAS institution level (LEVEL)
54
55
17 Field of study/major (detailed CIP codes) (MAJORS4Y)56
18 Field of study: undergraduate (10 categories) 5?
19 (MAJORS4) ; ~
20 Parent education (PAREOUC) 60
h up://tb jsweeki ned ucarion .blogspor.com/2006/02/
uad jrion a l-eo llegs:-sruden rs-no t -so. b tm I.
Swidey, N. (2009). The Four-Year College Myrh. The Boston
Globe Sunday Magazine. Retrieved November 25, 2009,
from
hnp://www.boston.com/bostonglobs:lmagazine/artkks/2
009/05/31/tbs: four year college myth/.
21 Tncomc percentile for dependent students (PCTDEP)
22
23
24
Income percentile for independent students
(PCTTNDEP)
61 Tierney, W. and Henrschke, G. {2007). New Players, Differmt
62 Game: Understanding the Rise ofFor-Profit Colleges and
63 Universities. Baltimore, MD: The j ohns Hopkins Univer-
Race/ethnicity (RACE)
64 sity Press.
25 I nstitution sector (SECTOR9)
26 Single parent (SJNGLPAR)
65 U.S. Governrnenr Accountability Office (GAO). {2009). Pro-
27 Undergraduate degree (UGDEG)
66
67
68
69
28 References
29 13erkner, L., Horn, L., and McCiw1e, M. (2000). Descriptive 70
30 Summnry of 1995-96 Beginning PostsecoruM.ry Students: 7I
31 Thm Yean Later (NCES 2000-154). NationaJ Center fol
2
32 Education Statistics, U.S. Department of Education. 73
33 Washington, DC. 74
34
35
36
37
38
39
Bradburn, E., Berger, R., Li, X., Peter, K., and Rooney, K. 75
(2003). A Descriptive Summary of1999-2000 Bachelor's 76
Degm Recipients 1 Year Later (NCES 2003-165). Nation77
al Center for Education Statistics, lnstlrute of Education 78
Sciences, U.S. Department of Education. Washington,
DC.
prietary SdJools: Stronger Department of Education Over-
sight Needed to HeLp Ensure OnLy Eligible Students Receive
FederaL Student Aid (GA0-09-600). Washington, DC:
Aurbor.
Wilson, R. (20 I 0). "for-Proflr Colleges Change Higher Edu-
cation's Landscape. Chronicle of Higher Education Febru-
ary 7, 20 I 0. Retrieved February 16, 2010 from
hnp://chronjcls:.com/arrjcls:/For-Profit-Colls:gs:s-
Changs:/640 12/.
Wei, C. (forthcoming) Web Tabks: Student Financing of Unrkr-
graduate Education {NCES 201 0-162). National Cemer
for Education Statistics, U.S. Department of Education.
Washington, DC.
1
www.ed.gov ies.ed.gov
Figure 1. WHERE UNDERGRADUATES ARE ENROLLED:
Percentage distribution of undergraduates, by type of
institution: 2007-08
Private
nonprofit
4-year
Community
colleges
44%
54%
1
Other includes public less-than-2-year and private nonprofit less-
than-4-year institutions.
2
Includes 4-, 2-, and less-than-2-year institutions.
NOTE: Estimates include students enrolled in Title IV eligible
postsecondary institutions in the 50 states, DC, and Puerto Rico.
Detail may not sum to totals because of rounding. Standard error
tables are available at http://nces.ed.gov/dasl librarv/ reports.asp.
SOURCE: U.S. Department of Education, National Center for
Education Statistics, 2007-08 National Postsecondary Student Aid
Study (NPSAS:08).
Figure 2. FOR-PROFIT INSTITUTIONS ARE AT ALL LEVELS:
Percentage distribution of undergraduates who
attended for-profit institutions, by level of institution:
2007-08
4-year
NOTE: Estimates include students enrolled in Title IV eligible post
secondary institutions in the 50 states. DC, and Puerto Rico. Detail
may not sum to totals because of rounding. Standard error tables
are available at http:lfnces.ed.gov/das/library/reports.asp .
SOURCE: U.S. Department of Education, National Center for
Education Statistics, 2007-08 National Postsecondary Student Aid
Study (NPSAS:08).
Figure 3. GENDER, FAMILY STATUS, AND RACE/ETHNCITY: Percentage of undergraduates with selected student characteristics, by type
of institution: 2007-{)8
Percent
100
80
60
40
20
For-profit'
Community
colleges
Public and private
nonprofit 4-year
0
0
69
57
Female
20
Percentage of undergraduates by selected characteristics
55
For-profit'
49
Percentage of
undergraduates with
children or dependents
Community
colleges
16
7
Percentage of
undergraduates who
were single with dependents
C Public and private
nonprofit 4-year
Percentage distribution of undergraduates by race/ethnicity
40 60 80
Percent
White OBiack CHispanic Asian Other or two or more races
1
Includes 4-, 2-, and less-than-2-year institutions.
100
NOTE: Black includes African American, Hispanic includes Latino, and Asian includes Pacific Islander. Other includes American Indian and
Alaska Native, and respondents having origins in a race/ethnicity not listed. Estimates include students enrolled in Tille IV eligible postsecondary
institutions in the 50 states, DC, and Puerto Rico. Detail may not sum to totals because of rounding. Standard error tables are available at
http:/fnces.ed.gov/das!library/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-Q8 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 4.
Percent
100
l
80
J
60
I
40
I
1
20
I
1
0
LOW-INCOME STATUS: Percentage of undergraduates who were independent and percentage in the low-income groups among
dependent and independent students, by type of institution: 2007-08
76
Independent
undergraduates
For-profit'
51
Low-income dependent
undergraduates'
Community
colleges
32
28
Low-income independent
undergraduates
2
oPublic and private
nonprofit 4-year
1
Low-income dependent undergraduates had family incomes of approximately $36.000 or less in 2006 (the year used for 2007-08 financial aid
calculations), which was at or below the 25th percentile of the income distribution for all dependent undergraduates.
2
Low-income independent undergraduates had personal (including spouse's income) incomes of approximately $11,000 or less in 2006 (the year
used for 2007-08 financial aid calculations), which was at or below the 25th percentile of the income distribution for all independent
undergraduates.
3
Includes 4-, 2-. and less-than-2-year institutions.
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Standard error
tables are available at http:/fnces.ed.gov/das/librarv/reports.asp .
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 5. AGE DISTRIBUTION: Average age and percentage distribution of undergraduates by age
and type of institution: 2007-08
Percent
100
..
80
G
60
40 36
20
0
All For-
undergraduates profit'
Average
age 26 28
1
Includes 4-, 2 , and less-than-2-year institutions.
30
Community
colleges
28
40 or older
45 0 24-29
Public and
private
nonprofit
4-year
24
20-23
19 or younger
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states,
DC, and Puerto Rico. Detail may not sum to totals because of rounding. Standard error tables are
available at http:/lnces.ed.qov/ dasllibrarv/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08
National Postsecondary Student Aid Study ( NPSAS:08).
Figure 6. STUDENTS' ACADEMIC BACKGROUND AND WHETHER FIRST IN FAMILY TO ATTEND COLLEGE: Percentage of undergraduates
who did not graduate from high school with a regular high school diploma, who reported having a high school GPA of 8 or
better, who reported taking 4 or more years of mathematics in high school, and who were the first in their family to attend
college, by type of institution: 2007-08
Percent
100
80
60
40
20
0
Students without
a regular high
school diploma'
For-profit'
81
61
54
Students with an
average high school
GPA equivalent
to grade B or better
Community colleges
76
48
54
Took 4 years or more
of mathematics
in high school
2
51
First in family to attend college
(neither parent attended
postsecondary education)
0 Public and private nonprofit 4-year
1
Undergraduates who did not have a regular high school diploma may have attended a foreign high school, been home schooled, or may have
earned a high school completion certificate, graduation equivalency diploma, or other equivalency, or have no high school degree or certificate.
2
Data on academic preparation in high school are available only for undergraduates younger than 30, who account for 77 percent of all
undergraduates.
3
Includes 4-, 2-, and less-than-2-year institutions.
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Standard error
tables are available at http://nces.ed .govfdas/libraryfreports.asp .
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007--08 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 7. FULL- AND PART-TIME ATTENDANCE: Percentage distribution of undergraduates by attendance status and by type of
i nstitution attended: 2007-08
Percent
100
80
~
a Part-time!
part-year
60
12
aPart-time/
33
full-year
14
40
Full-time/
part-year
20
11
a Full-time/
full-year
0
All undergraduates For-profit' Community Public and private
colleges nonprofit 4-year
1
Includes 4-, 2-, and less-than-2-year institutions.
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Detail may not
sum to totals because of rounding. Standard error tables are available at http://nces.ed.gov/das/library/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Postsecondary Student Aid Study
(NPSAS:08) .
Figure 8. DISTANCE EDUCATION: Percentage of all undergraduates who took a distance education course in 2007-08 and percentage of
all undergraduates who attended programs entirely taught through distance education, by institution type: 2007-08
Percent
100
80
60
40
20
0
L
21
24
17
Took a distance
education course in 2007-08
For-profit' Community colleges
1
Includes 4-, 2-, and less-than-2-year institutions.
12
3 2
Percent of undergraduates in programs
entirely taught through distance education
ClPublic and private nonprofit 4-year
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Detail may not
sum to totals because of rounding. Standard error tables are available at http://nces.ed.gov/das/ library/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Postsecondary Student Aid Study
(NPSAS:08) .
Figure 9. MAJOR FIELDS OF STUDY: Percentage distribution of undergraduates by major field of study and type of institution: 2007-08
Percent
100
80
60
40 33
20
22
20
li:n
70
6
9
6
2
..2.__o #
-
_ c:J
Health care fields Other applied' Business STEM
2
Humanities and Education General studies
social sciences and other'
For-profit
4
Community colleges a Public and private nonprofit 4-year
# Rounds to zero.
1
Includes personal and consumer services, manufacturing, construction. repair, and transportation, military technology and protective services,
architecture, communications, public administration and human services, design and applied arts, law and legal studies, and theology
and religious vocations.
2
Science, technology, engineering and math.
3
other includes basic skills, citizenship activities, leisure and recreational activities, personal awareness and self improvement, high school/
secondary diplomas and certificate programs, interpersonal and social skills.
4
Includes 4-. 2-, and less-than-2-year institutions.
NOTE: Excludes undergraduates who were not in a degree program or who had not declared a major field of study. Estimates include students
enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Detail may not sum to totals because of rounding.
Standard error tables are available at http://nces.ed.gov/das/librarv/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 10. MOST COMMON HEALTH CARE FIELDS: Percentage distribution of health majors at community colleges and for-profit
subbaccalaureate institutions by type of institution: 2007-08
Percent
100
80
60
40
20
!
0 '
57
40
11 10 10 10
14
5
26
m:;.,
22
__ _
Nursing Health
professions
and related
clinical sciences
:.J For-profit less-than-2-year
Health and
medical
administrative
services
li!l For-profit 2-year
Allied health
and medical
assisting
services
Other
health care fields'
Community colleges
1
Other health care fields include allied health diagnostic, intervention, and treatment fields, pharmacy and pharmaceutical sciences, and health
aides, among others.
NOTE: Health care fields accounted for 43 percent of majors at for-profit less-than-2-year institutions, 46 percent at for-profit 2-year institutions,
and 20 percent of majors at community colleges. Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states,
DC, and Puerto Rico. Standard error tables are available at http://nces.ed.gov/das/librarv/reports.asp .
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007--Q8 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 11. MAIN REASON FOR ENROLLI NG: Percentage distribution of undergraduates who attended community colleges and
2-year or less-than-2-year for-profit institutions, by their main reason for enrolling and type of institution: 2007-08
Percent
100
80
60
40
20
0
82
51
Acquire job-related
skills or credentials
t Not applicable.
12 12
t
Complete
associate's
degree
n For-profit less-than-2-year
1
From the institution in which they were primarily enrolled.
13
Solely for
personal interest
o For-profit 2-year
25
Prepare
to transfer
Community colleges
5
Earn course
credits at a
different
institution'
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Detail may not sum
to totals because of rounding. Standard error tables are available at http://nces.ed.govfdasllibrary/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-QS National Postsecondary Student Aid Study
(NPSAS:08).
Figure 12. COMBINING WORK AND STUDY: Percent age of all undergraduates who worked and worked full time whil e enrolled and f or
those who worked, reported their main role as being a student, by type of institution: 2007-08
Percent
100
80
60
40
20
0
81
73
69
Among all undergraduates, worked
while enrolled
For-profit
39 41
Among all undergraduates,
worked full time while enrolled
79
58 60
Of those who worked,
primary role as a student'
Community
colleges
o Public and private
nonprofit 4-year
1
Undergraduates who worked while enrolled were asked whether they regarded their primary role as a student or as an employee.
2
1ncludes 4-. 2-. and less-than-2-year institutions.
NOTE: Excludes work-study and assistantships. Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states,
DC, and Puerto Rico. Standard error tables are avai.lable at http://nces.ed.gov/ das/librarv/reports.asp .
SOURCE: U.S. Department of Education. National Center for Education Statistics, 2007--QS National Postsecondary Student Aid Study
(NPSAS:08) .
Table 1. DEGREE OR CERTIFICATE PROGRAM: Percentage distribution of undergraduates attending community colleges, for-profit,
and traditional 4-year institutions by program type: 2007-08
Associate' s Bachelor's No degree or
Program type Certificate degree degree certificate program
Public and private nonprofit 4-year 0.6 4.2 93.1
Community college 6.9 78.9 3.0
1
For-profit 34.4 38.2 25.8
Less-than-2-year 98.4
t t
2-year 35.7 61.8 0.5
4-year 5.6 42.6 50.4
t Not applicable.
1
Students who attended a community college and reported that they were working on a bachelor's degree may have attended a community
college that offers bachelor's degrees or planned to transfer to a 4-year institution.
2.1
11.2
1.6
L6
L9
1.4
NOTE: Detail may not sum to totals because of rounding. Estimates include students enrolled in Title IV eligible postsecondary institutions in the
50 states, DC, and Puerto Rico. Standard error tables are available at http:/lnces.ed.gov/das/librarv/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Postsecondary Student Aid Study
(NPSAS:OS).
Table 2. UNDERGRADUATE AWARDS CONFERRED in Title IV institutions, by type of institution: 2007
Institution type
Total
Type of institution
Total
awards
100.0
Traditional 4-year colleges and universities 53.6
Public 4-year or above 35.6
Private nonprofit 4-year and above 18.0
Public 2-year (community colleges) 28.2
Public less-than-2cyear 1 .0
Private nonprofit 2-year 0.5
Private nonprofit less-than-2-year 0.3
Private for-profit 16.4
4-year or above 4.8
2-year 5.5
Less-than-2-year 6.1
t Not applicable.
NOTE: Detail may not sum to totals because of rounding.
Percentage distribution of
undergraduate awards conferred in 2007
Associate's
Ceritifcates degree
100.0 100.0
5.1 14.2
2.9 9.2
2.2 5.0
46.2 69.1
4.3
t
1.1 0.8
1.3
t
41 .9 15.8
2.1 7.8
14.6 8.0
25.2 t
Bachelor's
degree
100.0
95.4
63.7
31 .7
t
t
t
t
4.6
4.6
t
t
SOURCE: U.S. Department of Education, National Center for Education Statistics, Integrated Postsecondary Education Data System (I PEDS),
"Completions Survey" and "Institutional Characteristics Survey," 2007.
Table A. Number of for-profit Title IV institutions, by level: United States, academic year
2004-05 to 2008-09.
-08
Total institutions, all controls 6383
6,441 6,536 6,551
Total for-profit institutions

Institutional Level
4-year 370 407 453 490 530
793 816 844 857 893
Less-than 2-vear 1318 1,34o 1,385 1,403
Change
3.9%
13.9%

1!l.6%
o.4%
SOURCE: U.S. Department of Education, National Center for Education Statistics, Integrated
Postsecondary Education Data System (!PEDS), Fall 2004to Fall 2008, Institutional
Characteristics Component
Table B. Fall
Enrollment by
institution control :
United States, Fall

Private For Profit
Private Nonprofit
Public
Total
SOURCE: U.S. Department
of Education, National
Center for Education
Statistics, Integrated
Postsecondary Education
Data System, Institutional
Characteristics and Fall
Enrollment Surveys: 2004
2008.
2008
6n,851 1,188,881 1,380,355 1, 797,563
3,137, 108 3,299,094 3,440,559 3,543,455 3,684,n3
11,891,450 12,883,071 13,081,358
15,701,409 17,710,798 19,574,395
Percent
Change

7.1%
7.7%
10. 5%
Table C. Percent
of Fall
Enrollment by
institution
control: United
States, Fall 2004

Private For Profit
Private Nonprofit
Public
SOURCE: Integrated
Postsecondary Education
Data System, Institutional
Characteristics and FaiJ
8nrollment Surveys: 2004
2008.

4.8%

75.7%

5.0%
19.4%
75.6%

6.7%
19.4%
73.9%
Percent
Change
7.6% 87%
19.5% 18.8% 8%
73.0% n.o% 3%
Table D. Number and percentage distribution of awards conferred at for-profit Title IV institutions, by level:
United States, academic year 2003-04 to 2007-08.
2003-04 2004-05 2005-06 2006-07 2007-08 Change
Total
All institutions2
2,998,114 3,085,464 3,165,016 3,232,300 3,314,978
10.6%
For-profit institutions
Number of awards2
356,128 382,146 405,542 425,377 444,307
24.8%
Percent of all awards2
11.9% 12.4% 12.8% 13.2% 13.4% 12.8%
4-year institutions}
All institutions2
2,213,225
2,293,350 2,371,219 2,443,619 2,521,319 13.9%
For-profit institutions
Number of awards2
111,586
134,081 161, 160 182,445 202,410 81.4%
Percent of all awards2 5.0% 5.8% 6.8% 7.5% 8.0% 59.2%
2-year institutionsl
All institutions2
542,177
557,172 564,964 563,875 571,964 5.5%
For-profit institutions
Number of awards2
55,080
55,259 57,461 59,381 58,415 6.1%
Percent of all awards2 10.2% 9.9% 10.2% 10.5% 10.2% 0.5%
Less-than 2-year institutions3
All institutions4
242,712
234,942 228,833 224,806 221,695 -8.7%
For-profit institutions
Number of awards4
189,462
192,806 186,921 183,551 183,482 -3.2%
Percent of all awards4 7R1% 82.1% 81.7% 81.6% 82.8% 6.0%
1 Includes degree-granting
institutions only
2 Excluding certificates
3 Includes all institutions, both
degree and non-degree granting
4 Includes all awards, including
both degrees and certificates
SOURCE: U.S. Department of Education, National Center for Education Statistics, Integrated Postsecondary
Education Data System (!PEDS), Fall 2003 to Fall 2008, Completions component.
1 The Expansion of Private Loans in
2 Postsecondary Education
3 Government officials and higher education associations are con-
4 cerned that, out of confusion or ignorance, some postsecondaty
5 students may not cake full advantage of federal loan programs and
6 turn instead to more costly private loans (e.g., Federal Trade
7 Commission 2008; King 2007). Others argue that federal loan lim-
8 its don't meet some students' education financing needs and that
9 such students, therefore, seek additional funds through private
10 loans (McSwain, Price, and Cunningham 2006).
Most students borrow money for postsecondary education through
federal loan programs, which have eligibility requirements and bor-
rowing limits. Students can also obtain private loans from banks
and other lending institutions. Private (or "alternative") loans are
not guaranteed by the government and can be relatively expensive,
as they generally have higher fees and interest rates than federal
student loans.
Private education loans are estimated to have reached a peak of
about $22 billion in 2007- 08 (College Board 2009). That same
year, many lenders increased their direct marketing to students,
highlighting a quick and easy application and approval process for
private loans; some of these lenders were accused of deceptive mar-
keting practices (U.S. Senate Committee 2007). According to the
U.S. Department of Education (2008), "[p]rivate loans and credit
cards are consumer loans and are vety expensive ways of financing
your education."
Since 2007-08, however, the volume of private loans for postse-
condaty education is thought to have declined substantially due to
a shortage of capital and higher underwriting standards by lenders
(Student Lending Analytics 2009a, 2009b). Recent data released by
I 2 I THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION
1 the College Board (2009) are consistent with this expla-
2 nation.
3 This Statistics in Brief examines how the use of private
4 education loans changed between 2003-04 and 2007-
5 08 by addressing the following study questions:
6 1. How did undergraduate borrowing from private
7 sources change between 2003-04 and 2007-08?
8 2. Who obtained private loans?
9 3. To what extent did undergraduates combine
10 private and public loans?
11 4. Did undergraduates borrow the maximum
12 amount from federal Stafford loans before turn-
13 ing to private loans?
14 5. How did private borrowing change among
15 graduate and professional students?
16 Key Findings
17 The percentage of undergraduates obtaining private
18 loans nearly tripled, from 5 to 14 percent.
19 Undergraduates at for-profit institutions had the
20
21
22
23
24
25
highest rate of borrowing from private sources-42
percent took out private loans in 2007- 08.
Dependent undergraduates from middle-income
families borrowed from private sources at higher
rates than did students from low- or high-income
families.
26 About one-half of full-time, full-year undergraduates
27 who obtained a private loan had borrowed the max-
28 imum federal (Stafford) loan amount.
29
30
MAJOR TYPES OF
HIGHER EDUCATION LOANS
Private loans. Private loans are education loans,
not guaranteed by the federal government, from
commercial lenders, credit unions, or other non-
profit entities. Their terms are determined by the
lender. Private loans carry a market interest rate,
usually variable and based on credit history, and
they generally have higher fees and interest rates
than federal student loans.(See question on p. 8)
Stafford loans. These student loans have fixed
interest rates and various repayment benefits and
are guaranteed by the federal government. They
have eligibility requirements and limits on amounts.
There are two types of federal Stafford Loans: sub-
sidized and unsubsidized. Subsidized Stafford
loans are awarded based on financial need, and
the federal government pays interest on the loan
until the student begins repayment and during au-
thorized periods of deferment thereafter. Unsubsi-
dized Stafford loans are not need based; students
are charged interest for the duration of the loan,
although it can be capitalized.
Parent PLUS loans. These federally guaranteed
loans are available only to the parents of depen-
dent students. The interest rate in 2007- 08 was
fixed at ei ther 7.9 or 8.5 percent. Borrowers cannot
have a negative credit history, and the amount is
limited to the cost of attendance minus other finan-
cial aid. The loans carry the benefits and protec-
tions of all federal loans.
Graduate PLUS loans. These are federally guar-
anteed loans for graduate and professional stu-
dents that became available in 2006. The terms are
the same as for Parent PLUS loans, with the same
interest rates, restrictions, and benefits.
THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION I 3
1 Detailed Findings 28 tutions was $5,500 in 2007-08; at private nonprofit 4-
2 How did undergraduate borrowing from private
29 year institutions, it was $17,800; and at for-profit insti-
3 sources change between 2003-Q4 and 2007-QS?
30 tutions, it was $10,200. Further, the for-profit sector is
3 1 more likely to enroll low-income students. The median
4 T he rate of undergraduate private loan borrowing in- 32 income for dependent students' parents at public 4-year
5 creased from 5 to 14 percent between 2003-04 and 33 institutions was $75,700 in 2007-08.
4
At for-profit in-
6 2007- 08 (figure 1).
1
But the amount borrowed held 34 stitutions, the corresponding median income was
7 steady after adj usting for inflation: the average private 35 $35,700.
8 loan in 2003-04 was $6,600 and $6,500 in 2007-08.
2
36 I n general, the higher the tuition, the higher the rate of
9 Stafford loan borrowing among undergraduates also in- 37 private borrowing. For example, the highest rates of bor-
1 0 creased from 32 percent to 35 percent, and the average 38 rowing (30- 32 percent) occurred among students whose
11 Stafford loan amount rose from $4,900 to $5,000. 39 tuition was more than $10,000 per year (figure 3) . In
12 The rate of any borrowing rose from 34 percent in
13 2003- 04 to 39 percent in 2007- 08.
3
T he average loan
14 amount from all sources, including Parent PLUS loans,
15 increased from $6,900 to $8, 100.
40 comparison, 22 percent of students paying $5,000 to
41 $9,999 in tuition took out private loans, as did 14 per-
42 cent of those paying $3,000 to $4,999 and 9 percent or
43 less of those paying under $3,000 in tuition.
16 Who obtained private loans?
44 Dependent undergraduates from middle-income fami li es
45 attending full-time full-year borrowed from private
17 In 2007-08, private loan borrowing varied by type of
46 sources at higher rates than students from low- or high-
18
. . . . . I 1 d d d d
4
7
tnstttuuon, tuttton eve , stu ent epen ency status, an
19 family income. T he rate of private borrowing was high-
4 8
20 est at for-profit institutions, where the rate about

2 1 from 13 percent to 42 percent, from 2003-04 to 2007-
50
22 08 (figure 2). Private borrowing at private nonprofit 4-
51
23 year institutions was also high and about doubled, from
52
24 11 percent to 25 percent. 53
income famil ies. One-fifth of studenrs in the two mid-
dle-income groups took out private loans, compared
with 15-16 percent of students in the low- and high-
income groups (figure 4). Looking at aU borrowing, in-
cluding Parent PLUS loans, dependent students from
high-income fan1Uies borrowed at the lowest rate (40
percent).
54 Among independent students, upper middle- and high-
25 T he higher rate of borrowing among students at private 55 income students took out private loans relatively more
26 institutions reflects to some extent the higher tuition at 56 often than low-income students. One-fifth of low-
27 these institutions. Average tuition at public 4-year insti- 57
income independent undergraduates took out private
58 loans, compared with 27- 28 percent of those in the up-
1 All comparisons of estimates were rested for statistical signi- 5
9
per middle- and high-income groups.
ftcance usi ng the Student's t-statistic, and all differences cited
are statistically significant at the p < .05 level. No adj ustments
for multiple comparisons were made. The standard errors for
the estimates can be found at
http://nces.ed.gov/das/li brary/reporrs.asp.
2
All dollar amounts for 2003-04 have been adj usted for infla-
tion to 2007 doll ars using the Consumer Price Index.
3
Includes Parent PLUS loans as well as Stafford, Perkins, and
private loans.
4
Independent students are age 24 or older and students under
24 who are married, have dependents, are veterans, or are an
orphan or ward of the courts. Other undergraduates under age
24 are considered to be dependent. Income for dependent
students is the income of their parents. For independent stu-
dents, income includes a spouse' s income if the st udent is mar-
ried.
I 4 I THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION
1 To what extent did undergraduates combine pri-
2 vate and public loans?
37 Did undergraduates borrow the maximum amount
38 from federal Stafford loans before turning to pri-
39 vate loans?
3 Most undergraduates who borrowed did so through a
4 federally guaranteed loan program. Some 63 percent of 40
5 undergraduates who borrowed obtained loans from pubA 1
6 lie,
5
mostly federal, sources exclusively, and another 27 42
7 percent borrowed from both public and private sources 43
8 (figme 5) . The remaining 9 percent borrowed only from44
9 private sources. Students borrowed from public and pri- 45
10 vate sources at different rates, however, depending on 46
In 2007- 08, one-half of full-time, full-year undergra-
duates exhausted their annual Stafford loan eligibility
before taking out private loans (figure 6). Independent
private loan borrowers exhausted Stafford el igibility at a
lower rate than did dependent borrowers.
Policymakers are concerned that some students seek pri-
vate loans because they are unaware of the advantages of
11 the types of institutions attended.
47 federal loan programs. They are also concerned that
48 some students may not borrow the maximum Stafford
12 Undergraduates in public 2-year institutions took out
49 loan amount before turning to private loans (U.S. De-
13 exclusively private loans at higher rates than those in oth-
50 partment of Education 2008). Applying for federal aid is
14 er types of institutions. At community colleges, 21 per-
51 necessaty to obtain federal loans. Consequently, it is use-
15 cent of borrowers took out only private loans, compared 52 ful to examine whether students did apply for federal aid
16 with 9 percent among students in all institutions. This
53 when comparing public and private borrowing.
6
17 higher rate of private borrowing, however, occurred with-
18 in the context of lower rates of borrowing overall. In 54 Among full-time, full-year undergraduates who took out
19 2007-08, some 13 percent of all community college stu-55 private loans, some 19 percent did not obtain a Stafford
20 dents borrowed from any SOLtrce, representing the lowest56 loan, and about half these students (10 percent of all
21 rate among students in all types of institutions (Wei et al57 private borrowers) did not apply for financial aid. 7
22 2009a, table 13). Some policy analysts believe that many53 Another 31 percent took out Stafford loans, but appear
23 community colleges do not encourage borrowing federal 59 to have borrowed less than the ma:ximwn amount. s The
24 loans (Cochrane, Shireman, and Sivashankaran 2008). 60 remaining 50 percent had taken out the maximum
25 The largest proportion of borrowers who took out pri-
26 vate loans either exclusively or in combination with pub-
27 lie loans (42 percent) was found among those enrolled at
28 for-profit institutions (figure 2), and those enrolled at
29 institutions with programs of 2 or more years had the
30 highest percentage who took out both private and public
31 loans (figure 5). Among those enrolled at for-profit insti-
32 tutions with programs taking 2 or more years to com-
33 plete, 45 percent of undergraduate borrowers took out
34 both private and public loans, while the overall percen-
35 cage of undergraduates who cook out both types of loans
36 at all institutions was 27 percent.
5
"Public loans" includes Stafford loans, other federal loans
(e.g., Perkins and PLUS), state, and institutional loans.
6
Foreign students and those attending part time or part yea.r
were excluded from this part of the analysis because they either
are ineligible for federal loans or qualifY for very little. Other
reasons that students cannot borrow from federal loan pro-
grams couJd not be identifled from the NPSAS data.
7
Some may have taken out other loans, such as Perkins, state,
or institutional loans, but those are comparatively rare.
8
The maximum Stafford amomu is limited for those attend-
ing part time or for less than a fttll year, so this analysis of
maximizing Stafford loans was limited to fuJI-time, full-year
students. There are other restrictions on Stafford borrowing
based on program or budget that were not taken into account.
Because of these restrictions, some borrowers who appear to be
borrowing below the maximum amount are in fact borrowing
at their maximum. Therefore, our estimates of the percentage
of borrowers who are borrowing at less than the maximum
couJd be overstated.
THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION I 5 I
1 amount allowed under the Stafford loan program. These24 points,
9
compared with an increase of9 percentage
2 students sought more money than was available from 25 points among undergraduates (figure 1 ).
3 federal programs to pay for their education expenses.
26 Graduate students differed significantly in borrowing
4 Undergraduate students have different annual Stafford 27 rates depending on the degree program in which they
5 loan limits based upon both. dependency status and class28 were enrolled (figure 7). Master's degree students' rate of
6 level. For example, for first-year undergraduates, the lim-29 total borrowing increased from 38 percent to 44 percent
7 its were $3,500 for dependent students and $7,500 for 30 over this period. The rate of private borrowing among
8 independent students. In 2007- 08, about one-third (35 31 master's degree students doubled between 2003- 04 and
9 percent) offull-time, full-year independent undergra- 32 2007-08 (from 6 percent to 12 percent), while their rate
10 duates who took out private loans also borrowed the 33 of Stafford borrowing increased from 36 percent to 39
11 maximum Stafford loan amount, compared with. 56 per-34 percent.
12 cent of dependent undergraduates who took out the
13 maximum Stafford loan amount (figure 6) . 35 Among doctoral students, changes in private, Stafford,
14
Annual Undergraduate
Stafford Loan Limits
(for loans taken out between July 1, 2007
and June 30, 2008)
36 or total borrowing were not evident. Between 5 and 7
7 percent took out private loans in both 2003- 04 and
8 2007-08.
9 Students in first-professional programs had the highest
f0 rate of private borrowing, but the proportion with pri-
Dependent Independent
Academic year student student i 1 vate loans declined over time, while borrowing from fed-
------''----------------- i2 eral sources increased.
10
In both years, students in frrst-
First year $3,500 $7,500 .
Second year 4,500 8,500 !3 professional programs borrowed from all sources at high-
Third and remaining years 5,500 10,500 i4 er rates than students in other degree progran1s. In
NOTE: Aside from dependency status and year in school, the i5 2003-04, about one-fourth (23 percent) of students in
amount a student can borrow under the Stafford loan program
can be further reduced depending on the cost of attendance, the
student's expected family contribution, attendance status,
whether the program is less than a year long, and how much
other financial aid is received.
SOURCE: U.S. Department of Education. (2007). 2007-oB
Federal Student Aid Handbook. Retrieved April 23, 2010, from
http://ifap.ed.gov/ifaplbyAwardYear.jsp?tyoe-fsahandbook&awa
i6 first-professional programs took out private loans, com-
i7 pared with 5 to 6 percent of master's and doctoral stu-
iS dents. By 2007-08, however, the rate of private
i9 borrowing among first-professional students had
;o dropped to 16 percent, while their rate of Stafford bor-
_::rd;:v::ea;:r;:=2::00=:
7
- ::2:;00:;8::,. _______________ , ; 1 rowing increased from 69 to 76 percent and one-fourth
15 How did private borrowing change among gradu-
52 (25 percent) took out Graduate PLUS loans.
16 ate and first-professional students? 53
54
17 ln 2006, graduate students became eligible for the feder-
55
18 al Graduate PLUS loan program. This gave graduate 56
19 students another source ofloans from a government-
20 guaranteed progran1 with competitive, ftxed-interest
21
22
23
rates, eliminating some of the need for private loans.
Between 2003- 04 and 2007- 08, the rate at which grad-
uate students took out private loans rose 4 percentage
Between 2003- 04 and 2007- 08, after adjusting for in-
flation, overall average loan amounts for all graduate and
first-professional students increased from $17,700 to
$18,500, largely due to the newly established Graduate
9
NPSA$:04 and NPSA$:08 Data Analysis System; data not
shown in figures or tables.
1
First professional programs include dentistry, medicine,
optometry, osteopathic medicine, pharmacy, podiatric medi-
cine, veterinaty medicine, chiropractic, law, and theology.
I 6 I THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION
1 PLUS loans. The average private loan amount for gradu-37
2 ate students decreased from $ l 0.500 to $8,400, and the
38
3 average Stafford loan amount decreased from $16, 100 to
39
Technical Notes
Dara in this report come from the National Postsecon-
dary Student Aid Study (NPSAS) administered for the
4 $15,600.
40 years 2003-04 and 2007-08. Conducted every 4 years,
5 Find Out More
4 1 NPSAS is based on a nationally representative sample of
42 al l students enroll ed in Title N -eligible institutions in
6 More detailed information on 2007-08 undergraduates 43 the United States between J uly 1 and J une 30 of a given
7 and graduate students enrolled in U.S. postsecondaty 44 academic year. The sampling design for coll ecting data
8 institutions can be found in Web Tables produced by 45 consists of two stages. The first stage involved selecting
9 the National Center for Education Statistics (NC.ES) 46 eligible institutions from which students would then be
10 using the 2007- 08 National Postsecondaty Student Aid 47 sampled in the second stage. In 2007- 08, approximately
11 Study (NPSAS:08) data. These Web Tables are a com- 48 114,000 undergraduate students were study respondents.
12 prehensive source of information on students during the49 Estimates were weighted to adjust for the unequal prob-
13 2007- 08 academic year. Included are estimates of de- 50 abili ty of selection into the sample and for nonresponse.
14 mographi cs, enroll ment, and employment characteris-
15 tics. Web Tables docw
11
enting how students pay for 51 NCES Statistical Standard 4-4-1 states that "[a]ny survey
16 their undergraduate education are also available. 52 stage of data collection wim a unit or item response rate
53 less than 85 percent must be evaluated for the potential
17 Web Tables-Profile of Undergraduate Students in US. Postse- 54 magnitude of nonresponse bias before the data or any
18 condary !mtitutiom: 2007-08 (NCES 20 I 0-205). [link 55 analysis using me data may be released" (U.S. Depart-
19 will be added)
56 ment of Education 2003). The only variable included in
20 Web Tables-Student Financing of Undergraduate Education: 57 this Bri ef that has a response rate of less than 85 percent
2 1 2007-08 (NCES 2010-162) . [link will be added) 58 is TOTLOAN.
22 Web Tables-Profile of Students in Graduate andFirst-
23 Professional Education: 2007-08 (NCES 20 10-177). [link59
For an overview of the survey memodology and a discus-
24 wil l be added] 60
sion of nonresponse bias, see appendix B of the report
6 1 2007- 08 National Postsecondary Student Aid Study
25 Readers may also be interested in the following NCES
26 products related to me topic of mis Statistics in Brief
62
63
27 2007-08 National Postsecondary Student Aid Study
28 (NPSAS:08): Student Finandal Aid Estimates: First Look 64
29 (NCES 2009-166).
30 hrrp:l/nces.ed .gov/pubs2009/2009l66.pdf
3 1 Undergraduate Financial Aid Estimates by Type of Imtitution in
32 2007-08 (NCES 2009-201).
33
34
35
36
h ttp:l/nces.ed.g,ov/pubs2009/2009201.pdf
Federal Loam in 2007-08: Undergraduate Cumulative Debt
and Students Who Bonow at the Maximum (NCES 2010-
151). [link to be added)
(NPSAS:OB): Student Financial Aid Estimates for 2007-08:
First Look (http:/ / nces.ed. gov/pubs2009/2009166.pdt).
THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION
1 References
39
40
41
2 The College Board. (2009). Trends in Student Aid. Table l.
3
4
5
Retrieved October 28, 2009, from
http://www.rrends-collegeboard.com/student aid/pdf/
2009 Trends Smdent Ald.pdf.
42
43
44
6 Cochrane, D. F., Shireman, R., and Sivashankaran, S. (2008,
7
8
9
10
11
12
13
14
15
16
17
18
19
20
2 1
22
April). Denied: Community College Students Lack Access to 45
A/fordable Loam. The Project on Smdent Debt. Retrieved46
September 9, 2009, from 47
48
http://ticas.org/ flies/pub/ denied .pdf.
Federal Trade Commission. (2008, June). Student Loam:
49
50
Avoiding Deceptive Offers. FTC Facts for Consumers. 51
Federal Trade Commission and U.S. Department of Edu-
cation. Retrieved September 9, 2009, from
h rrp://sruden taid.ed.gov/smden ts/auach menrs/
siteresources/loa nsA voidDecep. pdf.
King, J.. (2007). Who Borrows Private Loam? ACE Issue
52
53
54
55
56
Brief. Washington, DC: American Council on Education.
Retrieved October 28, 2008, fiom 57
http://www.acenet.edu/AJ\If/Template.cfm?Section-Searc58
h&seqion-issue briefs&templare-/CM/ContenrDisplay.59
cfm&ConrenrFileiD- 3445. 60
6 1
23 McSwain, C., Price, D., and Cwutingham, A. (2006). The 62
24 Future of Private Loam: Who Is Bonowing, and Why?
25 Washington, DC: lnstitute for Higher Education Policy. 63
64
26 Student Lending Analytics. (2009a, July 23). The Incredible 65
27 Shrinking Private Student Loan Mmket: Outlook for 2009"'66
28 10. Slog. Retrieved September 9, 2009, from 67
29 h up://studen tlendi nganalytic.qypepad.com/studen t lend68
30 ing analytics/2009/07/the-i ncredible-shrinking-private- 69
31 loan-market.html.
70
32 Student Lending Analytics. (2009b, August 17). Bank Lending
33 Standards May Remain Tight Through Second Half of
34 2010. Blog. Retrieved September 9, 2009, from
3 5 http://srudenrlendinganalytics. typepad.corn/smdent lend
36 ing analytics/2009/08/bank-lending-standards-may-
37 remain-tight-through-second-half-of-20 l O.htrnl.
38
U.S. Department of Education, Federal Student Aid, St udents
Charmel. (2008). Your Federal Student Loans: Learn the
Basics and Manage Your Debt. Washington, DC: Author.
U.S. Department of Education, National Center for Educa-
tion Statistics. (2003). NCES StatisticaL Standards (NCES
2003-601). Washington, DC.
U.S. Senate Committee on Banking, Housing and Urban Af-
fairs. (2007, J tme 6). Hearing on "Paying for College:
The Role of Private Student Lending." Washi ngton, DC.
Retrieved November l, 2009, from
h rtp://banki ng.senate.gov/ pub! ic/i ndex .cfm?Fuseaction-
Hearings.Hearing&Hearing ID-8c46al82-alff-4bc5-
af53-e0982b49527f.
Wei, C. C. (forthcoming). What is the Real Price of College?
Sticker, Net, and Out-of-Pocket Prices in 2007-08
(NCES 2010-175). National Center for Education Statis-
tics, Institute of Education Sciences, U.S. Department of
Education. Washington, DC.
Wei, C.C., Berkner, L. , He, S., ar1d Lew, S. (2009a). Web-
Tables: Undergraduate Financial Aid Estimates by Type of
Institution in 2007-08 (NCES 2009-201). National Cen-
ter for Education Statistics, Institute of Education
Sciences, U.S. Deparunenr of Education. Washington,
DC.
Wei, C. C., Berkner, L., He, S., Lew. S., Cominole, M., and
Siegel, P. (2009b). 2007-08 NationaL Postsecondmy Stu-
dent Aid Study (NPSAS:08): Student Financial Aid Esti-
mates for 2007-08 (flrst Look) (NCES 2009-166).
National Center for Education Statistics, Institute of
Education Sciences, U.S. Department of education.
Washington, DC.
I 8 I THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION
1 Run Your Own Analysis in Datalab 38 "During the 2007-2008 school year, did you take out
39
2 Direct access to the data used in this Statistics in Brief is
40
3 available through the NCES DataLab
any private or alternative loans from a financial institu-
tion? Some examples of commonly used private loans
include: 41
4 (http://nces.ed.gov/datalab/). You can generate your own
5 tables using one of three programs: PowerS tats, Quick- 42 Sallie Mae Signature Student Loan
6 Stats, or the Data Analysis System (DAS). 43 CitiAssist Undergraduate and Graduate Loan
7 All estimates presented in this Brief were produced using44 Chase Education One Private Student Loan
8 the DAS, a web-based software application that enables 45 Nellie Mae EXCEL Loan
9 users to generate tables for most of the postsecondary
46 Access Group Loans
10 surveys conducted by NCES. These figures were gener-
11 ated using the DAS, which contains a detailed descrip- 47 (Keep in mind that many lenders that offer private loans
12 tion of how each variable was created and includes the 48 might also offer federal Stafford, Graduate/Professional
13 wording of questions for variables coming directly from 49 PLUS loans, and Parent PLUS loans. For this question
14 the interview. 50 we want to know about private or alternative loans on-
51 ly.)"
15 You can replicate or expand upon the figures and tables
16 presented in this report by using DataLab. All variables 52 Package of private and nonprivate loans (PRJVPACK):
17 used to generate the figures and ta.bles in this Brief can
18 be found in the NPSAS:04 and NPSAS:08 DAS. More
19 information and descriptions of each variable can be
20 found at http://nces.ed.gov/DAS.
21 The variables used in this report are as follows:
53 Indicates the loan package by whether the loan received
54 was private (alternative) or not during the 2007- 08 aca-
55 demic year. For students with any loan (TOTLOAN2 >
56 0), indicates whether a borrower had only private oral-
57 ternative loans (PRIVLOAN), only nonprivate loans
58 consisting of federal loans (TFEDLN2, including
59 PLUS), state loans (STLNAMT), or institutional loans
60 (INLNAMT), or both private and nonprivate.
22 Private (alternative) loans (PRJVLOAN): Indicates the
23 amount of private or alternative loans received by stu-
24 dents for the 2003- 04 and 2007- 08 academic years.
61 Stafford total maximum (STAFCT3): Classifies the total
25 These are education loans from commercial lenders that 62 Stafford Joan amount received in 2007-08
26 are not guaJanteed by the federal government and that 63 (STAFFAMT) into categories based on the maximum
27 cany market interest rates based on credit scores. Exam- 64 Joan limits for subsidized and unsubsidized Stafford
28 pies of such loans are the Sallie Mae Signature Student 65
29
30
loans combined and includes a category for those who
didn't apply for federal aid. The normal maximum loan
an1ounts in 2007-08 for undergraduates were deter-
Loan, CitiAssist Loan, or Chase Education One Private 66
Student Loan. Private loans differ from Stafford, Parent 67
31 PLUS, Perkins, and Graduate PLUS loans, which are
32
33
68 mined by the student's undergraduate class level and
guaranteed by the federal government; private loans, 69 dependency status. This variable does not take into ac-
however, can be offered by the same lenders or by states 70 count further borrowing limits that might be imposed,
34 or other nonprofit institutions. Institutions do notal- 71 e.g., Stafford loan amounts cannot exceed a borrower's
35 ways have information on students' private or price of attendance, and subsidized Stafford loans cannot
36 loans, so this information came primarily from student 73 exceed a borrower's need amount.
37 interviews, in which students were asked the following:
1 Other Variables
2 Attendance pattern (A ITNST AT)
3 Citizenship status (CITIZEN2)
4 NPSAS institution control (CONTROL)
5 Dependency status (DEPEND)
6 Parents' income (DEPlNC)
7 Independent student's income (INDEPINC)
8 Applied for federal aid (FEDAPP)
9 Graduate PLUS loan total (GPLUSAMT)
10 Graduate degree program (GRADDEG)
11 Total federal Parent PLUS loans (PLUSAMT)
12 Type of institution (AID SECT)
13 Institution sector (SECI'OR4)
14 Federal subsidized loans (except PLUS) (SUBLOAN)
15 Total student loans from all sources (TOTLOAN)
16 Institution tuition and fees (TUITION2)
17
THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION I 9 I
www.ed.gov ies.ed.gov
Figure 1. Percentage of undergraduates who borrowed, by type of loan: 2003-04 and 2007-08
Percent
100
80
60
40
32
35
20
14
5
0
Private Stafford
2003-04 2007-08
34
39
All loans
Average undergraduate loan amounts,
by type of loan:
2003--04 and 2007-QS
2003--04 2007-QS
Private loans $6,600 $6,500
Stafford loans 4,900 5,000
All loans including PLUS 6,900 8,100
NOTE: Amounts for 2003-04 have been adjusted for inflation
using the Consumer Price Index for urban households (CPI
U). Arrounts are averages for those who received the
specified type of aid.
NOTE: Private loans are education loans from commercial lenders which are not guaranteed by t he federal government and carry market
interest rates.
SOURCE: U.S. Department of Education, Nationa.l Center for Education Statistics, 2003-04 and 2007-08 National Postsecondary Student Aid
Studies (NPSAS:04 and NPSAS:OS) .
Figure 2. Percentage of undergraduates who took out private loans, by institution type: 2003-04 and 2007-08
Percent
100
80
60
40
20
0
Public
2-year
14
Public
4-year
2003-04
25
11
Private
nonprofit
4-year
2007-08
42
13
For-profit
Average tuition and fees paid by
undergraduates, by institution type:
2007-QS
(includes full-time and part-time students)
Public 2-year
Public 4-year
Private nonprofit 4-year
For-profit
$1,200
5,500
17,800
10,200
Median income, by institution type: 2007-QB
Dependent lndepen-
students dent
parents students'
Public 2-year $54,200 $29,400
Public 4-year 75,700 22,000
Private nonprom 4-year 84,500 30,300
For-profit 35,700 19,500
1
Independent students are age 24 or older and 1udents under
24 who are married, have dependents, are veterans, or are an
orphan or ward of the oourts. Other undergraduates under age
24 are considered to be dependent. Income for dependent
students is the inoome of their parents. For independent
students. income includes a spouse's income if the student is
married.
NOTE: Private loans are education loans from commercial lenders which are not guaranteed by t he federal government and carry market
interest rates. Students who attended more than one inst itution or any type of institution not listed in the figure were not included due to the small
number of cases.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2003-04 and 2007-08 National Postsecondary Student Aid
St udies (NPSAS:04 and NPSAS:08).
Figure 3. Percentage of undergraduates who took out private loans, by
tuition amount: 2007-os
$15,000 or more
$10,000-$14,999
$5,000-$9,999
$3,000-$4,999 14
$1 ,500-$2,999
$1 ,499 or less
0 20
32
30
22
40 60
Percent
80
NOTE: Private loans are education loans from commercial lenders which
100
are not guaranteed by the federal government and carry market interest rates.
SOURCE: U.S. Department of Education. National Center for Education
Statistics, 2007- 08 National Postsecondary Student Aid St udy (NPSAS:08).
Figure 4. Percentage of fulltime, full-year undergraduates who took out private loans and any loans including Parent PLUS, by
dependency status and family income level: 2007-QS
Percent
100
80
60
54
56
61
66
69
52
40
40
27
28
21
20
0
60
Low-income Lower
middle-
income
Upper High-income
middle-
Low-income Lower
middle-
income
Upper High-income
middle-
income income
Dependent Independent
Private loans All borrowing including PLUS
NOTE: Family income categories were based upon the annual income distribution for 2006, from the lowest to the highest quartile ranges.
Among dependent students, low-income was less than $36,100: lower middle-income was $36.100-$66,600: upper middle-income was
$66,600-$104,600: and high-income was $104,600 or more. Among independent students, low-income was less than $11 ,000; lower middle-
income was $11,000-$26,000; upper middle-income was $26,000- $48,400; and high-income was $48,400 or more. Private loans are education
loans from commercial lenders which are not guaranteed by the federal government and carry market interest rates.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-()8 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 5. Percentage distribution of undergraduate borrowers, by borrowing package and institution type: 2007-QS
Percent
100
..

2
Exclusively
80 20
private
27
35 45 loans
60 Both
programs
40
Exclusively
20
public
loans
0
Total Public Public Private For-profit For-profit
2-year 4-year nonprofit less-than- 2 years
4-year 2-year or more
NOTE: Public loans includes Stafford loans, other federal loans (e.g., Perkins and PLUS), state, and institutional loans. Private loans are
education loans from commercial lenders which are not guaranteed by the federal government and carry market interest rates. Total estimates
also include those attending public less-than2-year institutions, those attending private nonprofit less-than-4-year institutions, and those
attending more than one institution but t hese are not shown separately. Detail may not sum t o totals because of rounding.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-QS National Postsecondary St udent Aid Study
(NPSAS:OS).
Figure 6. Percentage distribution of full-time, full-year undergraduates who took
out private loans, by status of Stafford borrowing, status of federal aid
application, and dependency status: 2007-QB
Percent
100
80
60
40
20
0
Total Dependent Independent
~ M a x i m u m
Stafford
DLessthan
maximum
Applied,
no Stafford
Did not
apply for
aid
NOTE: In this analysis, full-lime undergraduates excludes foreign students and
unclassified undergraduates. Private loans are education loans from commercial
lenders which are not guaranteed by the federal government and carry market
interest rates. Maximum Stafford is a loan in the amount of the annual limit for
Stafford loans based on dependency status and class level. Some restrictions on
Stafford borrowing were not taken into account when computing a student's Stafford
borrowing limit. Because of these restrictions, some borrowers who appear to be
borrowing below the maximum amount are in fact borrowing at their maximum.
Therefore, estimates of the percentage of borrowers who are borrowing at less than
the maximum may be overstated. Detail may not sum to totals because of
rounding.
SOURCE: U.S. Department of Education, National Center for Education Statistics,
2007-08 National Postsecondary Student Aid Study (NPSAS:08).
Figure 7. Percentage of graduate students who took out private, Stafford, Graduate PLUS, and all loans, by type of degree
program: 2003-04 and 2007-08
Private loans
Stafford loans
Graduate PLUS loans
Total loans
L6
~ 1 2
I 3
38
44
p,,. .. '""' ~ 5 7
Stafford loans ~
2 6
2 9
Graduate PLUS loans
Total loans
Private loans
Stafford loans
Graduate PLUS loans
Total loans
0 20
28
32
40
Percent
69
60
Average loan amounts received by graduate
students, by type of loan: 2003--{)4 and 2007-{)8
Private loans
Stafford loans
Graduate PLUS loans
Total loans
2003-Q4 2007--{)8
$10,500
16,100
17,700
$8.400
15,600
15,500
18,500
NOTE: Amounts for 2003-04 have been adjusted for inflation
using the Consumer Price Index for urban households (CPI-U).
Amounts are averages for those who received the specified
type of aid.
2003-04
2007-08
76
75
79
80 100
1
First-professional programs include dentistry, medicine, optometry, osteopat hic medicine, pharmacy, podiatric medicine, veterinary medicine,
chiropractic, law, and theology.
NOTE: Private loans are education loans f rom commercial tenders which are not guaranteed by the federal government and carry market interest
rates. Graduate PLUS loans were not available in 2003-Q4.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2003-Q4 and 2007-QS National Postsecondary Student Aid
Studies (NPSAS:04 and NPSA$:08).
U.S. DEPARTMENT OF EDUCATION APRIL 2010
What Is the Price
of College?
Sticker, Net, and
Out -of-Pocket Prices
in 2007-08
AUTHOR
Christina Chang Wei
MPR Associates, Inc.
PROJECT OFFICER
Tracy Hunt-White
National Center for Education Statistics
Students and parents see
college attendance as a principal avenue
to middle-class life, and, given the rising
price of postsecondary education, they
are apprehensive about their ability t o af-
f ord it.
1
In a recent survey of college
freshmen, about two-thirds (66 percent)
reported having concerns about being
able to fi nance their education.
2
Many U. S.
policymakers and researchers share their
concern, and are exploring ways to make
college more affordable.
3
Legislators have
required colleges and universities t o pro-
vide more extensive information about
tuition and prices, and in the 2008 Higher
Education Opportunity Act mandated a
host of price-related measures, including
institutional disclosure of net prices
charged to students, the reporting of net
price data to the U.S. Department of Edu-
cation, and the creation and disclosure of
"tuition watch lists."
4
This Statistics in Brief shows the amounts
U.S. undergraduates pay on average for
postsecondary education, wi th and wi th-
out financial aid. Drawing upon data from
the National Postsecondary Student Aid
Study, a nationally representative survey
of all postsecondary students, included in
this report are the average prices for pub-
lic 2-year, public 4-year, private nonprofit
4-year, and for-profit institutions.
1
A recent public opinion poll showed that increasing numbers of
Americans view college as a necessity for success (lmmerwahr et al.
2009). The College Board (2009) reports that the published, or stick
er, price of college has increased more rapidly than the price of other
goods and services over the past three decades. See also Marchand
(2010), which reported on these findings.
1
The Higher Education Research Institute at the University of Cali
fornia, Los Angeles, has conducted a survey of freshmen each year
since 1973 (Pryor et al. 2009). See also recent media anent ion to the
issue (Lewin 2009).
3
Both governmental and nongovernmental study commissions have
convened to address this. See, for example, The College Board
(2008) and Harvey et al. (1998).
4
Public Law 110315, title I, Sec. 132. August 14, 2008. Higher Edu
cation Opportunity Act .

1es
NATIONAL CENTER FOR
. . EDUCATION STATISTICS
STUDY QUESTIONS
What are the average
prices paid by full-time
undergraduate students
and how do these prices
vary by the type of insti-
tution attended?
Most of the figures in this Brief display
data only for full-time undergraduates
5
who attended one institution. These
students comprised about one-third
(35 percent) of all undergraduates in
2007-08.
6
Of all full-time undergra-
duates, 18 percent were enrolled in
public 2-year institutions, 43 percent
attended public 4-year institutions, 21
percent were at private nonprofit 4-
year institutions, and 8 percent were
enrolled in for-profit institutions (table
1).
7
Focusing on full-time students who
attended only one institution allows for
comparisons in tuition, price of atten-
dance, and financial aid. Those attend-
ing full time generally have higher
overall expenses than do all students.
They also generally qualify for federal
aid and other assistance not available
to many part-time students (table 2).
5
"full time" status is defined as having been enrolled full time
in one postsecondary institution for 9 months or more during
the academic year.
6
National Postsecondary Student Aid Study (NPSAS:OS) Data
Analysis System.
1
The remaining students were enrolled in other types of insti
tutions or in more than one institution during the academic
year.
What factors are re-
lated to variations in
average sticker and
net prices among
those attending these
institutions?
KEY FINDINGS
There is a wide range of prices for
postsecondary education. Students
enrolled at public 2-year institutions
had the lowest average sticker price
($12,600) while those at private
nonprofit 4-year institutions had the
highest {$35,500).
8
Those at public
4-year institutions had an average
sticker price of $18,900, and those at
for-profit institutions had an aver-
age sticker price of $28,600.
After all financial aid is received, {in-
cluding grants, loans, and work-
study), the average out-of-pocket
net price ranged from $9,100 at
public 2-year institutions to $10,300
at public 4-year institutions, $16,000
at for-profit institutions, and
$16,600 at private nonprofit 4-year
institutions.
3
All comparisons of estimates were tested for statistical signi
ficance using the Student's t-statistic, and all differences cited
are statistically significant at the p < .OS level. No adjustments
for multiple comparisons were made. The standard errors for
the estimates c a n ~ found at http://nces.ed.gov/dasf
librarv/reoorts.asp.
2
How do the net
prices paid by under-
graduates vary by
family income?
Lower sticker prices often mean a
lower need-or eligibility-for fi-
nancial aid. Those attending private
institutions had the highest tuition
but they also received the most fi-
nancial aid. Undergraduates at pri-
vate nonprofit 4-year institutions
received the greatest amount of in-
stitutional grant aid while those at
for-profit institutions had the largest
proportion of borrowers.
Among low-income undergra-
duates enrolled full time, those at-
tending for-profit institutions had
the highest average net price after
grants and average out-of-pocket
net price.
1
What are the average prices paid by full-time
undergraduate students and how do these
prices vary by the type of institution attended?
This Brief discusses three different
measures of the price of an undergra-
duate education: the sticker price, the
net price after grants, and the out-of-
pocket net price.
The Sticker Price
The sticker price is the total price of an
undergraduate education including
tuition
9
and all other nontuition and
living expenses such as books, sup-
plies, and housing. The sticker price va-
ried widely by the type of institution
attended in 2007-08, rangi ng from an
average of $12,600 among undergra-
duates enrolled full time at public 2-
year institutions, to $18,900 at public 4-
year institutions, $28,600 at for-prof it
institutions, and $35,500 at private
nonprofit 4-year institutions (figure 1 ).
9
1n this report, the term "tuition includes both tuition and
fees. Sometimes institutions treat tuition and fees as a single
charge, and sometimes as separate charges. Tuition is defined
as the price ofinstruction and fees as the price of other services
provided by the school. The tuition amounts shown here in
elude those charged to outof-state and out-of-district students
enrolled in public 4-year and public 1-year institutions, respec-
tively.
FIGURE 1.
THE PRICE OF AN UNDERGRADUATE EDUCATION
for full-time, full-year undergraduates,
by type of institution attended: 2007- 08
Price
$40,000
30,000
20,000
10,000
0
Public 2-year Public 4-year For-profit
Type of institution
Private nonprofit
4-year
Average sticker w Average net price ~ A v e r a g e out-of-
price after grants pocket net price
NOTE: The "sticker price" is the total price of attendance which includes tuition and fees, books and supplies, housing,
meals, transportation, and other miscellaneous, or personal, expenses. The "net price is the price that must be paid after
receiving grants. The "out-of-pocket net price subtracts from the sticker price all forms of financial aid, including grants,
student loans, Parent PLUS loans, work-study, employer aid, job training benefits, veterans benefits, and any other finan
cia I aid received. Both the net price after grants and the out-of-pocket net price are calculated for all full-time, full-year
students, regardless of whether or not they received any financial ald. "Full-time, full-year' is defined as having been
enrolled in one postsecondary institution for 9 months or more full time. Estimates include students enrolled in Title IV
eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Standard error tables are available at
hnpl/nces.ed.gov/das/library/reports.asp.
SOURCE: U5. Department of Education, National Center for Education Statistics, 1007-08 National Postsecondary Student
Aid Study (NPSAS:08).
3
For some institutions (private nonprofit known in financial aid parlance as the supporting families (Wei forthcoming),
4-year institutions in particular), the "student budget." Student budgets which raises their nontuition expenses
tuition is a large component of the vary according to students' attendance as compared with students who have
sticker price. In 2007-08, average tui- and dependency status, family respon- no family responsibilities.
tion was $2,400 at public 2-year institu- sibilities, and residence (e.g., living at
tions, $7,100 at public 4-year home with parents, in an on-campus Students attending public 2-year insti-
institutions, $11,900 at for-profit insti- dormitory, or in off-campus housing). tutions-mostly community colleges-
tutions, and $23,400 at private non-
profit 4-year institutions {figure 2).
In 2007-08, students enrolled full time
in for-profit institutions had the high-
Nontuition expenses, which include est average nontuition expenses
had the lowest nontuition expenses. A
larger proportion lived at home with
their parents, which reduces costs for
room and board, as compared with
students at for-profit institutions (many
of whom are supporting themselves or
their own families) and those at 4-year
institutions (where a larger proportion
lived on campus or in off-campus hous-
ing) (Wei forthcoming).
books and supplies, housing and ($16,700), when compared with those
meals, transportation, and personal at other types of institutions {where
{or miscellaneous) expenses, also can average nontuition expenses ranged
vary by institution type. College or uni- from $10,200 to $12,1 00). Many stu-
versity financial aid officers usually de- dents at for-profit institutions are fi-
velop an estimate of the sticker price, nancially independent and are
In-state vs. out-of-state tuition
At most public 4-year
institutions, tuition charges are
generally higher for out-of-state
students than for in-state
residents, reflecting the state
subsidies public institutions
receive. In 2007- 08, the average
in-state tuition was $6,200 and
the average out-of-state tuition
was $15,1 00 for full-time
undergraduates enrolled in
public 4-year institutions.*
NPSAS:OS Data Analysis System (data not shown).
FIGURE 2.
TUITION AND NONTUITION EXPENSES
for full-time, full-year undergraduates,
by type of institution attended: 2007- 08
Price
$40,000
30,000
20,000
10,000
0
Public 2-year Public 4ycar For profit
Type of institution
Private
nonprofit
4-year
tJ Average
nontuition
expenses
Average
tuition
NOTE:"full-time, full year is defined as havlng been enrolled in one postsecondary institution for 9 months or more full
time. "Tuition" includes all tuition and fees. Non tuition expenses" include books, supplies, housing, meals, transporta
tion, and miscellaneous 01 pe1>onal expenses. Estimates include students enrolled in Title IV eligible postsecondary institu-
tions in the 50 states, DC, and Puerto Rico. Standard error tables are available at
htto://nces.ed.gov/das/librarv/reoorts.aso.
SOURCE: U 5. Department of Education, National (enter for Education Statistics, 2007-08 National Postsecondary Student
Aid Study (NPSAS:OS).
4
The Net Price After Grants those at public 4-year institutions,
Policymakers and researchers generally grant aid to full-time undergraduates
subtract grants from the sticker price lowered the average sticker price of
when discussing the net price of at- $18,900 to an average net price after
tending a postsecondary institution. grants of $15,200.
For example, Congress recently re-
quired institutions to make public both For those at private institutions (both
the sticker price and the average net for-profit and nonprofit), the average
price of attendance, calculated as the net price after grants was about
total price of attendance minus all $25,700. For those attending for-profit
grants received. In this Brief, two
measures of net price are discussed:
the "net price after grants" and the
"out-of-pocket net price," which is the
price after all financial aid (i ncluding
loans, which must be repaid} is t aken
into account.
10
Grant aid helped to lower the average
st icker price among full-time under-
graduates at public 2-year institutions
from $12,600 to an average net price
after grants of $10,600 (figure 1 ). For
10
Both net price and out of-pocket net price averages are cal-
culated for all students, regardless of whether they received
any financial aid. This method of calculating net price averages
for all students differs from that used to calculate average aid
amounts in this report (see table 2). Average aid amounts are
calculated only for students receiving a particular type of aid.
Those not receiving a specific type of aid (i.e., zero values) are
not included in the average for that aid. The average grant
therefore, will be greater than the difference between the
sticker price and the net price after grants.
institutions, the difference between
the average sticker price and the net
price af ter grants was about $2,800. At
private nonprofit 4-year instituti ons,
however, that difference was $10,000.
The Out-of-Pocket Net Price
The "out-of-pocket net price," defined
as the sticker price less all fi nancial aid
received, takes into account all forms
of financial aid, including grants, loans,
work-study, and other aid (as well as
Parent PLUS loans). The out-of-pocket
net price represents the amount t hat
must be paid immediately to enroll in a
postsecondary institution f or that aca-
demic year.'' Because the out-of-
pocket net price subtracts loans from
the sticker price, it measures the net
price only in the short t erm. Loans off-
set immediate costs to students and
their families, but they must be repaid
overtime.
Full-time undergraduates enroll ed at
public 2-year institutions had the low-
est average out-of-pocket net price
($9,100}, reflecting the already lower
sticker price at these institutions (figure
1}. Those at public 4-year instituti ons
had a slightly higher average out-of-
pocket net price ($1 0,300}. Students at
private institutions had the highest av-
erage out-of-pocket net prices ($16,000
at for-profit institutions and $16,600 at
private nonprofit 4-year institutions).
11
Since 1998, the federal government has also provided postse-
condary students and their families with various federal tax
benefits. These are not included in the definition of financial aid
and are not used in the calculation of net price in this study.
2
What factors are related to variations
in average sticker and net prices among
those attending these institutions?
Prices vary by institution type for sev-
eral reasons. Institutions charge differ-
ent levels of tuition (based on whether
they are public or private and the
amount of state and local subsidies re-
ceived); the demographic characteris-
tics of students (and thus their
nontuition expenses and eligibility for
federal and state grant aid) vary by
type of institution; and institutional
policies for awarding institutional aid
differ.
As noted previously, the amount of tui-
tion charged by institutions is a prima-
ry factor in the total price of
attendance, but students also incur dif-
ferent nontuition expenses depending
on their family responsibilities and liv-
ing arrangements.
Average net prices are further affected
by differences in the proportion of aid
recipients at each type of institution.
For example, the number of students
eligible for federal Pell Grants or state-
funded grant aid at a particular school
will affect the average net price after
grants, as will a school's policies for
awarding grants from institutional
funds. The out-of-pocket net price fur-
ther depends on the level of borrowing
among students and their parents and
other types of aid received, such as
work-study.
Public 2-Year Institutions
Full-time students attending public 2-
year institutions had the lowest aver-
age sticker price ($12,600), net price af-
ter grants ($1 0,600), and out-of-pocket
net price ($9, 1 00) among all undergra-
duates (figure 1 ).
Because students at public 2-year insti-
tutions had a lower average sticker
price initially, their average net prices
also were lower-even though they
had the smallest proportions of grant
recipients (56 percent) and students
who borrowed (23 percent took out a
student loan) (table 2).
6
Public 4-Year Institutions
Tuition at public 4-year institutions was
higher than at public 2-year institu-
tions, but not as high as at the private
institutions (figure 2). Students at these
institutions also have slightly higher
nontuition expenses than those at
public 2-year institutions, with a larger
proportion living on campus or away
from home {see Wei forthcoming).
Among those enrolled full time at pub-
lic 4-year institutions, 60 percent re-
ceived grant aid, about one-half (53
percent) took out student loans, and 10
percent received work-study support
(table 2). Grant aid helped to lower the
net price after grants to an average of
$15,200, and the addition of loans,
work-study, and other aid resulted in
an average out-of-pocket net price of
$10,300 (figure 1 ). This compares to an
average out-of-pocket net price of
$9,100 at public 2-year institutions-a
difference of $1,200 in the average out-
of-pocket net price, even though the
difference in the average sticker price
was $6,300.
For-Profit Institutions
For-profit institutions are privately
owned and operated and the profits
they generate benefit individual own-
ers and shareholders. The programs
can range from less than 1 year to 4-
year bachelor's and graduate degrees.
Most undergraduates enrolled in less-
than-4-year for-profit institutions are
pursuing certificates or associate's de-
grees in occupational training pro-
grams (Staklis forthcoming).
Undergraduates at for-profit institu-
tions also tend to be older, financially
independent, and have family respon-
sibilities. This increases their nontuition
expenses and hence, their average
sticker price. More undergraduates at
for-profit institutions received federal
grants (62 percent) than did students
in any other type of institution in our
analysis (figure 3). However, a smaller
percentage of for-profit students re-
ceived state, institutional, or private
grants than students in other sectors.
On average, these students had a net
price after grants of $25,80Q-not
measurably different than that of pri-
vate nonprofit 4-year institutions, and
higher than that of public institutions
(figure 1 ).
FIGURE 3.
SOURCES OF GRANT AID
for full-time, full-year undergraduates,
by type of institution attended: 2007- 08
Percent
100
80
60
40
20
0
Public 2year Public 4-year
62
For-profit
Type of institution
67
Private nonprofit
4-year
Federal grants ~ s t a t e grant s Institutional grants n Private source grants
NOTE: "Grants" include S<holarships and tuition waivers. full-time, full-year" is defined as having been enrolled in one
postsecondary institution for 9 months or more full time. Estimates indude students enrolled in Title IV eligible postsecon-
dary institutions ln the 50 states, DC, and Puerto Rico. Standard error tables are available at
hnpJ/nces.ed.gov/dasllibrarv/reports.asp.
SOURCE: U.S. Oepartment of Education, National Center for Education Statistics, 2007- 08 National Postsecondary Student
Aid Study (NPSAS:OS).
7
Student loans were critical to reducing Private Nonprofit 4-Year Institutions next highest sticker price: for-profit in-
the average out-of-pocket net price for Even though they had the highest av- stitutions ($25,800 and $16,000, re-
al I undergraduates, but particularly for erage sticker price ($35,500) of the four spectively) (figure 1 ).
those at for-profit institutions. For-
profit institutions had the largest pro-
portion of full-time undergraduates
with at least one loan in thei r financial
aid package: 9 out of 10 (92 percent)
received an aid package containi ng a
loan (table 2 and figure 4}, compared
to 65 percent of those at private non-
profit 4-year institutions, 53 percent at
public 4-year institutions, and 23 per-
cent at public 2-year institutions. The
high level of student borrowing at for-
profit institutions resulted in an aver-
age out-of-pocket net price of $16,000
(figure 1 ).
FIGURE 4.
TYPE OF AID PACKAGE
major sectors, financial aid recipients at
private nonprofit 4-year institutions al-
so received the largest average amount
of total aid ($21, 100), when compared
with those at for-profit institutions
($13, 1 00), public 4-year institutions
($11 ,000}, and public 2-year institutions
($5,400) (table 2).
In fact, aid received by undergraduates
at private nonprofit 4-year institutions
resulted in an average net price of
$25,500 and an average out-of-pocket
net price of $16,600, both of which
were not measurably different from
undergraduates at schools with the
for full-time, full-year undergraduates, by aid package received
and type of institution attended: 2007- 08
Percent
100
80
60
40
20
0
Percent distribution receiving aid
16
Public 2ycar Public 4-year For-profit
Type of institution
Private
nonprofit
4-year
Aid package
Grants, work-
study, or other
aid, with loans
Loans only
;::: Grants, work-
study, or other
aid, without loans
No aid received
NOTE: "Grants" include scholarships and tuition waivers. "Loans" include federal, state, ins1itutional, or private student
loans, excludiog Parent PLUS loans. "Othe(' aid includes job training, veterans benefits, employer aid, and Parent PLUS
loans. "Full-time, full-year" is defined as having been enrolled In one postse<ondary institution for 9 months or more full
time. Estimates include stlldents enrolled in Title IV eligible postsecondary institutions in the 50 states, OC, and Puerto
Rico. Detail may not sum to totals because of rounding. Standard error tables are available at http://nces.ed.gov/das/
librarv/reports.asp.
SOURCE: U.S. Department of EduCiltion, National Center for Education Statistics, 2007-08 National Postsecondary Stlldent
Aid Study (NPSAS:08).
8
The receipt of institutional grants, in
particular, was critical in lowering the
price for those attending private non-
profit 4-year colleges and universities.
About two-thirds (67 percent) of stu-
dents in private nonprofit +year
schools received institutional grants or
tuition waivers, a larger proportion
than at any other type of institution (30
percent at public 4-year institutions, 17
percent at public 2-year institutions,
and 7 percent at for-profit institutions)
(f igure 3). The average institutional
grant received by those attending pri-
vate nonprofit 4-year institutions was
$10,400,
12
which helped reduce the av-
erage sticker price to an average net
price after grants of $25,500 (figure 1 ).
Work-study was also an important
source of aid to those at private non-
profit 4-year institutions. Nearly one-
third (31 percent) of all full-time un-
dergraduates at private nonprofit 4-
year institutions received work-study
aid, the highest percentage among all
full-time undergraduates (between 2
and 10 percent of undergraduates at
other types of institutions received
work-study aid) (table 2). With the aid
of student loans, work-study, and other
types of support, full-time undergra-
duates at private nonprofit 4-year insti-
tutions had an average out-of-pocket
net price of $16,600-not measurably
different from those attending for-
profit institutions ($16,000) (figure 1 ).
12
NPSAS:OS Data Analysis System (data not shown).
3
How do the net prices
paid by undergraduates
vary by family income?
Figure 5 shows the average net price
after grants among dependent under-
graduates by their family income and
type of institution attended. Among
low-income and low middle-income
dependent students, those with the
highest average net price after grants
were enrolled at for-profit institutions.
In contrast, among high middle-
income and high-income students, the
average net price af ter grants for those
at for-profit institutions was not mea-
surably different from those at private
nonprofit 4-year institutions.
FIGURE 5.
NET PRICE AFTER GRANTS BY INCOME
for full-time, full-year dependent undergraduates, by family income
category and type of institution attended: 2007-08
Price
$40,000
30,000
20,000
10,000
0
Low income
Average net price after grants
Low
middle-income
High
middle-income
Highincome
Public 2year D Public 4-year Forprofit 0 Private nonprofit 4year
NOTE: The "net price after grants" Is the price of attending a postsecondary institution after all grants have been received.
This amount is calculated by subtracting total grant aid from the sticker price. The sticker price is the total price of atten-
dance and includes tuition and fees, books and supplies, housing, meals, transportation, and other miscellaneous, or per-
sonal, expenses. Grant aid includes grants, scholarships, tuition waivers, and other gift aid that does not need to be repaid.
The net price after grants is calculated for all students, regardless of whether or notthey received any grant aid. Family
income categories were based upon parents' annual income in 2006. Dollar cutoffs are based on the distribution among all
dependent undergraduates: "low-income" was the lowest 25th percentile (less than $36, 100); "low middle-income"was
the 26tll to 50th percentile ($36, 100-$66,600); "High middle-income was the 51st to 75th percentile ($66,60Q-
$104,6001; and "High-income" was the highest 25th percentile ($104,600 or more). ' Full-time, full-yea(' is defined as
having been enrolled in one postsecondary institution for9 months or more full time. Estimates Include students enrolled
in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Standard error tables are available at
http://nct>5.ed.gov/dasllibrarv/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics. 2007-08 National Postsecondary Student
Aid Study (NPSAS:OS).
9
The average out-of-pocket net price,
on the other hand, shows a slightly dif-
ferent pattern (figure 6). After borrow-
ing, low-income and low middle-
income undergraduates enrolled at for-
prof it institutions continued to have
the highest average out-of-pocket net
price, when compared with those at
other institutions. However, among
those with incomes above t he median
(i.e., high middle-income and high-
income students), the average out-of-
pocket net price was highest for those
enrolled at private nonprofit 4-year in-
stitutions.
FIGURE 6.
OUT -OF-POCKET NET PRICE BY INCOME
for full-time, full-year dependent undergraduates,
by family il)come: 2007- 08
Price
$40,000
30,000
20,000
10,000
0
Low income
Average out-of-pocket net price
Low
middle-income
High
middle-income
Highincome
Public 2year 0 Public 4year For-profit IJ Private nonprofit 4-year
NOTE: The "out-of-pocket net price is the price of attending a post5econdaryinstitution after all aid has been received.
This amount is calculated by subtracting total financial aid from the sticker price. The sticker price is the total price of at
tendance and includes tuition and fees, books and supplies, housing, meals, transportation, and other miscellaneoos, or
personal, expenses. Financial aid includes grants, student loans, Parent PLUS loans, work-study, employer aid, job training
benefi\5, veterans benefits, and any other type of aid.lhe ootof-pocket net price is calculated for all students, regardless
of whether or not they received any aid. Family income categories were based upon parents' annual income in 2006. Dollar
cutoffs are based on the distribution among all dependent undergraduates: "low-income was the lowest 25th percentile
(less than $3-6,100); "low middle-income was the 26th to 50th percentile ($36, 106-$66,600); "High middle-income"
was the 51st to 75th peJcentile ($66,600-$104,600); and''Higlrincome was the highest 25th percentile ($104,600 or
more). "Full-time, full-year" is defined as having been enrolled in one postsecondary instirution for 9 months or more full
time. Estimates include studen\5 enrolled in Trtle IV eligible postsecondary institutions in the 50 states, OC, and Puerto
Rico. Standard error tables are available at htto:l/nm.ed.gov/das/librarv/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Post5econdary Student
Aid Study (NPSAS:OS).
10
SUPPLEMENTAL TABLES
TABLE 1.
UNDERGRADUATE ENROLLMENT, TUITION, AND STICKER PRICE
for all undergraduates and full-time, full-year undergraduates, by type of
institution attended: 2007- 08
All undergraduates Full-time,
Percent Average Percent Average
distribution Average sticker distribution Average sticker
enrolled tuition' price' enrolled tuition' price
1
Total 100.0 $5,800 $14,000 100.0 $10,300 $22,400
Type of institution
Public 2-year 40.0 1,200 7,000 17.8 2,400 12,600
Public 4-year 29.2 5,500 15,200 42.7 7,100 18,900
Private nonprofit 4-year 13.0 17,800 28,200 20.6 23,400 35,500
For-profit 9.0 10,200 20,600 8.5 11,900 28,600
Other, or more than
one institution 8.8 4,800 12,300 10.5 7,000 18,000
TABLE 2.
FINANCIAL AID
for all undergraduates and full-time, full-year undergraduates,
by type of institution attended: 2007- 08
1
Average tuition and average sticker price estimates
are shown for those attending one lnstiwtlon only.
NOTE: "full-time, full-year is defined as having been
enrolled in a postsecondary institution for 9 months
or more full time. "Tuition includes all tllition and
fees. "Stkker price" is thetotal price of attendance
which includes tllition and fees, books and supplies,
housing, meals, transportation, and other miscella-
neous, or personal, expenses. Estimates include stu-
dents enrolled in Title IV eligible postsecondary
institutions in the 50 states, DC, and Puerto Rico. De-
tail may not sum to totals because of rounding. Stan
dard error tables are available at

SOURCE: U.S. Department of Education, National Cen-
ter for Education Statistics, 2007- 08 National Postse
condary Student Ai d Study (NPSAS:OS).
Any grants Any loans Any work-study Any aid NOTE:"Grants" include scholarships
Average Average Average Average
and tuition waiver;. "Loans" Include
Percent amount Percent amount Percent amount Percent amount
federal, state, institutional, or private
student loans, excluding Parent PLUS
loans. rota!" aid includes grants,
All undergraduates 51.7 $4,900 38.5 $7,100 7.4 $2,400 65.6 $9,100
loans, job training, veterans benefits,
employer aid, and Parent PLUS loans.
"All undergraduates" include those
Type of institution
attending more than one institution.
"Full-time, full-yea(' is defined as
Public 2-year 39.6 2,200 13.2 4,100 3.3 3,000 47.6 3,400 having been enrolled in one postse-
Public4-year 52.9 5,200 46.2 6,600 7.7 2,500 71.3 9.400
condary institution for 9 months or
more full time. Average aid amounts
Private nonprofit 4-year 73.6 10,200 58.9 9,100 23.2 2,100 84.7 17,400 are calculated only for students re-
For-profit 70.4 3,200 91.6 8,100 2.0 3,500 96.3 10,800
ceiving a particular type of aid. Those
not receiving a specific type of aid
(i.e., zero values) are not included in
Full-time, full-year 65.3 7,200 53.1 8,000 13.7 2,300 80.1 12,900
theaverageforthat aid. Estimates
include students enrolled in mle IV
eligible pnstsecondary institutions in
Type of institution
the 50 states, DC, and Puerto Rico.
Standard error tables are available at
Public 2-year 55.7 3.700 22.5 4,900 6.9 2,600 65.7 5,400
Public4-year 60.4 6,100 52.7 7,100 10.5 2,400 78.3 11,000

SOURCE: U.S. Oepartmem ofEducation,
Private nonprofit 4-year 81.2 12,300 65.0 9,800 31.5 2,100 89.4 21,100 National Ceoter for fducation Statistics,
For-profit 71.9 4,000 91.6 9,600 2.0 3,600 96.7 13,100
2007-08 National Postsecondary Stu
dent Aid Study (NPSAS:08).
1l
FIND OUT MORE
For questions about content, ordering additional copies of this Statistics in Brief,
or to view this report online, go to:
http://nces.ed.gov/insert-url/
More detailed information on the price of undergra-
duate education and undergraduate financing can be
found in Web Tables produced by the National Center
for Education Statistics (NCES) using the 2007-08 Na-
tional Postsecondary Student Aid Study (NPSAS:08)
data. These Web Tables are a comprehensive source of
information on financial aid awarded to undergra-
duate students during the 2007-08 academic year. In-
cluded are estimates of tuition, price of attendance,
and financial aid. Additional information on the de-
mographic characteristics of 2007- 08 undergraduates
can be found in a second set of Web Tables.
Web Tables-Student Financing of Undergraduate Edu-
cation: 2007-08 (NCES 2010-162). (link will be added)
Web Tables-Profile of Undergraduate Students in U.S.
Postsecondary Institutions: 2007- 08 (NCES 2010-205).
(link will be added)
Readers may also be interested in the following
NCES products related to the topic of this Statistics in
Brief:
2007-08 National Postsecondary Student Aid Study
(NPSAS:08): Student Financial Aid Estimates: First Look
(NCES 2009-166).
http://nces.ed.gov /pu bs2009 /2009166.pdf
Undergraduate Financial Aid Estimates by Type of Insti-
tution in 2007-08 (NCES 2009-201).
http://nces.ed.gov/pubs2009/2009201.pdf
-------------------------------------- 12 --------------------------------------
TECHNICAL NOTES
Survey Methodology
The statistics provided in this Statistics
in Brief are based on data collected
through the 2007- 08 National Postse-
condary Student Aid Study (NPSAS:08).
NPSAS:08 is the seventh administration
of NPSAS, which has been conducted
every 3 to 4 years since 1986-87. The
NPSAS:08 target population includes
all eligible students enrolled in Title IV
institutions in the United States at any
time between July 1, 2007 and June 30,
2008, representing about 21 million
undergraduates and 3 million graduate
students enrolled in over 6,000 institu-
tions.
The institution sampling frame for
NPSAS:08 was constructed from the
2004-05 and 2005-061nstitutional
Characteristics, Fall Enrollment, and
Completions files of the Integrated
Postsecondary Education Data System
(IPEDS). The sampling design consisted
of first selecting eligible institutions,
from which students were sampled in
the second stage. Institutions were se-
lected with probabilities proportional
to a composite measure of size based
on expected enrollment. Approximate-
ly 1,700 institutions participated in the
study, yielding an unweighted re-
sponse rate of 89 percent. Approx-
imately 114,000 undergraduates and
14,000 graduate students were study
respondents, yielding an unweighted
response rate of 96 percent. Estimates
were weighted to adjust for the un-
equal probability of selection into the
sample and for nonresponse.
NPSAS covers broad topics concerning
student enrollment in postsecondary
education and how students, and their
families, finance their education. Res-
pondents provided data through in-
struments administered over the
Internet, by telephone, or in person. In
addition to respondent-provided data,
data on respondents were collected
from other relevant databases, includ-
ing U.S. Department of Education
records on student loan and grant pro-
grams and student financial aid appli-
cations.
VARIABLES USED
Two broad categories of error occur in
estimates generated from surveys:
sampling and nonsampling errors.
Sampling errors occur when observa-
tions are based on samples rather than
on entire populations. Nonsampling
errors can be attributed to several
sources: incomplete information about
all respondents (e.g., some students or
institutions refused to participate, or
students participated but answered
only certain items); differences among
respondents in question interpretation;
inability or unwillingness to give cor-
All estimates presented in this Statistics in Brief were produced using the
Data Analysis System (DAS). See "Run Your Own Analysis" for more informa-
tion on the DAS and other web-based software applications that enable us-
ers to generate tables for most of the postsecondary surveys conducted by
NCES. The program files that generated the statistics presented here can be
found at [insert-uri-here}. The variables used in these analyses include the
following:
Label
Attendance status
Family income for dependent students
Federal grants
Institutional grants
Net price after grants
Non-tuition expenses
Out-of-pocket net price
Private grants
State grants
Sticker price
Total financial aid
Total grants
Total student loans
Tuition
Type of aid package
Type of institution
Work-study
Name
ATINSTAT
PGDEP
TFEDGRT
INGRTAMT
NETCST3
BUDNONAJ
NETCSTl
PRIVAID
STGTAMT
BUDGETAJ
TOT AID
TOTGRT
TOTLOAN
TUITION2
AIDTYPE
SEGOR4
TOTWKST
-------------------------------------- 13 --------------------------------------
rect information; mistakes in recording
or coding data; and other errors of col-
lecting, processing, sampling, and im-
puting missing data.
For more information on NPSA$:08 me-
thodology, see 2007-08 National Post-
secondary Student Aid Study (NPSAS:08)
Full-scale Methodology Report (NCES
201 0-1 88) (forthcoming).
Item Response Rates
NCES Statistical Standard 4-4-1 states
that " [a)ny survey stage of data collec-
tion with a unit or item response rate
less than 85 percent must be evaluated
for the potential magnitude of nonres-
ponse bias before the data or any anal-
REFERENCES
The College Board (2009). Trends in
College Pricing 2009. Retrieved on
February 11,2010, from
http://www.trends-col legeboard.
com/college pricing/pdf/
2009 Trends College Pricing.pdf.
The College Board (2008). Fulfilling the
Commitment: Recommendations for
Reforming Federal Student Aid: TheRe-
port from the Rethinking Student Aid
Study Group. Retrieved on January 15,
201 0, from .!:!.llQ.;LL
professionals.collegeboard.com/
profdownload/rethinking-stu-aid-
fulfilling-commitment-
recom mendations.pdf.
Harvey, J., Williams, R.M., Kirshstein, RJ.,
O'Malley, A.S., and Wellman, J.V.
( 1998). Straight Talk About College
Costs and Prices. Report of the National
Commission on the Cost of Higher Edu-
cation. Phoenix, AZ: Oryx Press.
ysis using the data may be released"
(U.S. Department of Education 2003).
Several derived variables used in this
report were comprised of variables that
had response rates of less than 85 per-
cent before imputation. EMPLYAM1
and EMPL YAM2 (measures of employer
aid received, and components of
TOTGRT) had response rates of 79 per-
cent and 80 percent, respectively. Non-
response bias analyses showed that for
EMPL YAM 1, there was statistically sig-
nificant bias on 87 percent of the cha-
racteristics analyzed, and for
EMPLYAM2, there was significant bias
on 76 percent of all characteristics.
PRIVLOAN (private loans received, and
lmmerwahr, J., Johnson, J., Gasbarra, P.,
Ott, A., and Rochkind, J. (2009).
Squeeze Play 2009: The Public's Views
on College Costs Today. New York and
San Jose, CA: Public Agenda and The
National Center for Public Policy and
Higher Education.
Lewin, T. (2009, October 20). College
Costs Keep Rising, Report Says. The
New York Times.
Marchand, A. (201 0, January 21 ). Cost of
College Is a Big Worry of Freshmen in
National Survey. The Chronicle of High-
er Education.
Pryor, J.H., Hurtado, S., DeAngelo, L., Pa-
lucki Blake, l., and Tran, S. (2009). The
American Freshman: National Norms
Fall2009. Los Angeles: Higher Educa-
tion Research Instit ute, University of
California, Los Angeles.
a component ofTOTLOAN) had a re-
sponse rate of 67 percent. The nonres-
ponse bias analysis for this variable
showed that respondent s and nonres-
pondents were significantly different
on 80 percent of all analyzed characte-
ristics. PCTDEP had a response rate of
56 percent, and the non response bias
analysis also showed significant bias on
80 percent of all characteristics. For
more detailed information on nonres-
ponse bias analysis and an overview of
the survey methodology, see 2007-08
National Postsecondary Student Aid
Study (NPSAS:08) Full-scale Methodology
Report (NCES 201 0-188) (forthcoming).
Staklis, S. (forthcoming). Web Tables-
Profile of Undergraduate Students:
2007-08 (NCES 2010-205). National
Center for Education Statistics, Insti-
tute of Education Sciences, U.S. De-
partment of Education. Washington,
DC.
U.S. Department of Education, National
Center for Education Statistics. (2003).
NCES Statistical Standards (NCES 2003-
601 ). Washington, DC.
Wei, C. (forthcoming). Web Tables-
Student Financing of Undergraduate
Education: 2007- 08 (NCES 201 0-162).
National Center for Education Statis-
tics, Institute of Education Sciences,
U.S. Department of Education. Wash-
ington, DC.
-------------------------------------- 14 --------------------------------------
RUN YOUR OWN ANALYSIS WITH DATALAB
You can replicate or expand upon the figures and tables in this report, or
even create your own. Data lab has several different tools that allow you to
customize and generate output from a variety of different survey datasets.
Visit Data Lab at:
QuickS tats
Create a simple table
quickly
View )'O'Jr cutp-..rt as a
chart or table
Choose from many data
sets each with about one
hundred Y!lriables
Select from
postsecondary surmes
Cover artwork iStockphoto.<Om!centauria.
http:/ /nces.ed.gov/datalab/
Dotolab
PowerS tats
Produee- complex tables
Run linear and logistic

Choose from many data
sets each with thousands
of variabl&s
library
Search existing tables and
figures to find to your
questions Vistt NCES Tables
sod Figures
Comr119 soon Thousands of
pubhshed tables using
Po,verStats and OuickStats
Help
Need Ccntact
John Vuicka
MPR Assoc.iatas Inc.
d3tala!rhelp@mpnrte com
Aurora D'Amico. NCES
Au.ora gcw
Need access to restncted
data? Learn more on The
Restnctedll se Data
Procedures Manual
-------------------------------------- 15 --------------------------------------
www.ed.gov ies.ed.gov
(b){5)
G
From: Kanter, Martha
Sent Friday, April 30, 2010 2:49PM
To: Yuan, Georgia
Subject FW: Tony Miller briefing
l(b)(5)
From: Bergeron, David
Sent: Friday, April 30, 2010 2:37PM
To: Kanter, Martha
Subject FW: Tony Miller briefing
From: Weko, Tom
Sent: Friday, April 30, 2010 2:35PM
To: Bergeron, David
Cc: Miller, Elise; Hunt-White, Tracy
Subject: Tony Miller briefing
(b)(5)
From: Yuan, Georgia
To.: Cunningham,Peter
CC:
Date: 4/30/2010 3:11 :12PM
Subject: FW: Tony Miller briefing
(b)(5)
(b)(5)
(b)(5)
.................................................................................................................................................................................................................................................................................................
(b)(5)
1 The "New Majority" of Undergraduates:
2 Students in Community Colleges and
3 For-Profit Institutions
4 Introduction
5 Depictions of postsecondaty education in the news, films, and oth-
6 er media often focus on undergraduates who study at traditional
7 4-year public universities and nonprofit colleges (Swidey 2009;
8 Russo 2006; U.S. Department of Education 2002). Yet even two
9 decades ago these students accounted for just 55 percent of stu-
l 0 dents enrolled in U.S. postsecondaty education (Choy and Gifford
1990).
1
Now a majority of undergraduates in U.S. postsecondaty
education attend community colleges and for-profit institutions.
These two sectors together enrolled 54 percent of the approximate-
ly 21 million undergraduates who attended U.S. postsecondary
institutions in 2007-08. Some 44 percent of undergraduates at-
l6 tended community colleges, the single largest sector of U.S. postse-
condary education, and another 10 percent attended for-profit
institutions (figure 1). While the for-profit sector accounts for a
relatively small percentage of undergraduates, its enrollment is
growing at a rapid pace relative to other postsecondaty sectors
(Horn and Li 2009; Wilson 2010).
Who are the undergraduate students served by community colleges
and for-profit institutions, and how do they compare to students
who are enrolled at traditional 4-year institutions? Drawing on data
from a nationally representative sample of undergraduates enrolled
L Based on fall enrollment in the 1986-87 National Postsecondary Stu-
dent Aid Study (NPSAS:87).
THE "NEW MAJORITY" OF UNDERGRADUATES:
STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS
1 in U.S. postseconda1y institutions, this Statistics in Brief27 enrolled in bachelor's degree programs, while 43 percent
2 addresses the following questions. 28 were pursuing associate's degrees, and the remaining stu-
29 dents were either in certificate programs or not in a de-
3 How do students enrolled in for-profit institutions and 30 gree program (table 1).2 The average 2007-08 tuition
4 community colleges differ from each other and from 31 and fees for full-time, full-year students was $11,900,
5 those enrolled in traditional4-year institutions in terms 32 and for all students, $10,200 (Wei forthcoming).
6 of:
7
8
9
10
11
12
13
33 Community colleges: Also known as public 2-year institu-
1) demographic and socioeconomic characreristics?34 tions, community colleges offer associate's degrees in
2) indicators of high school academic preparation 35 occupational and academic fields, as well as occupational
and being the first in their family to attend col- 36 certificates. Unlike for-profit institutions, which offer
lege? 37 primarily occupational programs, community colleges
38 offer a variety of progran1s with different goals. These
3) enrollment patterns and major fields of study?
39 include academic transfer to 4-year institutions, voca-
4) the extent to which they combine work and
study?
40 tionalltechnical education, continuing education, and
41 developmental education (Cohen and Brawer 2003).
14 The three groups of postsecondary institutions comparcJ2 Conu11w1ity colleges, therefore, serve a range of students,
15 in this Brief are briefly defined below. 43 from those needing basic remedial courses to very high
44 achieving students meeting lower-division bachelor's
16 For-profit institutions: A recent report by the U.S. Gov- 45 degree requirements before transferring to a selective 4-
17 ernment Accountability Office defined propriety institu-46 year college or university (Provasnik and Planty 2008).
18 tions as "Institutions that are privately-owned whose net47 The average 2007-08 tuition and fees for full-time, full-
19 earnings can benefit a shareholder or individual; that is, 48 year students was $2,400, and for all students, $1,200
20 for-profit institutions." (U.S. GovernmentAccountabili-49 (WeiforthcominiJ.
21 ty Office 2009). Among undergraduates enrolled in for-
22 profit institutions in 2007-08, about half attended insti-50
23 tutions authorized to award 4-year (bachelor) degrees, 27
24 percent attended 2-year institutions, and 22 percent at-
25 tended less-than-2-year institutions (figure 2). Half of
26 those in for-profit 4-year institutions were
2
At community colleges, 79 percent of undergraduates were in
associate's degree programs, 7 percent in cerrillcate programs,
and 11 percent were not in a degree program. In traditional 4-
year institutions, 93 percent of lmdergraduates were enrolled
in bachelor's degree programs.
THE "NEW MAJORITY" OF UNDERGRADUATES: I I
STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS 3
1 Public and private nonprofit 4-year institutions: Public 28
2 and private nonprofit institutions serve as the main 29
3 comparison group in this analysis. Consistent with pre- 30
4 vious research, they are referred to as "traditional4-year 31
5 institutions" (Tierney and Hentschke 2007). Together, 32
6 students in these two sectors accounted for 46 percent of
33
7 all undergraduates in 2007-08-32 percent attending
34
8 public 4-year colleges and universities, and 14 percent
35
9 attending private nonprofit 4-year institutions (figure 1)
36
ic), and income (larger proportion oflow-
income students).
3
Community college students
also differ from traditional 4-year college stu-
dents on most of these characteristics, but to a
lesser extent.
4
The part-time status of the majority of commu-
nity coll ege students sets them apart from their
peers in for-profit and traditional 4-year institu-
tions, the majority of whom attend full time.
10 The average 2007-08 tuition and fees for full-time, full-
11 year students in 4-year public institutions was $5,700
37
12 and $6,300 and for all students, $4,100 and $6,300, at 38
Health care fields are the most common fields of
study among students in for-profit institutions,
and arean1ong the most common fields for
community college students.
13 non-doctorate- and doctorate-granting institutions, re- 39
14 spectively. Tuition and fees in private nonprofit institu- 40
15 tions totaled $20, 800 and $25,800 for full-time, full- 41
Regardless of where undergraduates enroll, a
majority work while enrolled.
16 year students and $15,200 and $20,700 for all students, 42
17 at non-doctorate- and doctorate-granting institutions,
18 respectively (Wei forthcoming).
19
20
UNDERGRADUATE AWARDS CONFERRED
Table 2 shows the distribution of undergraduate
awards for the institution comparison groups
(shaded) presented in this report. These comple-
tions data are reported by all institutions that partic-
ipate in Title IV student aid programs. Based on
these data, community colleges awarded 46 per-
cent of all certificates and 69 percent of associate's
degrees in 2007. For-profit institutions awarded 42
percent of all certificates and 16 percent of asso-
ciate's degrees. For-profit 4-year institutions
awarded about 5 percent of bachelor's degrees
and traditional 4-year colleges and universities
awarded the remaining 95 percent.
21 Key findings
43 Demographic and Socioeconomic
44 Characteristics
5 Many demographic characteristics, including gender,
6 race/ethnicity, family status, income and age distinguish
7 students enrolled in for-profit institutions and commu-
8 nity colleges from those enrolled in traditional 4-ycar
9 colleges. Students in the for-profit sector, for example,
0 were the most likely to be female, Black or Hispanic, to
1 have children or other dependents, and to be single with
2 dependents in 2007-08 (figure 3). Women constituted a
3 majority of undergraduates in all three sectors, but com-
4 prised 69 percent of students in for-profit institutions. fn
5 contrast, 57 percent of community college students and
6 55 percent of traditional4-year institution students were
7 women.
22 Students enrolled in for-profit institutions differ
3
All comparisons of estimates were tested for statistical signi-
ficance using the Student's !-statistic, and all differences cited
are statistically significant at the p < .05 level. No adjustments
for mltltiple comparisons were made. The standard errors for
the estimates can be found at
http://nces.ed.goy/das/library/rs:ports.asp.
23 from those enrolled in traditional 4-year institu-
24 tions across a wide range of demographic and
25 socioeconomic characteristics, including gender
26 (larger proportion of women), age (older),
27 race/ethnicity (nearly half are Black or Hispan-
4
Determined by a Student' s t-srarisric that rests the signillc-
ance of the difference of differences.
THE "NEW MAJORITY" OF UNDERGRADUATES:
STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS
1 Nearly half of the students in for-profit institutions were35 students were 31 percent for dependent students and 22
2 Black (25 percent) or Hispanic (21 percent). In commu-36 percent for independent students.
3 nity colleges, the percentages of Black (14 percent) and
4 Hispanic students (15 percent) were lower than those in 37 The income status of community college students rela-
5 for-profit institutions but higher than those in tradition-38 tive to their peers in traditional 4-year institutions dif-
6 al4-year institutions (1 1 and 12 percent, respectively, 39 fered for dependent and independent students. A larger
7 were Black or Hispanic).
40 percentage of dependent community college students
4 1 were from low-income families (31 vs. 20 percent), while
8 Community college students also fel l between for-profit 42 a smaller percentage of independent students earned low
9 and traditional 4-year institutions in terms of their like- 43 incomes (22 vs. 28 percent) .
10 lihood of having dependents and being single with de-
11 pendents. Some 32 percent of communi ty college
44 The average age of students enrolled at conummity col-
12 students had children or other dependents, versus 49 anJ5 leges and for-profit institutions was older than the aver-
13 14 percent in for-profit and traditional 4-year institu-
14
15
tions, respectively. The same pattern was found for the
percentage of undergraduates who were single with de-
16 pendents (16 vs. 31 and 7 percent) .
46 age age of students at traditional 4-year institutions (28
47 vs. 24) (figure 5) . Although the average age of students
48 in both for-profit institutions and communi ty coll eges
49 was 28, the distri bution of students by age group in each
50 type of institution differed (figure 5) . Undergraduates
17 When income is considered for postsecondaty students, 51 enroll ed in for-profit institutions were least likely to be
18 family income is reported for students considered finan- 52 traditional-aged students-23 or younger.
19 cially dependent on their parents, and student (and
20 spousal) income is reported for chose considered finan- 53
2 1 cially independent. Dependency status is based largely 54
22 on age: all students 24 or older are considered indepen- 55
23 dent. Married students and students who are parents are
6
24 also considered independent. Reflecting these characte-
5
25 ristics, just over three-quarters (76 percent) of students
57
26 in for-profit institutions were independent (figure 4). I n ~ ~
27 contrast, 57 percent of community college students and
60
28 32 percent of students at traditional4-year institutions
61
29 were independent.
62
30 Students in the for-profit sector were the most likely to
31
32
63
64
be low income; roughly half (51 percent) of dependent
students had low family incomes, and roughJy one-third 65
33 of independent students earned low incomes (32 per-
66
Indicators of High School Academic
Preparation and First in Family to Attend
College
Students in for-profit institutions appeared to be the
least academically prepared based on indicators of high
school completion and academic preparation available in
the data. Indicators include high school completion sta-
tus (standard diploma), average grades, and the number
of years of mathematics taken. As illustrated in figure 6,
compared with their peers in community colleges and
traditional 4-year institutions, students in for-profit in-
stitutions were the most likely co have enrolled in postse-
condaty education without a standard hi gh school
diploma (19 percent, vs. 14 percent and 7 percent, re-
34 cent) .s Comparable percentages for conununity college 67 spectively, of community college and traditional4-year
5
Low income is defmed as a parent (dependent wldergra-
duates) or student (including spouse's income, for indepen-
dent wldergraduates) income that is at or below the 25th
percentile of the income distribution for each group. In 2006
(the year used for 2007-08 financial ajd applications) the in-
come levels were at or below $36,000 for dependent students
and at or below $11,000 for independent students.
THE "NEW MAJORITY" OF UNDERGRADUATES: I
STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS 5
1 college students) .
6
They were also least likely to report 30 in for-profit institutions attended full time for the full
2 earning grades ofB or higher (54 vs. 6 1 and 81 percent),31 year and 33 percent attended full time for part of the
3 or completing 4 or more years of high school mathemat-32 year. A higher percentage of undergraduates attended
4 ics (48 vs. 54 and 76 percent) ? At the same time, about 33 full time, full year in traditional4-year institutions (59
5 half of for-profi t students were the first in their famil y to34 percent) than in for-profit institutions.
6 enroll in postsecondary education, as measured by the
7 level of education achieved by either parent (figure 6). 35 Not ali students attend classes on campuses, but instead
8 SpecificaHy, among for-profi t students, 51 percent were 36 participate in distance education. Students in communi-
9 from families in which neither parent had an education 37 ty coll eges and for-profit institutions di d so more often
10 beyond high school, compared with 39 and 25 percent, 38 chan their peers at traditional4-year colleges. About one-
11 respectively, of community coll ege and tradi tional 4-yea
1
39 quarter of community college students and about one-
12 college students. 40 fifth of students in for-profi t institutions reported taking
41 distance education courses during the 2007-08 academic
13 Enrollment Patterns and Major Fields of
14 Study
42 year (figure 8) . In contrast, 17 percent of students at tra-
43 di tional 4-year institutions reported the same. In addi-
44 cion, a larger percentage offor-proflt enrollees (12
15 The part-time enrollment patterns of conumrnity collegC45 percent) than those at commttnity colleges (3 percent)
16 students distinguish them from their peers in both for- 46 reported that their entire program was offered through
17 profi t and traditiona14-year institutions (figure 7) . Se- 47 distance education.
18 venty percent attended part time, while the same percen-
19 cage of students in traditional 4-year and for-profit 4 8 With respect to fields of study, community college stu-
20 institutions attended full time. 49 dents sought both academic and occupational prepara-
50 cion (figure 9) . For example, 7 percent of community
2 1
22
23
While part-time attendance offers a flexibility often
5 1 college students and 22 percent of students in traditional
needed by studenrs attending community colleges (Co- 52 4-year institutions majored in humanities and social
wen and Brower 2003) , it is also associated with lower 53 sciences, compared with 2 percent of students in for-
24 rates of degree completion relative to full-time atten-
25 dance (e.g., Berkner, He, and Cataldi 2002).
54 profit institutions. Another 20 percent at community
55 coll eges and 6 percent at tradi tional 4-year institutions
56 majored in general studies and other fields, compared
26 Looking at attendance patterns over the year, in
57
27 2007-08, some 38 percent of community college stu-
58
28 dents attended part time for part of the year and 32 per-
59
29 cent for the full year.
8
By contrast, 37 percent of
with less chan 1 percent at for-profit institutions. Com-
munity coll ege students majoring in humanities, social
sciences, and general studies are largely those who seek to
transfer to a 4-year college (Berkner, Horn, and Chrne
6 1 2000).
6
These students may have dropped out of high school, ob-
tained aGED or high school equivalency certificate, been 62 Health care fields accounted for one-third of undergra-
home schooled, or attended a foreign high school. 63 duace majors in for-profit institutions and about one-
7 Data on high school academic preparation (GPA and
courses) is available only for students younger than 30, which 64 fifth (21 percent) in community coll eges. In conuast, 9
accowued for 77 percent of alltmdergraduares (figure 5). 65 percent of students in traditional 4-year institutions ma-
8 A full year is defined as 9 or more months. Those attending 66 jored in health care fields. l n addition to health care
part year may have completed a program of study or left with-67 fields, students in for-profit institutions were the most
out completing, either temporari ly or permanently.
68 likely to major in fields that are grouped under the head-
I I
THE "NEW MAJORITY" OF UNDERGRADUATES:
6 STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS
1 ing "other applied" (26 vs. 16 and 14 percent at com- 16
2 munity coll eges and tradi t ional 17
3 4-year institutions, respectively). These fields include 18
4 criminal justice, communications, and design and ap- 19
5 plied arts, among others. 20
1
HEALTH MAJORS AT COMMUNITY COLLEGES AND
2
FORPROFIT LESS THAN4YEAR INSTITUTIONS 3
4
Although health care fields were among the most
5
common majors at less-than-2-year and 2-year for-
profit institutions and community colleges, the mix
6
of health majors at each institution type varied.
Figure 10 shows the percentage of health majors
7
at community colleges and 2-year and less-than-2- 8
year for-profit institutions in five fields of study. To-
9
gether, these fields account for about 43 and 46
0
percent of majors at less than-2-year and 2-year
for profit institutions, respectively, and 20 percent
1
at community colleges. A higher percentage of stu-
2
dents majoring in health care fields majored in
3
nursing at community colleges (59 percent) than at
less-than-2-year (12 percent) and 2-year (1 0 per-
4
cent) for-profit institutions. By contrast, higher per-
centages of students at for-profit i nstitutions
5
majored in health and medical administrative ser-
6
vices (21 and 29 percent for less-than-2-year and
2-year institutions, respectively, vs. 5 percent at
7
community colleges) and allied health and medical 8
assisting services (21 and 27 percent for less-than-
9
2-year and 2-year institutions, respectively, vs. 4
0
percent at community colleges) .
1
6
2
7 Combining Work and Study
43
44
8 More than two-thirds of students attending for-profit
9 institutions below the bachelor's degree level and about 45
10 one-half of community college students indicated that 4
6
11 th
fi 11 b
47
. .e1r pnmaty reason or enro mg was to acqUire JO re-
12 lated skills or credentials (figure 11). Many of these stu-
13 dents work while attending classes. l n fact, regardless of
14 where undergraduates enroll , a large majority of students
15 work while enrolled (figure 12).
Community college students were the most likely to
work while enrolled (81 percent) and to work full time
(41 percent of all community college students). Howev-
er, even though a smaller percentage of students in for-
profit institutions attended part time (figure 7), nearly
three-quarters (73 percent) worked whil e enrolled and
39 percent of all for-profit students worked full time
(figure 12). Students in traditional 4-year institutions
were the least likely to work while enrolled, yet 69 per-
cent did so.
When working students were asked if they considered
employment or going to school as their primaty activity,
a majority reported the latter.
9
This was the case for stu-
dents enrolled in all three institution types. About 60
percent of students in both community colleges and for-
profit institutions reported that their primaty role was as
a student, as did 79 percent of students in 4-year institu-
tions.
Find Out More
See Related NCES Products
More detailed information on 2007-08 undergraduates
enroll ed in U.S. postsecondary institutions can be found
in Web Tables produced by NCES using the NPSAS:08
data. These web Tables are a comprehensive source of
information on undergraduate students during the
2007- 08 academic year. Included are estimates of de-
mographics, enrollment, and employment characteris-
tics. In addition, Web Tables documenting how students
pay for their undergraduate education are also avail able.
Web Tables-Profile of Undergraduate Students in U.S Postse-
condmy Institutions: 2007-08 (NCES 2010-205).
[link will be added]
9
Students who worked while enrolled were asked the foliow-
tng question: "Would you say you were primarily a student
working to meer expenses or an employee who decided ro
enroll in school. "
1 Web Tables-Student Financing of Undergraduate Education:
2 2007-08 (NCES 2010-162). [link wil l be added]
3 Readers may also be interested in the following NCES
4 prod ucts related to the topic of this Statistics in Brief:
5 Changes in Awards Below the Bachelors Degree (NCES 201 0-
6 167). 1 OL2Ql QlGZ.pdf
THE "NEW MAJORITY" OF UNDERGRADUATES: I I
STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS 7
33
34
35
36
37
38
39
pondents. Estimates were weighted to adj ust for the un-
equal probabili ty of selection into the sample and for
non response. For an overview of the survey methodolo-
gy, see appendix B of the report 2007-08 National Post-
secondary Student Aid Study (NPSAS:08): Student
FinanciaL Aid Estimates for 2007- 08: First Look
(http://nces.ed.gov /pubs2009/2009166.pdf).
7 The Price of Undergraduate Education: 2007-08 (NCES 201 0-
8 175). [link to be added] 40
T he fi ndings presented here are purely descri ptive in
nature. Al l estimates presented in this Statistics in Brief
were produced using the Data Analysis System (DAS), a
web-based software application that enables users to gen-
9 Student hnancing of Undergraduate Education: 2007-08
10 (N CES 2009-175). [li11 k to be added]
4 1
42
43
1 1 Undergraduate Financial Aid Estimates by Type oflmtitution in 44 erate tables for most of the postsecondary surveys con-
12 2007-08 (NCES 2009-201).
13 http:Unces.ed.gQvlpubs2Q09/2Q092Ql .pdf
14 2007-08 National Postsecondary Student Aid Study
15 (NPSAS:08): Student Finan,ial Aid Estimates: First Look
16 (NCES 2009-166).
17 http://nces.ed.gov/pubs2009/2009166.pdf
4 5 ducted by the National Center for Education Statistics
46 (NCES). T he DAS also contains a detailed description
47 of how each variable was created and incl udes the word-
4 8 ing of questions for variables coming d irectly from the
49 interview. With the DAS, users can repli cate or expand
50 upon the figures and tables presented in this report. For
18 Run Your Own Analysis 51 a description of al l available options, users should access
19 Direct access to the data used in this Statistics in Brief is 52 the DAS website. If users are new to the DAS, the DAS
20 avail able through NCES' s DataLab site 53 User Help Center provides online tutorials on how to
2 1 (http://nces.ed.gov/datalab/). 54 use al l functions of the DAS. The DAS is available at
22 Technical Notes
55 http://nces.ed.gov/DAS.
23 The data in this report come from the National Postse- 56 Variables Used in This Statistics in Brief
4 d d Ps S) dm d
57 Information on al l the variables used in this report can
2 condary Student Ai Stu y (N A a inistere over
25 the 2007- 08 year. T he NPSAS target population in-
58 be obtained from the NPSAS:08 DAS. Users can search
26 II d I
59 the DAS for the label or variable nan1es specified below.
dudes all eligibl e students enro e in T it eN institu-
27 tions in the United States at any time between July 1, 60 T he foll owing variables included in chis Brief had a
28 2007 and June 30, 2008.
10
The sampli ng design for col- 6
1
response race ofless than 85 percent:
29 lecting data consisted of two stages. T he fi rst stage in- 62 ATTEND MR, DEPANY, DISTED UC,
30 volved selecting eligible institutions from which students63 DISTALL, H CYSMATH, H SGPA, JOBENR,
31 would then be sampled in the second stage. Approx- 64 ]OBRO LE, PAREDUC, PCT DEP, PCTlNDEP.
32 imately 114,000 undergraduate students were study res-
10
Title IV institutions are those eligible to participate in the
federal fmancial aid programs incl uded in Title IV of the
Higher Ed ucation Act. These programs include the Pel ! Grant,
federal student loans, work-st udy, and other federal aid.
I I
THE "NEW MAJORITY" OF UNDERGRADUATES:
8 STUDENTS IN COMMUNITY COLLEGES AND FOR-PROFIT INSTITUTIONS
1 Age (AGE) 40 Choy, S. (200 1). Stur.knts Whose Parmts Did Not Go To Col-
2 Main reasons for attending institution (AITENDMR)
3 Attendance pattern (ATfNSTKD
4 Has dependents (DEPANY)
41
42
43
lege: PostucoruM.ry A cuss, Pnsistmu, and Attainmmt
(NCES 2001-126). Narlonal Center for Education Statis-
tics, U.S. Department of Education. Washington, DC.
5 Dependency status (DEPEND) 44 Cohen, A.M., and Brawer, F.B. (2003). The Ammcan Com-
6 Distance education: took courses in 2007-08
7 (DISTEDUC)
45 munity College {tlth ed.). San Francisco: Jossey-Bass.
8 Distance education: entire program (DIST ALL)
9 Gender (GENDER)
46
47
48
49
Horn, L. and Li, X. Changes in Awards Below the Bachelor's
Degree (NCES 2010-167). National Center for Educa-
tion Statistics, Insti tute of Ed ucation Sciences, U.S. De-
partment of Education. Washington, DC.
10 High school math courses planned/taken
J 1 (HCYSMATH)
50 Russo, A. (2006). Traditional College Students Not So Tradi-
12 High school degree type (HSDEG) 51 tiona!. This Week in Education: Alexander Russo's WeekLy
13 High school grade point average (HSGPA)
14 Work intensity whi le enrolled QOBENR)
52 Roundup of the Best Education News and Analysis. Re-
53 trieved December 16, 2009, from
15 Primary role as student or employee QOBROLE)
16 NPSAS institution level (LEVEL)
54
55
17 Field of study/major (detailed CIP codes) (MAJORS4Y)56
18 Field of study: undergraduate (10 categories) 5?
19 (MAJORS4) ; ~
20 Parent education (PAREOUC) 60
h up://tb jsweeki ned ucarion .blogspor.com/2006/02/
uad jrion a l-eo llegs:-sruden rs-no t -so. b tm I.
Swidey, N. (2009). The Four-Year College Myrh. The Boston
Globe Sunday Magazine. Retrieved November 25, 2009,
from
hnp://www.boston.com/bostonglobs:lmagazine/artkks/2
009/05/31/tbs: four year college myth/.
21 Tncomc percentile for dependent students (PCTDEP)
22
23
24
Income percentile for independent students
(PCTTNDEP)
61 Tierney, W. and Henrschke, G. {2007). New Players, Differmt
62 Game: Understanding the Rise ofFor-Profit Colleges and
63 Universities. Baltimore, MD: The j ohns Hopkins Univer-
Race/ethnicity (RACE)
64 sity Press.
25 I nstitution sector (SECTOR9)
26 Single parent (SJNGLPAR)
65 U.S. Governrnenr Accountability Office (GAO). {2009). Pro-
27 Undergraduate degree (UGDEG)
66
67
68
69
28 References
29 13erkner, L., Horn, L., and McCiw1e, M. (2000). Descriptive 70
30 Summnry of 1995-96 Beginning PostsecoruM.ry Students: 7I
31 Thm Yean Later (NCES 2000-154). NationaJ Center fol
2
32 Education Statistics, U.S. Department of Education. 73
33 Washington, DC. 74
34
35
36
37
38
39
Bradburn, E., Berger, R., Li, X., Peter, K., and Rooney, K. 75
(2003). A Descriptive Summary of1999-2000 Bachelor's 76
Degm Recipients 1 Year Later (NCES 2003-165). Nation77
al Center for Education Statistics, lnstlrute of Education 78
Sciences, U.S. Department of Education. Washington,
DC.
prietary SdJools: Stronger Department of Education Over-
sight Needed to HeLp Ensure OnLy Eligible Students Receive
FederaL Student Aid (GA0-09-600). Washington, DC:
Aurbor.
Wilson, R. (20 I 0). "for-Proflr Colleges Change Higher Edu-
cation's Landscape. Chronicle of Higher Education Febru-
ary 7, 20 I 0. Retrieved February 16, 2010 from
hnp://chronjcls:.com/arrjcls:/For-Profit-Colls:gs:s-
Changs:/640 12/.
Wei, C. (forthcoming) Web Tabks: Student Financing of Unrkr-
graduate Education {NCES 201 0-162). National Cemer
for Education Statistics, U.S. Department of Education.
Washington, DC.
1
www.ed.gov ies.ed.gov
Figure 1. WHERE UNDERGRADUATES ARE ENROLLED:
Percentage distribution of undergraduates, by type of
institution: 2007-08
Private
nonprofit
4-year
Community
colleges
44%
54%
1
Other includes public less-than-2-year and private nonprofit less-
than-4-year institutions.
2
Includes 4-, 2-, and less-than-2-year institutions.
NOTE: Estimates include students enrolled in Title IV eligible
postsecondary institutions in the 50 states, DC, and Puerto Rico.
Detail may not sum to totals because of rounding. Standard error
tables are available at http://nces.ed.gov/dasl librarv/ reports.asp.
SOURCE: U.S. Department of Education, National Center for
Education Statistics, 2007-08 National Postsecondary Student Aid
Study (NPSAS:08).
Figure 2. FOR-PROFIT INSTITUTIONS ARE AT ALL LEVELS:
Percentage distribution of undergraduates who
attended for-profit institutions, by level of institution:
2007-08
4-year
NOTE: Estimates include students enrolled in Title IV eligible post
secondary institutions in the 50 states. DC, and Puerto Rico. Detail
may not sum to totals because of rounding. Standard error tables
are available at http:lfnces.ed.gov/das/library/reports.asp .
SOURCE: U.S. Department of Education, National Center for
Education Statistics, 2007-08 National Postsecondary Student Aid
Study (NPSAS:08).
Figure 3. GENDER, FAMILY STATUS, AND RACE/ETHNCITY: Percentage of undergraduates with selected student characteristics, by type
of institution: 2007-{)8
Percent
100
80
60
40
20
For-profit'
Community
colleges
Public and private
nonprofit 4-year
0
0
69
57
Female
20
Percentage of undergraduates by selected characteristics
55
For-profit'
49
Percentage of
undergraduates with
children or dependents
Community
colleges
16
7
Percentage of
undergraduates who
were single with dependents
C Public and private
nonprofit 4-year
Percentage distribution of undergraduates by race/ethnicity
40 60 80
Percent
White OBiack CHispanic Asian Other or two or more races
1
Includes 4-, 2-, and less-than-2-year institutions.
100
NOTE: Black includes African American, Hispanic includes Latino, and Asian includes Pacific Islander. Other includes American Indian and
Alaska Native, and respondents having origins in a race/ethnicity not listed. Estimates include students enrolled in Tille IV eligible postsecondary
institutions in the 50 states, DC, and Puerto Rico. Detail may not sum to totals because of rounding. Standard error tables are available at
http:/fnces.ed.gov/das!library/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-Q8 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 4.
Percent
100
l
80
J
60
I
40
I
1
20
I
1
0
LOW-INCOME STATUS: Percentage of undergraduates who were independent and percentage in the low-income groups among
dependent and independent students, by type of institution: 2007-08
76
Independent
undergraduates
For-profit'
51
Low-income dependent
undergraduates'
Community
colleges
32
28
Low-income independent
undergraduates
2
oPublic and private
nonprofit 4-year
1
Low-income dependent undergraduates had family incomes of approximately $36.000 or less in 2006 (the year used for 2007-08 financial aid
calculations), which was at or below the 25th percentile of the income distribution for all dependent undergraduates.
2
Low-income independent undergraduates had personal (including spouse's income) incomes of approximately $11,000 or less in 2006 (the year
used for 2007-08 financial aid calculations), which was at or below the 25th percentile of the income distribution for all independent
undergraduates.
3
Includes 4-, 2-. and less-than-2-year institutions.
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Standard error
tables are available at http:/fnces.ed.gov/das/librarv/reports.asp .
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 5. AGE DISTRIBUTION: Average age and percentage distribution of undergraduates by age
and type of institution: 2007-08
Percent
100
..
80
G
60
40 36
20
0
All For-
undergraduates profit'
Average
age 26 28
1
Includes 4-, 2 , and less-than-2-year institutions.
30
Community
colleges
28
40 or older
45 0 24-29
Public and
private
nonprofit
4-year
24
20-23
19 or younger
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states,
DC, and Puerto Rico. Detail may not sum to totals because of rounding. Standard error tables are
available at http:/lnces.ed.qov/ dasllibrarv/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08
National Postsecondary Student Aid Study ( NPSAS:08).
Figure 6. STUDENTS' ACADEMIC BACKGROUND AND WHETHER FIRST IN FAMILY TO ATTEND COLLEGE: Percentage of undergraduates
who did not graduate from high school with a regular high school diploma, who reported having a high school GPA of 8 or
better, who reported taking 4 or more years of mathematics in high school, and who were the first in their family to attend
college, by type of institution: 2007-08
Percent
100
80
60
40
20
0
Students without
a regular high
school diploma'
For-profit'
81
61
54
Students with an
average high school
GPA equivalent
to grade B or better
Community colleges
76
48
54
Took 4 years or more
of mathematics
in high school
2
51
First in family to attend college
(neither parent attended
postsecondary education)
0 Public and private nonprofit 4-year
1
Undergraduates who did not have a regular high school diploma may have attended a foreign high school, been home schooled, or may have
earned a high school completion certificate, graduation equivalency diploma, or other equivalency, or have no high school degree or certificate.
2
Data on academic preparation in high school are available only for undergraduates younger than 30, who account for 77 percent of all
undergraduates.
3
Includes 4-, 2-, and less-than-2-year institutions.
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Standard error
tables are available at http://nces.ed .govfdas/libraryfreports.asp .
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007--08 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 7. FULL- AND PART-TIME ATTENDANCE: Percentage distribution of undergraduates by attendance status and by type of
i nstitution attended: 2007-08
Percent
100
80
~
a Part-time!
part-year
60
12
aPart-time/
33
full-year
14
40
Full-time/
part-year
20
11
a Full-time/
full-year
0
All undergraduates For-profit' Community Public and private
colleges nonprofit 4-year
1
Includes 4-, 2-, and less-than-2-year institutions.
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Detail may not
sum to totals because of rounding. Standard error tables are available at http://nces.ed.gov/das/library/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Postsecondary Student Aid Study
(NPSAS:08) .
Figure 8. DISTANCE EDUCATION: Percentage of all undergraduates who took a distance education course in 2007-08 and percentage of
all undergraduates who attended programs entirely taught through distance education, by institution type: 2007-08
Percent
100
80
60
40
20
0
L
21
24
17
Took a distance
education course in 2007-08
For-profit' Community colleges
1
Includes 4-, 2-, and less-than-2-year institutions.
12
3 2
Percent of undergraduates in programs
entirely taught through distance education
ClPublic and private nonprofit 4-year
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Detail may not
sum to totals because of rounding. Standard error tables are available at http://nces.ed.gov/das/ library/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Postsecondary Student Aid Study
(NPSAS:08) .
Figure 9. MAJOR FIELDS OF STUDY: Percentage distribution of undergraduates by major field of study and type of institution: 2007-08
Percent
100
80
60
40 33
20
22
20
li:n
70
6
9
6
2
..2.__o #
-
_ c:J
Health care fields Other applied' Business STEM
2
Humanities and Education General studies
social sciences and other'
For-profit
4
Community colleges a Public and private nonprofit 4-year
# Rounds to zero.
1
Includes personal and consumer services, manufacturing, construction. repair, and transportation, military technology and protective services,
architecture, communications, public administration and human services, design and applied arts, law and legal studies, and theology
and religious vocations.
2
Science, technology, engineering and math.
3
other includes basic skills, citizenship activities, leisure and recreational activities, personal awareness and self improvement, high school/
secondary diplomas and certificate programs, interpersonal and social skills.
4
Includes 4-. 2-, and less-than-2-year institutions.
NOTE: Excludes undergraduates who were not in a degree program or who had not declared a major field of study. Estimates include students
enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Detail may not sum to totals because of rounding.
Standard error tables are available at http://nces.ed.gov/das/librarv/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 10. MOST COMMON HEALTH CARE FIELDS: Percentage distribution of health majors at community colleges and for-profit
subbaccalaureate institutions by type of institution: 2007-08
Percent
100
80
60
40
20
!
0 '
57
40
11 10 10 10
14
5
26
m:;.,
22
__ _
Nursing Health
professions
and related
clinical sciences
:.J For-profit less-than-2-year
Health and
medical
administrative
services
li!l For-profit 2-year
Allied health
and medical
assisting
services
Other
health care fields'
Community colleges
1
Other health care fields include allied health diagnostic, intervention, and treatment fields, pharmacy and pharmaceutical sciences, and health
aides, among others.
NOTE: Health care fields accounted for 43 percent of majors at for-profit less-than-2-year institutions, 46 percent at for-profit 2-year institutions,
and 20 percent of majors at community colleges. Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states,
DC, and Puerto Rico. Standard error tables are available at http://nces.ed.gov/das/librarv/reports.asp .
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007--Q8 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 11. MAIN REASON FOR ENROLLI NG: Percentage distribution of undergraduates who attended community colleges and
2-year or less-than-2-year for-profit institutions, by their main reason for enrolling and type of institution: 2007-08
Percent
100
80
60
40
20
0
82
51
Acquire job-related
skills or credentials
t Not applicable.
12 12
t
Complete
associate's
degree
n For-profit less-than-2-year
1
From the institution in which they were primarily enrolled.
13
Solely for
personal interest
o For-profit 2-year
25
Prepare
to transfer
Community colleges
5
Earn course
credits at a
different
institution'
NOTE: Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Detail may not sum
to totals because of rounding. Standard error tables are available at http://nces.ed.govfdasllibrary/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-QS National Postsecondary Student Aid Study
(NPSAS:08).
Figure 12. COMBINING WORK AND STUDY: Percent age of all undergraduates who worked and worked full time whil e enrolled and f or
those who worked, reported their main role as being a student, by type of institution: 2007-08
Percent
100
80
60
40
20
0
81
73
69
Among all undergraduates, worked
while enrolled
For-profit
39 41
Among all undergraduates,
worked full time while enrolled
79
58 60
Of those who worked,
primary role as a student'
Community
colleges
o Public and private
nonprofit 4-year
1
Undergraduates who worked while enrolled were asked whether they regarded their primary role as a student or as an employee.
2
1ncludes 4-. 2-. and less-than-2-year institutions.
NOTE: Excludes work-study and assistantships. Estimates include students enrolled in Title IV eligible postsecondary institutions in the 50 states,
DC, and Puerto Rico. Standard error tables are avai.lable at http://nces.ed.gov/ das/librarv/reports.asp .
SOURCE: U.S. Department of Education. National Center for Education Statistics, 2007--QS National Postsecondary Student Aid Study
(NPSAS:08) .
Table 1. DEGREE OR CERTIFICATE PROGRAM: Percentage distribution of undergraduates attending community colleges, for-profit,
and traditional 4-year institutions by program type: 2007-08
Associate' s Bachelor's No degree or
Program type Certificate degree degree certificate program
Public and private nonprofit 4-year 0.6 4.2 93.1
Community college 6.9 78.9 3.0
1
For-profit 34.4 38.2 25.8
Less-than-2-year 98.4
t t
2-year 35.7 61.8 0.5
4-year 5.6 42.6 50.4
t Not applicable.
1
Students who attended a community college and reported that they were working on a bachelor's degree may have attended a community
college that offers bachelor's degrees or planned to transfer to a 4-year institution.
2.1
11.2
1.6
L6
L9
1.4
NOTE: Detail may not sum to totals because of rounding. Estimates include students enrolled in Title IV eligible postsecondary institutions in the
50 states, DC, and Puerto Rico. Standard error tables are available at http:/lnces.ed.gov/das/librarv/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Postsecondary Student Aid Study
(NPSAS:OS).
Table 2. UNDERGRADUATE AWARDS CONFERRED in Title IV institutions, by type of institution: 2007
Institution type
Total
Type of institution
Total
awards
100.0
Traditional 4-year colleges and universities 53.6
Public 4-year or above 35.6
Private nonprofit 4-year and above 18.0
Public 2-year (community colleges) 28.2
Public less-than-2cyear 1 .0
Private nonprofit 2-year 0.5
Private nonprofit less-than-2-year 0.3
Private for-profit 16.4
4-year or above 4.8
2-year 5.5
Less-than-2-year 6.1
t Not applicable.
NOTE: Detail may not sum to totals because of rounding.
Percentage distribution of
undergraduate awards conferred in 2007
Associate's
Ceritifcates degree
100.0 100.0
5.1 14.2
2.9 9.2
2.2 5.0
46.2 69.1
4.3
t
1.1 0.8
1.3
t
41 .9 15.8
2.1 7.8
14.6 8.0
25.2 t
Bachelor's
degree
100.0
95.4
63.7
31 .7
t
t
t
t
4.6
4.6
t
t
SOURCE: U.S. Department of Education, National Center for Education Statistics, Integrated Postsecondary Education Data System (I PEDS),
"Completions Survey" and "Institutional Characteristics Survey," 2007.
Table A. Number of for-profit Title IV institutions, by level: United States, academic year
2004-05 to 2008-09.
-08
Total institutions, all controls 6383
6,441 6,536 6,551
Total for-profit institutions

Institutional Level
4-year 370 407 453 490 530
793 816 844 857 893
Less-than 2-vear 1318 1,34o 1,385 1,403
Change
3.9%
13.9%

1!l.6%
o.4%
SOURCE: U.S. Department of Education, National Center for Education Statistics, Integrated
Postsecondary Education Data System (!PEDS), Fall 2004to Fall 2008, Institutional
Characteristics Component
Table B. Fall
Enrollment by
institution control :
United States, Fall

Private For Profit
Private Nonprofit
Public
Total
SOURCE: U.S. Department
of Education, National
Center for Education
Statistics, Integrated
Postsecondary Education
Data System, Institutional
Characteristics and Fall
Enrollment Surveys: 2004
2008.
2008
6n,851 1,188,881 1,380,355 1, 797,563
3,137, 108 3,299,094 3,440,559 3,543,455 3,684,n3
11,891,450 12,883,071 13,081,358
15,701,409 17,710,798 19,574,395
Percent
Change

7.1%
7.7%
10. 5%
Table C. Percent
of Fall
Enrollment by
institution
control: United
States, Fall 2004

Private For Profit
Private Nonprofit
Public
SOURCE: Integrated
Postsecondary Education
Data System, Institutional
Characteristics and FaiJ
8nrollment Surveys: 2004
2008.

4.8%

75.7%

5.0%
19.4%
75.6%

6.7%
19.4%
73.9%
Percent
Change
7.6% 87%
19.5% 18.8% 8%
73.0% n.o% 3%
Table D. Number and percentage distribution of awards conferred at for-profit Title IV institutions, by level:
United States, academic year 2003-04 to 2007-08.
2003-04 2004-05 2005-06 2006-07 2007-08 Change
Total
All institutions2
2,998,114 3,085,464 3,165,016 3,232,300 3,314,978
10.6%
For-profit institutions
Number of awards2
356,128 382,146 405,542 425,377 444,307
24.8%
Percent of all awards2
11.9% 12.4% 12.8% 13.2% 13.4% 12.8%
4-year institutions}
All institutions2
2,213,225
2,293,350 2,371,219 2,443,619 2,521,319 13.9%
For-profit institutions
Number of awards2
111,586
134,081 161, 160 182,445 202,410 81.4%
Percent of all awards2 5.0% 5.8% 6.8% 7.5% 8.0% 59.2%
2-year institutionsl
All institutions2
542,177
557,172 564,964 563,875 571,964 5.5%
For-profit institutions
Number of awards2
55,080
55,259 57,461 59,381 58,415 6.1%
Percent of all awards2 10.2% 9.9% 10.2% 10.5% 10.2% 0.5%
Less-than 2-year institutions3
All institutions4
242,712
234,942 228,833 224,806 221,695 -8.7%
For-profit institutions
Number of awards4
189,462
192,806 186,921 183,551 183,482 -3.2%
Percent of all awards4 7R1% 82.1% 81.7% 81.6% 82.8% 6.0%
1 Includes degree-granting
institutions only
2 Excluding certificates
3 Includes all institutions, both
degree and non-degree granting
4 Includes all awards, including
both degrees and certificates
SOURCE: U.S. Department of Education, National Center for Education Statistics, Integrated Postsecondary
Education Data System (!PEDS), Fall 2003 to Fall 2008, Completions component.
1 The Expansion of Private Loans in
2 Postsecondary Education
3 Government officials and higher education associations are con-
4 cerned that, out of confusion or ignorance, some postsecondaty
5 students may not cake full advantage of federal loan programs and
6 turn instead to more costly private loans (e.g., Federal Trade
7 Commission 2008; King 2007). Others argue that federal loan lim-
8 its don't meet some students' education financing needs and that
9 such students, therefore, seek additional funds through private
10 loans (McSwain, Price, and Cunningham 2006).
Most students borrow money for postsecondary education through
federal loan programs, which have eligibility requirements and bor-
rowing limits. Students can also obtain private loans from banks
and other lending institutions. Private (or "alternative") loans are
not guaranteed by the government and can be relatively expensive,
as they generally have higher fees and interest rates than federal
student loans.
Private education loans are estimated to have reached a peak of
about $22 billion in 2007- 08 (College Board 2009). That same
year, many lenders increased their direct marketing to students,
highlighting a quick and easy application and approval process for
private loans; some of these lenders were accused of deceptive mar-
keting practices (U.S. Senate Committee 2007). According to the
U.S. Department of Education (2008), "[p]rivate loans and credit
cards are consumer loans and are vety expensive ways of financing
your education."
Since 2007-08, however, the volume of private loans for postse-
condaty education is thought to have declined substantially due to
a shortage of capital and higher underwriting standards by lenders
(Student Lending Analytics 2009a, 2009b). Recent data released by
I 2 I THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION
1 the College Board (2009) are consistent with this expla-
2 nation.
3 This Statistics in Brief examines how the use of private
4 education loans changed between 2003-04 and 2007-
5 08 by addressing the following study questions:
6 1. How did undergraduate borrowing from private
7 sources change between 2003-04 and 2007-08?
8 2. Who obtained private loans?
9 3. To what extent did undergraduates combine
10 private and public loans?
11 4. Did undergraduates borrow the maximum
12 amount from federal Stafford loans before turn-
13 ing to private loans?
14 5. How did private borrowing change among
15 graduate and professional students?
16 Key Findings
17 The percentage of undergraduates obtaining private
18 loans nearly tripled, from 5 to 14 percent.
19 Undergraduates at for-profit institutions had the
20
21
22
23
24
25
highest rate of borrowing from private sources-42
percent took out private loans in 2007- 08.
Dependent undergraduates from middle-income
families borrowed from private sources at higher
rates than did students from low- or high-income
families.
26 About one-half of full-time, full-year undergraduates
27 who obtained a private loan had borrowed the max-
28 imum federal (Stafford) loan amount.
29
30
MAJOR TYPES OF
HIGHER EDUCATION LOANS
Private loans. Private loans are education loans,
not guaranteed by the federal government, from
commercial lenders, credit unions, or other non-
profit entities. Their terms are determined by the
lender. Private loans carry a market interest rate,
usually variable and based on credit history, and
they generally have higher fees and interest rates
than federal student loans.(See question on p. 8)
Stafford loans. These student loans have fixed
interest rates and various repayment benefits and
are guaranteed by the federal government. They
have eligibility requirements and limits on amounts.
There are two types of federal Stafford Loans: sub-
sidized and unsubsidized. Subsidized Stafford
loans are awarded based on financial need, and
the federal government pays interest on the loan
until the student begins repayment and during au-
thorized periods of deferment thereafter. Unsubsi-
dized Stafford loans are not need based; students
are charged interest for the duration of the loan,
although it can be capitalized.
Parent PLUS loans. These federally guaranteed
loans are available only to the parents of depen-
dent students. The interest rate in 2007- 08 was
fixed at ei ther 7.9 or 8.5 percent. Borrowers cannot
have a negative credit history, and the amount is
limited to the cost of attendance minus other finan-
cial aid. The loans carry the benefits and protec-
tions of all federal loans.
Graduate PLUS loans. These are federally guar-
anteed loans for graduate and professional stu-
dents that became available in 2006. The terms are
the same as for Parent PLUS loans, with the same
interest rates, restrictions, and benefits.
THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION I 3
1 Detailed Findings 28 tutions was $5,500 in 2007-08; at private nonprofit 4-
2 How did undergraduate borrowing from private
29 year institutions, it was $17,800; and at for-profit insti-
3 sources change between 2003-Q4 and 2007-QS?
30 tutions, it was $10,200. Further, the for-profit sector is
3 1 more likely to enroll low-income students. The median
4 T he rate of undergraduate private loan borrowing in- 32 income for dependent students' parents at public 4-year
5 creased from 5 to 14 percent between 2003-04 and 33 institutions was $75,700 in 2007-08.
4
At for-profit in-
6 2007- 08 (figure 1).
1
But the amount borrowed held 34 stitutions, the corresponding median income was
7 steady after adj usting for inflation: the average private 35 $35,700.
8 loan in 2003-04 was $6,600 and $6,500 in 2007-08.
2
36 I n general, the higher the tuition, the higher the rate of
9 Stafford loan borrowing among undergraduates also in- 37 private borrowing. For example, the highest rates of bor-
1 0 creased from 32 percent to 35 percent, and the average 38 rowing (30- 32 percent) occurred among students whose
11 Stafford loan amount rose from $4,900 to $5,000. 39 tuition was more than $10,000 per year (figure 3) . In
12 The rate of any borrowing rose from 34 percent in
13 2003- 04 to 39 percent in 2007- 08.
3
T he average loan
14 amount from all sources, including Parent PLUS loans,
15 increased from $6,900 to $8, 100.
40 comparison, 22 percent of students paying $5,000 to
41 $9,999 in tuition took out private loans, as did 14 per-
42 cent of those paying $3,000 to $4,999 and 9 percent or
43 less of those paying under $3,000 in tuition.
16 Who obtained private loans?
44 Dependent undergraduates from middle-income fami li es
45 attending full-time full-year borrowed from private
17 In 2007-08, private loan borrowing varied by type of
46 sources at higher rates than students from low- or high-
18
. . . . . I 1 d d d d
4
7
tnstttuuon, tuttton eve , stu ent epen ency status, an
19 family income. T he rate of private borrowing was high-
4 8
20 est at for-profit institutions, where the rate about

2 1 from 13 percent to 42 percent, from 2003-04 to 2007-
50
22 08 (figure 2). Private borrowing at private nonprofit 4-
51
23 year institutions was also high and about doubled, from
52
24 11 percent to 25 percent. 53
income famil ies. One-fifth of studenrs in the two mid-
dle-income groups took out private loans, compared
with 15-16 percent of students in the low- and high-
income groups (figure 4). Looking at aU borrowing, in-
cluding Parent PLUS loans, dependent students from
high-income fan1Uies borrowed at the lowest rate (40
percent).
54 Among independent students, upper middle- and high-
25 T he higher rate of borrowing among students at private 55 income students took out private loans relatively more
26 institutions reflects to some extent the higher tuition at 56 often than low-income students. One-fifth of low-
27 these institutions. Average tuition at public 4-year insti- 57
income independent undergraduates took out private
58 loans, compared with 27- 28 percent of those in the up-
1 All comparisons of estimates were rested for statistical signi- 5
9
per middle- and high-income groups.
ftcance usi ng the Student's t-statistic, and all differences cited
are statistically significant at the p < .05 level. No adj ustments
for multiple comparisons were made. The standard errors for
the estimates can be found at
http://nces.ed.gov/das/li brary/reporrs.asp.
2
All dollar amounts for 2003-04 have been adj usted for infla-
tion to 2007 doll ars using the Consumer Price Index.
3
Includes Parent PLUS loans as well as Stafford, Perkins, and
private loans.
4
Independent students are age 24 or older and students under
24 who are married, have dependents, are veterans, or are an
orphan or ward of the courts. Other undergraduates under age
24 are considered to be dependent. Income for dependent
students is the income of their parents. For independent stu-
dents, income includes a spouse' s income if the st udent is mar-
ried.
I 4 I THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION
1 To what extent did undergraduates combine pri-
2 vate and public loans?
37 Did undergraduates borrow the maximum amount
38 from federal Stafford loans before turning to pri-
39 vate loans?
3 Most undergraduates who borrowed did so through a
4 federally guaranteed loan program. Some 63 percent of 40
5 undergraduates who borrowed obtained loans from pubA 1
6 lie,
5
mostly federal, sources exclusively, and another 27 42
7 percent borrowed from both public and private sources 43
8 (figme 5) . The remaining 9 percent borrowed only from44
9 private sources. Students borrowed from public and pri- 45
10 vate sources at different rates, however, depending on 46
In 2007- 08, one-half of full-time, full-year undergra-
duates exhausted their annual Stafford loan eligibility
before taking out private loans (figure 6). Independent
private loan borrowers exhausted Stafford el igibility at a
lower rate than did dependent borrowers.
Policymakers are concerned that some students seek pri-
vate loans because they are unaware of the advantages of
11 the types of institutions attended.
47 federal loan programs. They are also concerned that
48 some students may not borrow the maximum Stafford
12 Undergraduates in public 2-year institutions took out
49 loan amount before turning to private loans (U.S. De-
13 exclusively private loans at higher rates than those in oth-
50 partment of Education 2008). Applying for federal aid is
14 er types of institutions. At community colleges, 21 per-
51 necessaty to obtain federal loans. Consequently, it is use-
15 cent of borrowers took out only private loans, compared 52 ful to examine whether students did apply for federal aid
16 with 9 percent among students in all institutions. This
53 when comparing public and private borrowing.
6
17 higher rate of private borrowing, however, occurred with-
18 in the context of lower rates of borrowing overall. In 54 Among full-time, full-year undergraduates who took out
19 2007-08, some 13 percent of all community college stu-55 private loans, some 19 percent did not obtain a Stafford
20 dents borrowed from any SOLtrce, representing the lowest56 loan, and about half these students (10 percent of all
21 rate among students in all types of institutions (Wei et al57 private borrowers) did not apply for financial aid. 7
22 2009a, table 13). Some policy analysts believe that many53 Another 31 percent took out Stafford loans, but appear
23 community colleges do not encourage borrowing federal 59 to have borrowed less than the ma:ximwn amount. s The
24 loans (Cochrane, Shireman, and Sivashankaran 2008). 60 remaining 50 percent had taken out the maximum
25 The largest proportion of borrowers who took out pri-
26 vate loans either exclusively or in combination with pub-
27 lie loans (42 percent) was found among those enrolled at
28 for-profit institutions (figure 2), and those enrolled at
29 institutions with programs of 2 or more years had the
30 highest percentage who took out both private and public
31 loans (figure 5). Among those enrolled at for-profit insti-
32 tutions with programs taking 2 or more years to com-
33 plete, 45 percent of undergraduate borrowers took out
34 both private and public loans, while the overall percen-
35 cage of undergraduates who cook out both types of loans
36 at all institutions was 27 percent.
5
"Public loans" includes Stafford loans, other federal loans
(e.g., Perkins and PLUS), state, and institutional loans.
6
Foreign students and those attending part time or part yea.r
were excluded from this part of the analysis because they either
are ineligible for federal loans or qualifY for very little. Other
reasons that students cannot borrow from federal loan pro-
grams couJd not be identifled from the NPSAS data.
7
Some may have taken out other loans, such as Perkins, state,
or institutional loans, but those are comparatively rare.
8
The maximum Stafford amomu is limited for those attend-
ing part time or for less than a fttll year, so this analysis of
maximizing Stafford loans was limited to fuJI-time, full-year
students. There are other restrictions on Stafford borrowing
based on program or budget that were not taken into account.
Because of these restrictions, some borrowers who appear to be
borrowing below the maximum amount are in fact borrowing
at their maximum. Therefore, our estimates of the percentage
of borrowers who are borrowing at less than the maximum
couJd be overstated.
THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION I 5 I
1 amount allowed under the Stafford loan program. These24 points,
9
compared with an increase of9 percentage
2 students sought more money than was available from 25 points among undergraduates (figure 1 ).
3 federal programs to pay for their education expenses.
26 Graduate students differed significantly in borrowing
4 Undergraduate students have different annual Stafford 27 rates depending on the degree program in which they
5 loan limits based upon both. dependency status and class28 were enrolled (figure 7). Master's degree students' rate of
6 level. For example, for first-year undergraduates, the lim-29 total borrowing increased from 38 percent to 44 percent
7 its were $3,500 for dependent students and $7,500 for 30 over this period. The rate of private borrowing among
8 independent students. In 2007- 08, about one-third (35 31 master's degree students doubled between 2003- 04 and
9 percent) offull-time, full-year independent undergra- 32 2007-08 (from 6 percent to 12 percent), while their rate
10 duates who took out private loans also borrowed the 33 of Stafford borrowing increased from 36 percent to 39
11 maximum Stafford loan amount, compared with. 56 per-34 percent.
12 cent of dependent undergraduates who took out the
13 maximum Stafford loan amount (figure 6) . 35 Among doctoral students, changes in private, Stafford,
14
Annual Undergraduate
Stafford Loan Limits
(for loans taken out between July 1, 2007
and June 30, 2008)
36 or total borrowing were not evident. Between 5 and 7
7 percent took out private loans in both 2003- 04 and
8 2007-08.
9 Students in first-professional programs had the highest
f0 rate of private borrowing, but the proportion with pri-
Dependent Independent
Academic year student student i 1 vate loans declined over time, while borrowing from fed-
------''----------------- i2 eral sources increased.
10
In both years, students in frrst-
First year $3,500 $7,500 .
Second year 4,500 8,500 !3 professional programs borrowed from all sources at high-
Third and remaining years 5,500 10,500 i4 er rates than students in other degree progran1s. In
NOTE: Aside from dependency status and year in school, the i5 2003-04, about one-fourth (23 percent) of students in
amount a student can borrow under the Stafford loan program
can be further reduced depending on the cost of attendance, the
student's expected family contribution, attendance status,
whether the program is less than a year long, and how much
other financial aid is received.
SOURCE: U.S. Department of Education. (2007). 2007-oB
Federal Student Aid Handbook. Retrieved April 23, 2010, from
http://ifap.ed.gov/ifaplbyAwardYear.jsp?tyoe-fsahandbook&awa
i6 first-professional programs took out private loans, com-
i7 pared with 5 to 6 percent of master's and doctoral stu-
iS dents. By 2007-08, however, the rate of private
i9 borrowing among first-professional students had
;o dropped to 16 percent, while their rate of Stafford bor-
_::rd;:v::ea;:r;:=2::00=:
7
- ::2:;00:;8::,. _______________ , ; 1 rowing increased from 69 to 76 percent and one-fourth
15 How did private borrowing change among gradu-
52 (25 percent) took out Graduate PLUS loans.
16 ate and first-professional students? 53
54
17 ln 2006, graduate students became eligible for the feder-
55
18 al Graduate PLUS loan program. This gave graduate 56
19 students another source ofloans from a government-
20 guaranteed progran1 with competitive, ftxed-interest
21
22
23
rates, eliminating some of the need for private loans.
Between 2003- 04 and 2007- 08, the rate at which grad-
uate students took out private loans rose 4 percentage
Between 2003- 04 and 2007- 08, after adjusting for in-
flation, overall average loan amounts for all graduate and
first-professional students increased from $17,700 to
$18,500, largely due to the newly established Graduate
9
NPSA$:04 and NPSA$:08 Data Analysis System; data not
shown in figures or tables.
1
First professional programs include dentistry, medicine,
optometry, osteopathic medicine, pharmacy, podiatric medi-
cine, veterinaty medicine, chiropractic, law, and theology.
I 6 I THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION
1 PLUS loans. The average private loan amount for gradu-37
2 ate students decreased from $ l 0.500 to $8,400, and the
38
3 average Stafford loan amount decreased from $16, 100 to
39
Technical Notes
Dara in this report come from the National Postsecon-
dary Student Aid Study (NPSAS) administered for the
4 $15,600.
40 years 2003-04 and 2007-08. Conducted every 4 years,
5 Find Out More
4 1 NPSAS is based on a nationally representative sample of
42 al l students enroll ed in Title N -eligible institutions in
6 More detailed information on 2007-08 undergraduates 43 the United States between J uly 1 and J une 30 of a given
7 and graduate students enrolled in U.S. postsecondaty 44 academic year. The sampling design for coll ecting data
8 institutions can be found in Web Tables produced by 45 consists of two stages. The first stage involved selecting
9 the National Center for Education Statistics (NC.ES) 46 eligible institutions from which students would then be
10 using the 2007- 08 National Postsecondaty Student Aid 47 sampled in the second stage. In 2007- 08, approximately
11 Study (NPSAS:08) data. These Web Tables are a com- 48 114,000 undergraduate students were study respondents.
12 prehensive source of information on students during the49 Estimates were weighted to adjust for the unequal prob-
13 2007- 08 academic year. Included are estimates of de- 50 abili ty of selection into the sample and for nonresponse.
14 mographi cs, enroll ment, and employment characteris-
15 tics. Web Tables docw
11
enting how students pay for 51 NCES Statistical Standard 4-4-1 states that "[a]ny survey
16 their undergraduate education are also available. 52 stage of data collection wim a unit or item response rate
53 less than 85 percent must be evaluated for the potential
17 Web Tables-Profile of Undergraduate Students in US. Postse- 54 magnitude of nonresponse bias before the data or any
18 condary !mtitutiom: 2007-08 (NCES 20 I 0-205). [link 55 analysis using me data may be released" (U.S. Depart-
19 will be added)
56 ment of Education 2003). The only variable included in
20 Web Tables-Student Financing of Undergraduate Education: 57 this Bri ef that has a response rate of less than 85 percent
2 1 2007-08 (NCES 2010-162) . [link will be added) 58 is TOTLOAN.
22 Web Tables-Profile of Students in Graduate andFirst-
23 Professional Education: 2007-08 (NCES 20 10-177). [link59
For an overview of the survey memodology and a discus-
24 wil l be added] 60
sion of nonresponse bias, see appendix B of the report
6 1 2007- 08 National Postsecondary Student Aid Study
25 Readers may also be interested in the following NCES
26 products related to me topic of mis Statistics in Brief
62
63
27 2007-08 National Postsecondary Student Aid Study
28 (NPSAS:08): Student Finandal Aid Estimates: First Look 64
29 (NCES 2009-166).
30 hrrp:l/nces.ed .gov/pubs2009/2009l66.pdf
3 1 Undergraduate Financial Aid Estimates by Type of Imtitution in
32 2007-08 (NCES 2009-201).
33
34
35
36
h ttp:l/nces.ed.g,ov/pubs2009/2009201.pdf
Federal Loam in 2007-08: Undergraduate Cumulative Debt
and Students Who Bonow at the Maximum (NCES 2010-
151). [link to be added)
(NPSAS:OB): Student Financial Aid Estimates for 2007-08:
First Look (http:/ / nces.ed. gov/pubs2009/2009166.pdt).
THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION
1 References
39
40
41
2 The College Board. (2009). Trends in Student Aid. Table l.
3
4
5
Retrieved October 28, 2009, from
http://www.rrends-collegeboard.com/student aid/pdf/
2009 Trends Smdent Ald.pdf.
42
43
44
6 Cochrane, D. F., Shireman, R., and Sivashankaran, S. (2008,
7
8
9
10
11
12
13
14
15
16
17
18
19
20
2 1
22
April). Denied: Community College Students Lack Access to 45
A/fordable Loam. The Project on Smdent Debt. Retrieved46
September 9, 2009, from 47
48
http://ticas.org/ flies/pub/ denied .pdf.
Federal Trade Commission. (2008, June). Student Loam:
49
50
Avoiding Deceptive Offers. FTC Facts for Consumers. 51
Federal Trade Commission and U.S. Department of Edu-
cation. Retrieved September 9, 2009, from
h rrp://sruden taid.ed.gov/smden ts/auach menrs/
siteresources/loa nsA voidDecep. pdf.
King, J.. (2007). Who Borrows Private Loam? ACE Issue
52
53
54
55
56
Brief. Washington, DC: American Council on Education.
Retrieved October 28, 2008, fiom 57
http://www.acenet.edu/AJ\If/Template.cfm?Section-Searc58
h&seqion-issue briefs&templare-/CM/ContenrDisplay.59
cfm&ConrenrFileiD- 3445. 60
6 1
23 McSwain, C., Price, D., and Cwutingham, A. (2006). The 62
24 Future of Private Loam: Who Is Bonowing, and Why?
25 Washington, DC: lnstitute for Higher Education Policy. 63
64
26 Student Lending Analytics. (2009a, July 23). The Incredible 65
27 Shrinking Private Student Loan Mmket: Outlook for 2009"'66
28 10. Slog. Retrieved September 9, 2009, from 67
29 h up://studen tlendi nganalytic.qypepad.com/studen t lend68
30 ing analytics/2009/07/the-i ncredible-shrinking-private- 69
31 loan-market.html.
70
32 Student Lending Analytics. (2009b, August 17). Bank Lending
33 Standards May Remain Tight Through Second Half of
34 2010. Blog. Retrieved September 9, 2009, from
3 5 http://srudenrlendinganalytics. typepad.corn/smdent lend
36 ing analytics/2009/08/bank-lending-standards-may-
37 remain-tight-through-second-half-of-20 l O.htrnl.
38
U.S. Department of Education, Federal Student Aid, St udents
Charmel. (2008). Your Federal Student Loans: Learn the
Basics and Manage Your Debt. Washington, DC: Author.
U.S. Department of Education, National Center for Educa-
tion Statistics. (2003). NCES StatisticaL Standards (NCES
2003-601). Washington, DC.
U.S. Senate Committee on Banking, Housing and Urban Af-
fairs. (2007, J tme 6). Hearing on "Paying for College:
The Role of Private Student Lending." Washi ngton, DC.
Retrieved November l, 2009, from
h rtp://banki ng.senate.gov/ pub! ic/i ndex .cfm?Fuseaction-
Hearings.Hearing&Hearing ID-8c46al82-alff-4bc5-
af53-e0982b49527f.
Wei, C. C. (forthcoming). What is the Real Price of College?
Sticker, Net, and Out-of-Pocket Prices in 2007-08
(NCES 2010-175). National Center for Education Statis-
tics, Institute of Education Sciences, U.S. Department of
Education. Washington, DC.
Wei, C.C., Berkner, L. , He, S., ar1d Lew, S. (2009a). Web-
Tables: Undergraduate Financial Aid Estimates by Type of
Institution in 2007-08 (NCES 2009-201). National Cen-
ter for Education Statistics, Institute of Education
Sciences, U.S. Deparunenr of Education. Washington,
DC.
Wei, C. C., Berkner, L., He, S., Lew. S., Cominole, M., and
Siegel, P. (2009b). 2007-08 NationaL Postsecondmy Stu-
dent Aid Study (NPSAS:08): Student Financial Aid Esti-
mates for 2007-08 (flrst Look) (NCES 2009-166).
National Center for Education Statistics, Institute of
Education Sciences, U.S. Department of education.
Washington, DC.
I 8 I THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION
1 Run Your Own Analysis in Datalab 38 "During the 2007-2008 school year, did you take out
39
2 Direct access to the data used in this Statistics in Brief is
40
3 available through the NCES DataLab
any private or alternative loans from a financial institu-
tion? Some examples of commonly used private loans
include: 41
4 (http://nces.ed.gov/datalab/). You can generate your own
5 tables using one of three programs: PowerS tats, Quick- 42 Sallie Mae Signature Student Loan
6 Stats, or the Data Analysis System (DAS). 43 CitiAssist Undergraduate and Graduate Loan
7 All estimates presented in this Brief were produced using44 Chase Education One Private Student Loan
8 the DAS, a web-based software application that enables 45 Nellie Mae EXCEL Loan
9 users to generate tables for most of the postsecondary
46 Access Group Loans
10 surveys conducted by NCES. These figures were gener-
11 ated using the DAS, which contains a detailed descrip- 47 (Keep in mind that many lenders that offer private loans
12 tion of how each variable was created and includes the 48 might also offer federal Stafford, Graduate/Professional
13 wording of questions for variables coming directly from 49 PLUS loans, and Parent PLUS loans. For this question
14 the interview. 50 we want to know about private or alternative loans on-
51 ly.)"
15 You can replicate or expand upon the figures and tables
16 presented in this report by using DataLab. All variables 52 Package of private and nonprivate loans (PRJVPACK):
17 used to generate the figures and ta.bles in this Brief can
18 be found in the NPSAS:04 and NPSAS:08 DAS. More
19 information and descriptions of each variable can be
20 found at http://nces.ed.gov/DAS.
21 The variables used in this report are as follows:
53 Indicates the loan package by whether the loan received
54 was private (alternative) or not during the 2007- 08 aca-
55 demic year. For students with any loan (TOTLOAN2 >
56 0), indicates whether a borrower had only private oral-
57 ternative loans (PRIVLOAN), only nonprivate loans
58 consisting of federal loans (TFEDLN2, including
59 PLUS), state loans (STLNAMT), or institutional loans
60 (INLNAMT), or both private and nonprivate.
22 Private (alternative) loans (PRJVLOAN): Indicates the
23 amount of private or alternative loans received by stu-
24 dents for the 2003- 04 and 2007- 08 academic years.
61 Stafford total maximum (STAFCT3): Classifies the total
25 These are education loans from commercial lenders that 62 Stafford Joan amount received in 2007-08
26 are not guaJanteed by the federal government and that 63 (STAFFAMT) into categories based on the maximum
27 cany market interest rates based on credit scores. Exam- 64 Joan limits for subsidized and unsubsidized Stafford
28 pies of such loans are the Sallie Mae Signature Student 65
29
30
loans combined and includes a category for those who
didn't apply for federal aid. The normal maximum loan
an1ounts in 2007-08 for undergraduates were deter-
Loan, CitiAssist Loan, or Chase Education One Private 66
Student Loan. Private loans differ from Stafford, Parent 67
31 PLUS, Perkins, and Graduate PLUS loans, which are
32
33
68 mined by the student's undergraduate class level and
guaranteed by the federal government; private loans, 69 dependency status. This variable does not take into ac-
however, can be offered by the same lenders or by states 70 count further borrowing limits that might be imposed,
34 or other nonprofit institutions. Institutions do notal- 71 e.g., Stafford loan amounts cannot exceed a borrower's
35 ways have information on students' private or price of attendance, and subsidized Stafford loans cannot
36 loans, so this information came primarily from student 73 exceed a borrower's need amount.
37 interviews, in which students were asked the following:
1 Other Variables
2 Attendance pattern (A ITNST AT)
3 Citizenship status (CITIZEN2)
4 NPSAS institution control (CONTROL)
5 Dependency status (DEPEND)
6 Parents' income (DEPlNC)
7 Independent student's income (INDEPINC)
8 Applied for federal aid (FEDAPP)
9 Graduate PLUS loan total (GPLUSAMT)
10 Graduate degree program (GRADDEG)
11 Total federal Parent PLUS loans (PLUSAMT)
12 Type of institution (AID SECT)
13 Institution sector (SECI'OR4)
14 Federal subsidized loans (except PLUS) (SUBLOAN)
15 Total student loans from all sources (TOTLOAN)
16 Institution tuition and fees (TUITION2)
17
THE EXPANSION OF PRIVATE LOANS IN POSTSECONDARY EDUCATION I 9 I
www.ed.gov ies.ed.gov
Figure 1. Percentage of undergraduates who borrowed, by type of loan: 2003-04 and 2007-08
Percent
100
80
60
40
32
35
20
14
5
0
Private Stafford
2003-04 2007-08
34
39
All loans
Average undergraduate loan amounts,
by type of loan:
2003--04 and 2007-QS
2003--04 2007-QS
Private loans $6,600 $6,500
Stafford loans 4,900 5,000
All loans including PLUS 6,900 8,100
NOTE: Amounts for 2003-04 have been adjusted for inflation
using the Consumer Price Index for urban households (CPI
U). Arrounts are averages for those who received the
specified type of aid.
NOTE: Private loans are education loans from commercial lenders which are not guaranteed by t he federal government and carry market
interest rates.
SOURCE: U.S. Department of Education, Nationa.l Center for Education Statistics, 2003-04 and 2007-08 National Postsecondary Student Aid
Studies (NPSAS:04 and NPSAS:OS) .
Figure 2. Percentage of undergraduates who took out private loans, by institution type: 2003-04 and 2007-08
Percent
100
80
60
40
20
0
Public
2-year
14
Public
4-year
2003-04
25
11
Private
nonprofit
4-year
2007-08
42
13
For-profit
Average tuition and fees paid by
undergraduates, by institution type:
2007-QS
(includes full-time and part-time students)
Public 2-year
Public 4-year
Private nonprofit 4-year
For-profit
$1,200
5,500
17,800
10,200
Median income, by institution type: 2007-QB
Dependent lndepen-
students dent
parents students'
Public 2-year $54,200 $29,400
Public 4-year 75,700 22,000
Private nonprom 4-year 84,500 30,300
For-profit 35,700 19,500
1
Independent students are age 24 or older and 1udents under
24 who are married, have dependents, are veterans, or are an
orphan or ward of the oourts. Other undergraduates under age
24 are considered to be dependent. Income for dependent
students is the inoome of their parents. For independent
students. income includes a spouse's income if the student is
married.
NOTE: Private loans are education loans from commercial lenders which are not guaranteed by t he federal government and carry market
interest rates. Students who attended more than one inst itution or any type of institution not listed in the figure were not included due to the small
number of cases.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2003-04 and 2007-08 National Postsecondary Student Aid
St udies (NPSAS:04 and NPSAS:08).
Figure 3. Percentage of undergraduates who took out private loans, by
tuition amount: 2007-os
$15,000 or more
$10,000-$14,999
$5,000-$9,999
$3,000-$4,999 14
$1 ,500-$2,999
$1 ,499 or less
0 20
32
30
22
40 60
Percent
80
NOTE: Private loans are education loans from commercial lenders which
100
are not guaranteed by the federal government and carry market interest rates.
SOURCE: U.S. Department of Education. National Center for Education
Statistics, 2007- 08 National Postsecondary Student Aid St udy (NPSAS:08).
Figure 4. Percentage of fulltime, full-year undergraduates who took out private loans and any loans including Parent PLUS, by
dependency status and family income level: 2007-QS
Percent
100
80
60
54
56
61
66
69
52
40
40
27
28
21
20
0
60
Low-income Lower
middle-
income
Upper High-income
middle-
Low-income Lower
middle-
income
Upper High-income
middle-
income income
Dependent Independent
Private loans All borrowing including PLUS
NOTE: Family income categories were based upon the annual income distribution for 2006, from the lowest to the highest quartile ranges.
Among dependent students, low-income was less than $36,100: lower middle-income was $36.100-$66,600: upper middle-income was
$66,600-$104,600: and high-income was $104,600 or more. Among independent students, low-income was less than $11 ,000; lower middle-
income was $11,000-$26,000; upper middle-income was $26,000- $48,400; and high-income was $48,400 or more. Private loans are education
loans from commercial lenders which are not guaranteed by the federal government and carry market interest rates.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-()8 National Postsecondary Student Aid Study
(NPSAS:08).
Figure 5. Percentage distribution of undergraduate borrowers, by borrowing package and institution type: 2007-QS
Percent
100
..

2
Exclusively
80 20
private
27
35 45 loans
60 Both
programs
40
Exclusively
20
public
loans
0
Total Public Public Private For-profit For-profit
2-year 4-year nonprofit less-than- 2 years
4-year 2-year or more
NOTE: Public loans includes Stafford loans, other federal loans (e.g., Perkins and PLUS), state, and institutional loans. Private loans are
education loans from commercial lenders which are not guaranteed by the federal government and carry market interest rates. Total estimates
also include those attending public less-than2-year institutions, those attending private nonprofit less-than-4-year institutions, and those
attending more than one institution but t hese are not shown separately. Detail may not sum t o totals because of rounding.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-QS National Postsecondary St udent Aid Study
(NPSAS:OS).
Figure 6. Percentage distribution of full-time, full-year undergraduates who took
out private loans, by status of Stafford borrowing, status of federal aid
application, and dependency status: 2007-QB
Percent
100
80
60
40
20
0
Total Dependent Independent
~ M a x i m u m
Stafford
DLessthan
maximum
Applied,
no Stafford
Did not
apply for
aid
NOTE: In this analysis, full-lime undergraduates excludes foreign students and
unclassified undergraduates. Private loans are education loans from commercial
lenders which are not guaranteed by the federal government and carry market
interest rates. Maximum Stafford is a loan in the amount of the annual limit for
Stafford loans based on dependency status and class level. Some restrictions on
Stafford borrowing were not taken into account when computing a student's Stafford
borrowing limit. Because of these restrictions, some borrowers who appear to be
borrowing below the maximum amount are in fact borrowing at their maximum.
Therefore, estimates of the percentage of borrowers who are borrowing at less than
the maximum may be overstated. Detail may not sum to totals because of
rounding.
SOURCE: U.S. Department of Education, National Center for Education Statistics,
2007-08 National Postsecondary Student Aid Study (NPSAS:08).
Figure 7. Percentage of graduate students who took out private, Stafford, Graduate PLUS, and all loans, by type of degree
program: 2003-04 and 2007-08
Private loans
Stafford loans
Graduate PLUS loans
Total loans
L6
~ 1 2
I 3
38
44
p,,. .. '""' ~ 5 7
Stafford loans ~
2 6
2 9
Graduate PLUS loans
Total loans
Private loans
Stafford loans
Graduate PLUS loans
Total loans
0 20
28
32
40
Percent
69
60
Average loan amounts received by graduate
students, by type of loan: 2003--{)4 and 2007-{)8
Private loans
Stafford loans
Graduate PLUS loans
Total loans
2003-Q4 2007--{)8
$10,500
16,100
17,700
$8.400
15,600
15,500
18,500
NOTE: Amounts for 2003-04 have been adjusted for inflation
using the Consumer Price Index for urban households (CPI-U).
Amounts are averages for those who received the specified
type of aid.
2003-04
2007-08
76
75
79
80 100
1
First-professional programs include dentistry, medicine, optometry, osteopat hic medicine, pharmacy, podiatric medicine, veterinary medicine,
chiropractic, law, and theology.
NOTE: Private loans are education loans f rom commercial tenders which are not guaranteed by the federal government and carry market interest
rates. Graduate PLUS loans were not available in 2003-Q4.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2003-Q4 and 2007-QS National Postsecondary Student Aid
Studies (NPSAS:04 and NPSA$:08).
U.S. DEPARTMENT OF EDUCATION APRIL 2010
What Is the Price
of College?
Sticker, Net, and
Out -of-Pocket Prices
in 2007-08
AUTHOR
Christina Chang Wei
MPR Associates, Inc.
PROJECT OFFICER
Tracy Hunt-White
National Center for Education Statistics
Students and parents see
college attendance as a principal avenue
to middle-class life, and, given the rising
price of postsecondary education, they
are apprehensive about their ability t o af-
f ord it.
1
In a recent survey of college
freshmen, about two-thirds (66 percent)
reported having concerns about being
able to fi nance their education.
2
Many U. S.
policymakers and researchers share their
concern, and are exploring ways to make
college more affordable.
3
Legislators have
required colleges and universities t o pro-
vide more extensive information about
tuition and prices, and in the 2008 Higher
Education Opportunity Act mandated a
host of price-related measures, including
institutional disclosure of net prices
charged to students, the reporting of net
price data to the U.S. Department of Edu-
cation, and the creation and disclosure of
"tuition watch lists."
4
This Statistics in Brief shows the amounts
U.S. undergraduates pay on average for
postsecondary education, wi th and wi th-
out financial aid. Drawing upon data from
the National Postsecondary Student Aid
Study, a nationally representative survey
of all postsecondary students, included in
this report are the average prices for pub-
lic 2-year, public 4-year, private nonprofit
4-year, and for-profit institutions.
1
A recent public opinion poll showed that increasing numbers of
Americans view college as a necessity for success (lmmerwahr et al.
2009). The College Board (2009) reports that the published, or stick
er, price of college has increased more rapidly than the price of other
goods and services over the past three decades. See also Marchand
(2010), which reported on these findings.
1
The Higher Education Research Institute at the University of Cali
fornia, Los Angeles, has conducted a survey of freshmen each year
since 1973 (Pryor et al. 2009). See also recent media anent ion to the
issue (Lewin 2009).
3
Both governmental and nongovernmental study commissions have
convened to address this. See, for example, The College Board
(2008) and Harvey et al. (1998).
4
Public Law 110315, title I, Sec. 132. August 14, 2008. Higher Edu
cation Opportunity Act .

1es
NATIONAL CENTER FOR
. . EDUCATION STATISTICS
STUDY QUESTIONS
What are the average
prices paid by full-time
undergraduate students
and how do these prices
vary by the type of insti-
tution attended?
Most of the figures in this Brief display
data only for full-time undergraduates
5
who attended one institution. These
students comprised about one-third
(35 percent) of all undergraduates in
2007-08.
6
Of all full-time undergra-
duates, 18 percent were enrolled in
public 2-year institutions, 43 percent
attended public 4-year institutions, 21
percent were at private nonprofit 4-
year institutions, and 8 percent were
enrolled in for-profit institutions (table
1).
7
Focusing on full-time students who
attended only one institution allows for
comparisons in tuition, price of atten-
dance, and financial aid. Those attend-
ing full time generally have higher
overall expenses than do all students.
They also generally qualify for federal
aid and other assistance not available
to many part-time students (table 2).
5
"full time" status is defined as having been enrolled full time
in one postsecondary institution for 9 months or more during
the academic year.
6
National Postsecondary Student Aid Study (NPSAS:OS) Data
Analysis System.
1
The remaining students were enrolled in other types of insti
tutions or in more than one institution during the academic
year.
What factors are re-
lated to variations in
average sticker and
net prices among
those attending these
institutions?
KEY FINDINGS
There is a wide range of prices for
postsecondary education. Students
enrolled at public 2-year institutions
had the lowest average sticker price
($12,600) while those at private
nonprofit 4-year institutions had the
highest {$35,500).
8
Those at public
4-year institutions had an average
sticker price of $18,900, and those at
for-profit institutions had an aver-
age sticker price of $28,600.
After all financial aid is received, {in-
cluding grants, loans, and work-
study), the average out-of-pocket
net price ranged from $9,100 at
public 2-year institutions to $10,300
at public 4-year institutions, $16,000
at for-profit institutions, and
$16,600 at private nonprofit 4-year
institutions.
3
All comparisons of estimates were tested for statistical signi
ficance using the Student's t-statistic, and all differences cited
are statistically significant at the p < .OS level. No adjustments
for multiple comparisons were made. The standard errors for
the estimates c a n ~ found at http://nces.ed.gov/dasf
librarv/reoorts.asp.
2
How do the net
prices paid by under-
graduates vary by
family income?
Lower sticker prices often mean a
lower need-or eligibility-for fi-
nancial aid. Those attending private
institutions had the highest tuition
but they also received the most fi-
nancial aid. Undergraduates at pri-
vate nonprofit 4-year institutions
received the greatest amount of in-
stitutional grant aid while those at
for-profit institutions had the largest
proportion of borrowers.
Among low-income undergra-
duates enrolled full time, those at-
tending for-profit institutions had
the highest average net price after
grants and average out-of-pocket
net price.
1
What are the average prices paid by full-time
undergraduate students and how do these
prices vary by the type of institution attended?
This Brief discusses three different
measures of the price of an undergra-
duate education: the sticker price, the
net price after grants, and the out-of-
pocket net price.
The Sticker Price
The sticker price is the total price of an
undergraduate education including
tuition
9
and all other nontuition and
living expenses such as books, sup-
plies, and housing. The sticker price va-
ried widely by the type of institution
attended in 2007-08, rangi ng from an
average of $12,600 among undergra-
duates enrolled full time at public 2-
year institutions, to $18,900 at public 4-
year institutions, $28,600 at for-prof it
institutions, and $35,500 at private
nonprofit 4-year institutions (figure 1 ).
9
1n this report, the term "tuition includes both tuition and
fees. Sometimes institutions treat tuition and fees as a single
charge, and sometimes as separate charges. Tuition is defined
as the price ofinstruction and fees as the price of other services
provided by the school. The tuition amounts shown here in
elude those charged to outof-state and out-of-district students
enrolled in public 4-year and public 1-year institutions, respec-
tively.
FIGURE 1.
THE PRICE OF AN UNDERGRADUATE EDUCATION
for full-time, full-year undergraduates,
by type of institution attended: 2007- 08
Price
$40,000
30,000
20,000
10,000
0
Public 2-year Public 4-year For-profit
Type of institution
Private nonprofit
4-year
Average sticker w Average net price ~ A v e r a g e out-of-
price after grants pocket net price
NOTE: The "sticker price" is the total price of attendance which includes tuition and fees, books and supplies, housing,
meals, transportation, and other miscellaneous, or personal, expenses. The "net price is the price that must be paid after
receiving grants. The "out-of-pocket net price subtracts from the sticker price all forms of financial aid, including grants,
student loans, Parent PLUS loans, work-study, employer aid, job training benefits, veterans benefits, and any other finan
cia I aid received. Both the net price after grants and the out-of-pocket net price are calculated for all full-time, full-year
students, regardless of whether or not they received any financial ald. "Full-time, full-year' is defined as having been
enrolled in one postsecondary institution for 9 months or more full time. Estimates include students enrolled in Title IV
eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Standard error tables are available at
hnpl/nces.ed.gov/das/library/reports.asp.
SOURCE: U5. Department of Education, National Center for Education Statistics, 1007-08 National Postsecondary Student
Aid Study (NPSAS:08).
3
For some institutions (private nonprofit known in financial aid parlance as the supporting families (Wei forthcoming),
4-year institutions in particular), the "student budget." Student budgets which raises their nontuition expenses
tuition is a large component of the vary according to students' attendance as compared with students who have
sticker price. In 2007-08, average tui- and dependency status, family respon- no family responsibilities.
tion was $2,400 at public 2-year institu- sibilities, and residence (e.g., living at
tions, $7,100 at public 4-year home with parents, in an on-campus Students attending public 2-year insti-
institutions, $11,900 at for-profit insti- dormitory, or in off-campus housing). tutions-mostly community colleges-
tutions, and $23,400 at private non-
profit 4-year institutions {figure 2).
In 2007-08, students enrolled full time
in for-profit institutions had the high-
Nontuition expenses, which include est average nontuition expenses
had the lowest nontuition expenses. A
larger proportion lived at home with
their parents, which reduces costs for
room and board, as compared with
students at for-profit institutions (many
of whom are supporting themselves or
their own families) and those at 4-year
institutions (where a larger proportion
lived on campus or in off-campus hous-
ing) (Wei forthcoming).
books and supplies, housing and ($16,700), when compared with those
meals, transportation, and personal at other types of institutions {where
{or miscellaneous) expenses, also can average nontuition expenses ranged
vary by institution type. College or uni- from $10,200 to $12,1 00). Many stu-
versity financial aid officers usually de- dents at for-profit institutions are fi-
velop an estimate of the sticker price, nancially independent and are
In-state vs. out-of-state tuition
At most public 4-year
institutions, tuition charges are
generally higher for out-of-state
students than for in-state
residents, reflecting the state
subsidies public institutions
receive. In 2007- 08, the average
in-state tuition was $6,200 and
the average out-of-state tuition
was $15,1 00 for full-time
undergraduates enrolled in
public 4-year institutions.*
NPSAS:OS Data Analysis System (data not shown).
FIGURE 2.
TUITION AND NONTUITION EXPENSES
for full-time, full-year undergraduates,
by type of institution attended: 2007- 08
Price
$40,000
30,000
20,000
10,000
0
Public 2-year Public 4ycar For profit
Type of institution
Private
nonprofit
4-year
tJ Average
nontuition
expenses
Average
tuition
NOTE:"full-time, full year is defined as havlng been enrolled in one postsecondary institution for 9 months or more full
time. "Tuition" includes all tuition and fees. Non tuition expenses" include books, supplies, housing, meals, transporta
tion, and miscellaneous 01 pe1>onal expenses. Estimates include students enrolled in Title IV eligible postsecondary institu-
tions in the 50 states, DC, and Puerto Rico. Standard error tables are available at
htto://nces.ed.gov/das/librarv/reoorts.aso.
SOURCE: U 5. Department of Education, National (enter for Education Statistics, 2007-08 National Postsecondary Student
Aid Study (NPSAS:OS).
4
The Net Price After Grants those at public 4-year institutions,
Policymakers and researchers generally grant aid to full-time undergraduates
subtract grants from the sticker price lowered the average sticker price of
when discussing the net price of at- $18,900 to an average net price after
tending a postsecondary institution. grants of $15,200.
For example, Congress recently re-
quired institutions to make public both For those at private institutions (both
the sticker price and the average net for-profit and nonprofit), the average
price of attendance, calculated as the net price after grants was about
total price of attendance minus all $25,700. For those attending for-profit
grants received. In this Brief, two
measures of net price are discussed:
the "net price after grants" and the
"out-of-pocket net price," which is the
price after all financial aid (i ncluding
loans, which must be repaid} is t aken
into account.
10
Grant aid helped to lower the average
st icker price among full-time under-
graduates at public 2-year institutions
from $12,600 to an average net price
after grants of $10,600 (figure 1 ). For
10
Both net price and out of-pocket net price averages are cal-
culated for all students, regardless of whether they received
any financial aid. This method of calculating net price averages
for all students differs from that used to calculate average aid
amounts in this report (see table 2). Average aid amounts are
calculated only for students receiving a particular type of aid.
Those not receiving a specific type of aid (i.e., zero values) are
not included in the average for that aid. The average grant
therefore, will be greater than the difference between the
sticker price and the net price after grants.
institutions, the difference between
the average sticker price and the net
price af ter grants was about $2,800. At
private nonprofit 4-year instituti ons,
however, that difference was $10,000.
The Out-of-Pocket Net Price
The "out-of-pocket net price," defined
as the sticker price less all fi nancial aid
received, takes into account all forms
of financial aid, including grants, loans,
work-study, and other aid (as well as
Parent PLUS loans). The out-of-pocket
net price represents the amount t hat
must be paid immediately to enroll in a
postsecondary institution f or that aca-
demic year.'' Because the out-of-
pocket net price subtracts loans from
the sticker price, it measures the net
price only in the short t erm. Loans off-
set immediate costs to students and
their families, but they must be repaid
overtime.
Full-time undergraduates enroll ed at
public 2-year institutions had the low-
est average out-of-pocket net price
($9,100}, reflecting the already lower
sticker price at these institutions (figure
1}. Those at public 4-year instituti ons
had a slightly higher average out-of-
pocket net price ($1 0,300}. Students at
private institutions had the highest av-
erage out-of-pocket net prices ($16,000
at for-profit institutions and $16,600 at
private nonprofit 4-year institutions).
11
Since 1998, the federal government has also provided postse-
condary students and their families with various federal tax
benefits. These are not included in the definition of financial aid
and are not used in the calculation of net price in this study.
2
What factors are related to variations
in average sticker and net prices among
those attending these institutions?
Prices vary by institution type for sev-
eral reasons. Institutions charge differ-
ent levels of tuition (based on whether
they are public or private and the
amount of state and local subsidies re-
ceived); the demographic characteris-
tics of students (and thus their
nontuition expenses and eligibility for
federal and state grant aid) vary by
type of institution; and institutional
policies for awarding institutional aid
differ.
As noted previously, the amount of tui-
tion charged by institutions is a prima-
ry factor in the total price of
attendance, but students also incur dif-
ferent nontuition expenses depending
on their family responsibilities and liv-
ing arrangements.
Average net prices are further affected
by differences in the proportion of aid
recipients at each type of institution.
For example, the number of students
eligible for federal Pell Grants or state-
funded grant aid at a particular school
will affect the average net price after
grants, as will a school's policies for
awarding grants from institutional
funds. The out-of-pocket net price fur-
ther depends on the level of borrowing
among students and their parents and
other types of aid received, such as
work-study.
Public 2-Year Institutions
Full-time students attending public 2-
year institutions had the lowest aver-
age sticker price ($12,600), net price af-
ter grants ($1 0,600), and out-of-pocket
net price ($9, 1 00) among all undergra-
duates (figure 1 ).
Because students at public 2-year insti-
tutions had a lower average sticker
price initially, their average net prices
also were lower-even though they
had the smallest proportions of grant
recipients (56 percent) and students
who borrowed (23 percent took out a
student loan) (table 2).
6
Public 4-Year Institutions
Tuition at public 4-year institutions was
higher than at public 2-year institu-
tions, but not as high as at the private
institutions (figure 2). Students at these
institutions also have slightly higher
nontuition expenses than those at
public 2-year institutions, with a larger
proportion living on campus or away
from home {see Wei forthcoming).
Among those enrolled full time at pub-
lic 4-year institutions, 60 percent re-
ceived grant aid, about one-half (53
percent) took out student loans, and 10
percent received work-study support
(table 2). Grant aid helped to lower the
net price after grants to an average of
$15,200, and the addition of loans,
work-study, and other aid resulted in
an average out-of-pocket net price of
$10,300 (figure 1 ). This compares to an
average out-of-pocket net price of
$9,100 at public 2-year institutions-a
difference of $1,200 in the average out-
of-pocket net price, even though the
difference in the average sticker price
was $6,300.
For-Profit Institutions
For-profit institutions are privately
owned and operated and the profits
they generate benefit individual own-
ers and shareholders. The programs
can range from less than 1 year to 4-
year bachelor's and graduate degrees.
Most undergraduates enrolled in less-
than-4-year for-profit institutions are
pursuing certificates or associate's de-
grees in occupational training pro-
grams (Staklis forthcoming).
Undergraduates at for-profit institu-
tions also tend to be older, financially
independent, and have family respon-
sibilities. This increases their nontuition
expenses and hence, their average
sticker price. More undergraduates at
for-profit institutions received federal
grants (62 percent) than did students
in any other type of institution in our
analysis (figure 3). However, a smaller
percentage of for-profit students re-
ceived state, institutional, or private
grants than students in other sectors.
On average, these students had a net
price after grants of $25,80Q-not
measurably different than that of pri-
vate nonprofit 4-year institutions, and
higher than that of public institutions
(figure 1 ).
FIGURE 3.
SOURCES OF GRANT AID
for full-time, full-year undergraduates,
by type of institution attended: 2007- 08
Percent
100
80
60
40
20
0
Public 2year Public 4-year
62
For-profit
Type of institution
67
Private nonprofit
4-year
Federal grants ~ s t a t e grant s Institutional grants n Private source grants
NOTE: "Grants" include S<holarships and tuition waivers. full-time, full-year" is defined as having been enrolled in one
postsecondary institution for 9 months or more full time. Estimates indude students enrolled in Title IV eligible postsecon-
dary institutions ln the 50 states, DC, and Puerto Rico. Standard error tables are available at
hnpJ/nces.ed.gov/dasllibrarv/reports.asp.
SOURCE: U.S. Oepartment of Education, National Center for Education Statistics, 2007- 08 National Postsecondary Student
Aid Study (NPSAS:OS).
7
Student loans were critical to reducing Private Nonprofit 4-Year Institutions next highest sticker price: for-profit in-
the average out-of-pocket net price for Even though they had the highest av- stitutions ($25,800 and $16,000, re-
al I undergraduates, but particularly for erage sticker price ($35,500) of the four spectively) (figure 1 ).
those at for-profit institutions. For-
profit institutions had the largest pro-
portion of full-time undergraduates
with at least one loan in thei r financial
aid package: 9 out of 10 (92 percent)
received an aid package containi ng a
loan (table 2 and figure 4}, compared
to 65 percent of those at private non-
profit 4-year institutions, 53 percent at
public 4-year institutions, and 23 per-
cent at public 2-year institutions. The
high level of student borrowing at for-
profit institutions resulted in an aver-
age out-of-pocket net price of $16,000
(figure 1 ).
FIGURE 4.
TYPE OF AID PACKAGE
major sectors, financial aid recipients at
private nonprofit 4-year institutions al-
so received the largest average amount
of total aid ($21, 100), when compared
with those at for-profit institutions
($13, 1 00), public 4-year institutions
($11 ,000}, and public 2-year institutions
($5,400) (table 2).
In fact, aid received by undergraduates
at private nonprofit 4-year institutions
resulted in an average net price of
$25,500 and an average out-of-pocket
net price of $16,600, both of which
were not measurably different from
undergraduates at schools with the
for full-time, full-year undergraduates, by aid package received
and type of institution attended: 2007- 08
Percent
100
80
60
40
20
0
Percent distribution receiving aid
16
Public 2ycar Public 4-year For-profit
Type of institution
Private
nonprofit
4-year
Aid package
Grants, work-
study, or other
aid, with loans
Loans only
;::: Grants, work-
study, or other
aid, without loans
No aid received
NOTE: "Grants" include scholarships and tuition waivers. "Loans" include federal, state, ins1itutional, or private student
loans, excludiog Parent PLUS loans. "Othe(' aid includes job training, veterans benefits, employer aid, and Parent PLUS
loans. "Full-time, full-year" is defined as having been enrolled In one postse<ondary institution for 9 months or more full
time. Estimates include stlldents enrolled in Title IV eligible postsecondary institutions in the 50 states, OC, and Puerto
Rico. Detail may not sum to totals because of rounding. Standard error tables are available at http://nces.ed.gov/das/
librarv/reports.asp.
SOURCE: U.S. Department of EduCiltion, National Center for Education Statistics, 2007-08 National Postsecondary Stlldent
Aid Study (NPSAS:08).
8
The receipt of institutional grants, in
particular, was critical in lowering the
price for those attending private non-
profit 4-year colleges and universities.
About two-thirds (67 percent) of stu-
dents in private nonprofit +year
schools received institutional grants or
tuition waivers, a larger proportion
than at any other type of institution (30
percent at public 4-year institutions, 17
percent at public 2-year institutions,
and 7 percent at for-profit institutions)
(f igure 3). The average institutional
grant received by those attending pri-
vate nonprofit 4-year institutions was
$10,400,
12
which helped reduce the av-
erage sticker price to an average net
price after grants of $25,500 (figure 1 ).
Work-study was also an important
source of aid to those at private non-
profit 4-year institutions. Nearly one-
third (31 percent) of all full-time un-
dergraduates at private nonprofit 4-
year institutions received work-study
aid, the highest percentage among all
full-time undergraduates (between 2
and 10 percent of undergraduates at
other types of institutions received
work-study aid) (table 2). With the aid
of student loans, work-study, and other
types of support, full-time undergra-
duates at private nonprofit 4-year insti-
tutions had an average out-of-pocket
net price of $16,600-not measurably
different from those attending for-
profit institutions ($16,000) (figure 1 ).
12
NPSAS:OS Data Analysis System (data not shown).
3
How do the net prices
paid by undergraduates
vary by family income?
Figure 5 shows the average net price
after grants among dependent under-
graduates by their family income and
type of institution attended. Among
low-income and low middle-income
dependent students, those with the
highest average net price after grants
were enrolled at for-profit institutions.
In contrast, among high middle-
income and high-income students, the
average net price af ter grants for those
at for-profit institutions was not mea-
surably different from those at private
nonprofit 4-year institutions.
FIGURE 5.
NET PRICE AFTER GRANTS BY INCOME
for full-time, full-year dependent undergraduates, by family income
category and type of institution attended: 2007-08
Price
$40,000
30,000
20,000
10,000
0
Low income
Average net price after grants
Low
middle-income
High
middle-income
Highincome
Public 2year D Public 4-year Forprofit 0 Private nonprofit 4year
NOTE: The "net price after grants" Is the price of attending a postsecondary institution after all grants have been received.
This amount is calculated by subtracting total grant aid from the sticker price. The sticker price is the total price of atten-
dance and includes tuition and fees, books and supplies, housing, meals, transportation, and other miscellaneous, or per-
sonal, expenses. Grant aid includes grants, scholarships, tuition waivers, and other gift aid that does not need to be repaid.
The net price after grants is calculated for all students, regardless of whether or notthey received any grant aid. Family
income categories were based upon parents' annual income in 2006. Dollar cutoffs are based on the distribution among all
dependent undergraduates: "low-income" was the lowest 25th percentile (less than $36, 100); "low middle-income"was
the 26tll to 50th percentile ($36, 100-$66,600); "High middle-income was the 51st to 75th percentile ($66,60Q-
$104,6001; and "High-income" was the highest 25th percentile ($104,600 or more). ' Full-time, full-yea(' is defined as
having been enrolled in one postsecondary institution for9 months or more full time. Estimates Include students enrolled
in Title IV eligible postsecondary institutions in the 50 states, DC, and Puerto Rico. Standard error tables are available at
http://nct>5.ed.gov/dasllibrarv/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics. 2007-08 National Postsecondary Student
Aid Study (NPSAS:OS).
9
The average out-of-pocket net price,
on the other hand, shows a slightly dif-
ferent pattern (figure 6). After borrow-
ing, low-income and low middle-
income undergraduates enrolled at for-
prof it institutions continued to have
the highest average out-of-pocket net
price, when compared with those at
other institutions. However, among
those with incomes above t he median
(i.e., high middle-income and high-
income students), the average out-of-
pocket net price was highest for those
enrolled at private nonprofit 4-year in-
stitutions.
FIGURE 6.
OUT -OF-POCKET NET PRICE BY INCOME
for full-time, full-year dependent undergraduates,
by family il)come: 2007- 08
Price
$40,000
30,000
20,000
10,000
0
Low income
Average out-of-pocket net price
Low
middle-income
High
middle-income
Highincome
Public 2year 0 Public 4year For-profit IJ Private nonprofit 4-year
NOTE: The "out-of-pocket net price is the price of attending a post5econdaryinstitution after all aid has been received.
This amount is calculated by subtracting total financial aid from the sticker price. The sticker price is the total price of at
tendance and includes tuition and fees, books and supplies, housing, meals, transportation, and other miscellaneoos, or
personal, expenses. Financial aid includes grants, student loans, Parent PLUS loans, work-study, employer aid, job training
benefi\5, veterans benefits, and any other type of aid.lhe ootof-pocket net price is calculated for all students, regardless
of whether or not they received any aid. Family income categories were based upon parents' annual income in 2006. Dollar
cutoffs are based on the distribution among all dependent undergraduates: "low-income was the lowest 25th percentile
(less than $3-6,100); "low middle-income was the 26th to 50th percentile ($36, 106-$66,600); "High middle-income"
was the 51st to 75th peJcentile ($66,600-$104,600); and''Higlrincome was the highest 25th percentile ($104,600 or
more). "Full-time, full-year" is defined as having been enrolled in one postsecondary instirution for 9 months or more full
time. Estimates include studen\5 enrolled in Trtle IV eligible postsecondary institutions in the 50 states, OC, and Puerto
Rico. Standard error tables are available at htto:l/nm.ed.gov/das/librarv/reports.asp.
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2007-08 National Post5econdary Student
Aid Study (NPSAS:OS).
10
SUPPLEMENTAL TABLES
TABLE 1.
UNDERGRADUATE ENROLLMENT, TUITION, AND STICKER PRICE
for all undergraduates and full-time, full-year undergraduates, by type of
institution attended: 2007- 08
All undergraduates Full-time,
Percent Average Percent Average
distribution Average sticker distribution Average sticker
enrolled tuition' price' enrolled tuition' price
1
Total 100.0 $5,800 $14,000 100.0 $10,300 $22,400
Type of institution
Public 2-year 40.0 1,200 7,000 17.8 2,400 12,600
Public 4-year 29.2 5,500 15,200 42.7 7,100 18,900
Private nonprofit 4-year 13.0 17,800 28,200 20.6 23,400 35,500
For-profit 9.0 10,200 20,600 8.5 11,900 28,600
Other, or more than
one institution 8.8 4,800 12,300 10.5 7,000 18,000
TABLE 2.
FINANCIAL AID
for all undergraduates and full-time, full-year undergraduates,
by type of institution attended: 2007- 08
1
Average tuition and average sticker price estimates
are shown for those attending one lnstiwtlon only.
NOTE: "full-time, full-year is defined as having been
enrolled in a postsecondary institution for 9 months
or more full time. "Tuition includes all tllition and
fees. "Stkker price" is thetotal price of attendance
which includes tllition and fees, books and supplies,
housing, meals, transportation, and other miscella-
neous, or personal, expenses. Estimates include stu-
dents enrolled in Title IV eligible postsecondary
institutions in the 50 states, DC, and Puerto Rico. De-
tail may not sum to totals because of rounding. Stan
dard error tables are available at

SOURCE: U.S. Department of Education, National Cen-
ter for Education Statistics, 2007- 08 National Postse
condary Student Ai d Study (NPSAS:OS).
Any grants Any loans Any work-study Any aid NOTE:"Grants" include scholarships
Average Average Average Average
and tuition waiver;. "Loans" Include
Percent amount Percent amount Percent amount Percent amount
federal, state, institutional, or private
student loans, excluding Parent PLUS
loans. rota!" aid includes grants,
All undergraduates 51.7 $4,900 38.5 $7,100 7.4 $2,400 65.6 $9,100
loans, job training, veterans benefits,
employer aid, and Parent PLUS loans.
"All undergraduates" include those
Type of institution
attending more than one institution.
"Full-time, full-yea(' is defined as
Public 2-year 39.6 2,200 13.2 4,100 3.3 3,000 47.6 3,400 having been enrolled in one postse-
Public4-year 52.9 5,200 46.2 6,600 7.7 2,500 71.3 9.400
condary institution for 9 months or
more full time. Average aid amounts
Private nonprofit 4-year 73.6 10,200 58.9 9,100 23.2 2,100 84.7 17,400 are calculated only for students re-
For-profit 70.4 3,200 91.6 8,100 2.0 3,500 96.3 10,800
ceiving a particular type of aid. Those
not receiving a specific type of aid
(i.e., zero values) are not included in
Full-time, full-year 65.3 7,200 53.1 8,000 13.7 2,300 80.1 12,900
theaverageforthat aid. Estimates
include students enrolled in mle IV
eligible pnstsecondary institutions in
Type of institution
the 50 states, DC, and Puerto Rico.
Standard error tables are available at
Public 2-year 55.7 3.700 22.5 4,900 6.9 2,600 65.7 5,400
Public4-year 60.4 6,100 52.7 7,100 10.5 2,400 78.3 11,000

SOURCE: U.S. Oepartmem ofEducation,
Private nonprofit 4-year 81.2 12,300 65.0 9,800 31.5 2,100 89.4 21,100 National Ceoter for fducation Statistics,
For-profit 71.9 4,000 91.6 9,600 2.0 3,600 96.7 13,100
2007-08 National Postsecondary Stu
dent Aid Study (NPSAS:08).
1l
FIND OUT MORE
For questions about content, ordering additional copies of this Statistics in Brief,
or to view this report online, go to:
http://nces.ed.gov/insert-url/
More detailed information on the price of undergra-
duate education and undergraduate financing can be
found in Web Tables produced by the National Center
for Education Statistics (NCES) using the 2007-08 Na-
tional Postsecondary Student Aid Study (NPSAS:08)
data. These Web Tables are a comprehensive source of
information on financial aid awarded to undergra-
duate students during the 2007-08 academic year. In-
cluded are estimates of tuition, price of attendance,
and financial aid. Additional information on the de-
mographic characteristics of 2007- 08 undergraduates
can be found in a second set of Web Tables.
Web Tables-Student Financing of Undergraduate Edu-
cation: 2007-08 (NCES 2010-162). (link will be added)
Web Tables-Profile of Undergraduate Students in U.S.
Postsecondary Institutions: 2007- 08 (NCES 2010-205).
(link will be added)
Readers may also be interested in the following
NCES products related to the topic of this Statistics in
Brief:
2007-08 National Postsecondary Student Aid Study
(NPSAS:08): Student Financial Aid Estimates: First Look
(NCES 2009-166).
http://nces.ed.gov /pu bs2009 /2009166.pdf
Undergraduate Financial Aid Estimates by Type of Insti-
tution in 2007-08 (NCES 2009-201).
http://nces.ed.gov/pubs2009/2009201.pdf
-------------------------------------- 12 --------------------------------------
TECHNICAL NOTES
Survey Methodology
The statistics provided in this Statistics
in Brief are based on data collected
through the 2007- 08 National Postse-
condary Student Aid Study (NPSAS:08).
NPSAS:08 is the seventh administration
of NPSAS, which has been conducted
every 3 to 4 years since 1986-87. The
NPSAS:08 target population includes
all eligible students enrolled in Title IV
institutions in the United States at any
time between July 1, 2007 and June 30,
2008, representing about 21 million
undergraduates and 3 million graduate
students enrolled in over 6,000 institu-
tions.
The institution sampling frame for
NPSAS:08 was constructed from the
2004-05 and 2005-061nstitutional
Characteristics, Fall Enrollment, and
Completions files of the Integrated
Postsecondary Education Data System
(IPEDS). The sampling design consisted
of first selecting eligible institutions,
from which students were sampled in
the second stage. Institutions were se-
lected with probabilities proportional
to a composite measure of size based
on expected enrollment. Approximate-
ly 1,700 institutions participated in the
study, yielding an unweighted re-
sponse rate of 89 percent. Approx-
imately 114,000 undergraduates and
14,000 graduate students were study
respondents, yielding an unweighted
response rate of 96 percent. Estimates
were weighted to adjust for the un-
equal probability of selection into the
sample and for nonresponse.
NPSAS covers broad topics concerning
student enrollment in postsecondary
education and how students, and their
families, finance their education. Res-
pondents provided data through in-
struments administered over the
Internet, by telephone, or in person. In
addition to respondent-provided data,
data on respondents were collected
from other relevant databases, includ-
ing U.S. Department of Education
records on student loan and grant pro-
grams and student financial aid appli-
cations.
VARIABLES USED
Two broad categories of error occur in
estimates generated from surveys:
sampling and nonsampling errors.
Sampling errors occur when observa-
tions are based on samples rather than
on entire populations. Nonsampling
errors can be attributed to several
sources: incomplete information about
all respondents (e.g., some students or
institutions refused to participate, or
students participated but answered
only certain items); differences among
respondents in question interpretation;
inability or unwillingness to give cor-
All estimates presented in this Statistics in Brief were produced using the
Data Analysis System (DAS). See "Run Your Own Analysis" for more informa-
tion on the DAS and other web-based software applications that enable us-
ers to generate tables for most of the postsecondary surveys conducted by
NCES. The program files that generated the statistics presented here can be
found at [insert-uri-here}. The variables used in these analyses include the
following:
Label
Attendance status
Family income for dependent students
Federal grants
Institutional grants
Net price after grants
Non-tuition expenses
Out-of-pocket net price
Private grants
State grants
Sticker price
Total financial aid
Total grants
Total student loans
Tuition
Type of aid package
Type of institution
Work-study
Name
ATINSTAT
PGDEP
TFEDGRT
INGRTAMT
NETCST3
BUDNONAJ
NETCSTl
PRIVAID
STGTAMT
BUDGETAJ
TOT AID
TOTGRT
TOTLOAN
TUITION2
AIDTYPE
SEGOR4
TOTWKST
-------------------------------------- 13 --------------------------------------
rect information; mistakes in recording
or coding data; and other errors of col-
lecting, processing, sampling, and im-
puting missing data.
For more information on NPSA$:08 me-
thodology, see 2007-08 National Post-
secondary Student Aid Study (NPSAS:08)
Full-scale Methodology Report (NCES
201 0-1 88) (forthcoming).
Item Response Rates
NCES Statistical Standard 4-4-1 states
that " [a)ny survey stage of data collec-
tion with a unit or item response rate
less than 85 percent must be evaluated
for the potential magnitude of nonres-
ponse bias before the data or any anal-
REFERENCES
The College Board (2009). Trends in
College Pricing 2009. Retrieved on
February 11,2010, from
http://www.trends-col legeboard.
com/college pricing/pdf/
2009 Trends College Pricing.pdf.
The College Board (2008). Fulfilling the
Commitment: Recommendations for
Reforming Federal Student Aid: TheRe-
port from the Rethinking Student Aid
Study Group. Retrieved on January 15,
201 0, from .!:!.llQ.;LL
professionals.collegeboard.com/
profdownload/rethinking-stu-aid-
fulfilling-commitment-
recom mendations.pdf.
Harvey, J., Williams, R.M., Kirshstein, RJ.,
O'Malley, A.S., and Wellman, J.V.
( 1998). Straight Talk About College
Costs and Prices. Report of the National
Commission on the Cost of Higher Edu-
cation. Phoenix, AZ: Oryx Press.
ysis using the data may be released"
(U.S. Department of Education 2003).
Several derived variables used in this
report were comprised of variables that
had response rates of less than 85 per-
cent before imputation. EMPLYAM1
and EMPL YAM2 (measures of employer
aid received, and components of
TOTGRT) had response rates of 79 per-
cent and 80 percent, respectively. Non-
response bias analyses showed that for
EMPL YAM 1, there was statistically sig-
nificant bias on 87 percent of the cha-
racteristics analyzed, and for
EMPLYAM2, there was significant bias
on 76 percent of all characteristics.
PRIVLOAN (private loans received, and
lmmerwahr, J., Johnson, J., Gasbarra, P.,
Ott, A., and Rochkind, J. (2009).
Squeeze Play 2009: The Public's Views
on College Costs Today. New York and
San Jose, CA: Public Agenda and The
National Center for Public Policy and
Higher Education.
Lewin, T. (2009, October 20). College
Costs Keep Rising, Report Says. The
New York Times.
Marchand, A. (201 0, January 21 ). Cost of
College Is a Big Worry of Freshmen in
National Survey. The Chronicle of High-
er Education.
Pryor, J.H., Hurtado, S., DeAngelo, L., Pa-
lucki Blake, l., and Tran, S. (2009). The
American Freshman: National Norms
Fall2009. Los Angeles: Higher Educa-
tion Research Instit ute, University of
California, Los Angeles.
a component ofTOTLOAN) had a re-
sponse rate of 67 percent. The nonres-
ponse bias analysis for this variable
showed that respondent s and nonres-
pondents were significantly different
on 80 percent of all analyzed characte-
ristics. PCTDEP had a response rate of
56 percent, and the non response bias
analysis also showed significant bias on
80 percent of all characteristics. For
more detailed information on nonres-
ponse bias analysis and an overview of
the survey methodology, see 2007-08
National Postsecondary Student Aid
Study (NPSAS:08) Full-scale Methodology
Report (NCES 201 0-188) (forthcoming).
Staklis, S. (forthcoming). Web Tables-
Profile of Undergraduate Students:
2007-08 (NCES 2010-205). National
Center for Education Statistics, Insti-
tute of Education Sciences, U.S. De-
partment of Education. Washington,
DC.
U.S. Department of Education, National
Center for Education Statistics. (2003).
NCES Statistical Standards (NCES 2003-
601 ). Washington, DC.
Wei, C. (forthcoming). Web Tables-
Student Financing of Undergraduate
Education: 2007- 08 (NCES 201 0-162).
National Center for Education Statis-
tics, Institute of Education Sciences,
U.S. Department of Education. Wash-
ington, DC.
-------------------------------------- 14 --------------------------------------
RUN YOUR OWN ANALYSIS WITH DATALAB
You can replicate or expand upon the figures and tables in this report, or
even create your own. Data lab has several different tools that allow you to
customize and generate output from a variety of different survey datasets.
Visit Data Lab at:
QuickS tats
Create a simple table
quickly
View )'O'Jr cutp-..rt as a
chart or table
Choose from many data
sets each with about one
hundred Y!lriables
Select from
postsecondary surmes
Cover artwork iStockphoto.<Om!centauria.
http:/ /nces.ed.gov/datalab/
Dotolab
PowerS tats
Produee- complex tables
Run linear and logistic

Choose from many data
sets each with thousands
of variabl&s
library
Search existing tables and
figures to find to your
questions Vistt NCES Tables
sod Figures
Comr119 soon Thousands of
pubhshed tables using
Po,verStats and OuickStats
Help
Need Ccntact
John Vuicka
MPR Assoc.iatas Inc.
d3tala!rhelp@mpnrte com
Aurora D'Amico. NCES
Au.ora gcw
Need access to restncted
data? Learn more on The
Restnctedll se Data
Procedures Manual
-------------------------------------- 15 --------------------------------------
www.ed.gov ies.ed.gov
l(b)(5)
-----Original Message-----
From: Gomez, Gabriella
Sent: Friday, April30, 2010 9:22AM
From: Kanter Martha
To.: Yuan. Georgia
CC:
Date: 4/30/2010 9:52:48 AM
Subject: FW: Transcript of Shjreman speech
To: Rogers, Margot; Kanter, Martha; Miller, Tony; Rose, Charlie; Cunningham, Peter; Martin, Carmel
Subject: FW: Transcript of Shireman speech
(b)(5)
Robert Shireman
Speech to NASASPS
ApriJ 28, 2010
Transcript provided by the Career Education Review
Michael J. Cooney, Editor
mcooney@workforce-com.com
. .. And I had forgotten to put the FAFSA simplification applause line into my actual
remarks so I have applaud when he said it so it never fails and of course- it's not often-
sometimes it's the thing I get questions about but more often its about things like two Pell
awards, two programs in an award year, things more detailed like that. But thank you so
much for that introduction.
Two and a half or three years ago, we started to see a serious economic slide downward
in this country, credit markets had seized up, the sub prime mortgage issue was a major
cause of that and we stmied seeing ppllosing their jobs. We saw ppl in their jobs feeling
much more insecure, much less secure about their ability to invest in higher education,
their ability to buy a home w the collapse of the credit markets and the way to solve that
- long term- is to invest in improving our nations economy, to invest in the kind of
innovation that comes from education, the productivity increases that come from job
training.
Jn order to follow up on that President Obama laid out a bold goal for he country, he said
that by 2002 we want to regain our place as the number one country in the world in terms
of adults with post sec credentials, college degrees, certificates and other job training
programs. In the recovery legislation, now about a yr and a half ago, that included an
expansion of the tax credits that [x] hoping to create in the 90s, an expansion of that tax
credit to $2500, making it for 4 yrs and actually covering more of the types of expenses
that students and families have for higher education. Increases in Pell grants- the usual
approach and what you have seen in your own states, are in an econ downtum, more ppl
are poor, more ppl want to go to school , but instead of following up on that need by
putting more money onto the grant and scholarship programs actually less money goes
into the grant and scholarship programs be of the state budgets.
Fed govt took the opposite approach really, what needs to be countercyclical spending
that helps- like unemployment insurance- spending that needs to follow up on and help
to address the new gaps that families are seeing - so the tax credits were part of that, the
increases in Pell grants are not only meeting the new demand for Pell grant dollars but
actually increasing he size of the Pell grants and proving those increases into the future
with the follow up legislation passed a few weeks ago. Also restoring some certainty to
the student loan program, and making sure that no one has any reason to doubt whether
they will be able to get the federal student loans that they need, again with the refmms
that were implemented a few weeks ago.
I mentioned that when people are losing their jobs, when ppl become insecure in their
jobs, they look for higher education, they look to find what kind of job training can I get,
what kind of skills can I add to my repertoire, what are the skills that I have, how can I
make them better so that ill keep my job, so that ifl lose my job I'll have options. And at
the same t ime, while we saw this increase in demand, which is helpful and useful given
what the President had to say about the need to train our population, we saw state tax
revenue declining in all but a few states, we saw resulting cuts in the budgets of state
colleges and universities and community colleges, resulting in a combination of very
large increases in tuition in some cases and reduced enrollments, fewer seats. So
increased demand - ppl wanting more education and training, and public institution either
had fewer seats and charging more tuition or might not declare going to be enrolling
fewer ppl but their course offerings are cut, fewer kinds of course offerings, so the result
is they are not able to demand for higher education.
Tuition-driven institutions didn't react that way because they're tuition-driven institutions,
and the non-profit institutions have done pretty well despite significant declines in their
endowments because there continued to be significant demand for higher education. The
public., the non profit private colleges did well and in particular the for profit institutions
have come in with investors making sure that there was capacity to be able to serve
additional students, and they knew that those students would come with those federal
dollars, Pell grants, student loans, tax credits and that hat would help them to not only be
consumers who want higher ed but consumers who can pay for that higher education with
that federal support so the for-profit industry, more than any other in this economic
difficult times, has responded.
I want to give you some specific numbers, we now post now on one of the [X] websites,
the quarterly numbers ofPell grants by different kinds of schools s I looked at what the
first 3 quarters- the total of the first 3 quarters of this award year compared to the last
award year for some of the schools that I knew would be here today. So for example,
Corinthian Colleges- 38% increase for first 3 quarters this year compared to last year for
a total of $800M
DeVry- a couple people here from DeVry?- 42% increase up to $1.78
ITT- you guys here? A 44% increase up to $623M
Strayer- still here? Is that you? Well this one- 95% increase, may be something about
the quarters, but up to $414M
APE!- Wally here? And Russell? 94% increase up to $44M
Kaplan- they here? So this total is actually all the Washington Post owned entities, 33%
increase up to $909M, and again this is the first 3 quarters of the year so the totals for the
year are obviously more than that
Career Education Corporation- 29% increase up to $1B this first three quarters
EDMC- several folks here; a 16% increase, $1.1B
Capella- over there? 40% increase to $378M
And I think I've just got a couple of others:
Grand Canyon- 55% increase to $260M
And University of phoenix- you there?- 9% increase but obviously that's on a larger
base. So probably that increase is as much as a lot of others' total dollars, and that
increase is $2.7b total
And Bridgepoint- you guys here? - 6 l% increase, $393M
I think those were all that I had numbers for, obviously I know that there's a few others
here as well.
So I wanted to begin just by thanking the for-profit industry for responding to the critical
demands from ppl out there who need higher education. I'd like everybody to give them a
hand. Now, others of us in the room have the responsibility for making sure those federal
funds I just listed, for education and training, that it's all totally above board. That those
significant increases in fed spending for higher education- loans, grants- are serving
students and tax payers as well as they possibly can. and that is what the Triad is about-
and I know I can say triad in front of this audience because I heard somebody say it
earlier. I want to talk for a second about some things going on in Washington right now,
and I don't mean negotiated rulemaking- I will get to that in a few minutes, but there is a
wall street reform debate going on right now in Washington. What happened in that
credit crisis a couple of years ago had something to do with credit rating agencies-
agencies like S&P, Fitch, other agencies that were responsible for rating instruments-
:financial instruments, looking at what is the quality of these things that have names that
cause people's eyes to roll over- things like collateralized debt obligations, and other
kinds of securitizations- so what is the quality of the loans, mortgages, are they going to
be repaid, how likely are theses loans to be repaid so that an investor purchasing this,
how confident can they be that when they purchase, when they invest in this particular
instrument, that they will get the money back that they are expecting.
The business model for these rating agencies has come under fire in these meetings in
Washington, part of this has to do with the business model of the rating agencies, on the
one hand, their responsibility, their job, the core of their business was to make sure they
did a good job providing an honest rating tor the instrument that they were analyzing. On
the other hand, they relied on the income from the companies who asked them to rate the
instrument, and ill read to you from- a NYTimes- some of the emails that have been
coming out recently.
In 2004, well before wall street's bets on subprime mortgages became widely known,
employees at Standard & Poor's credit rating agency were feeling pressure to expand the
business. One employee warned in an internal email that the company would lose
business if it failed to give high enough ratings to collateralized debt obligations, the
investments that later emerged at the heart of the financial crisis.
Quote, " we are meeting with your group this week to discuss adjusting criteria for rating
CDOs of Real estate assets be of the ongoing threat of losing deals. Lose the CDO and
lose the base business. A self reinforcing loop"
In other words, if we don't loosen up, we don't loosen up in our assessment of these
instruments, nobody is going to come have their instruments assessed by us anymore.
And this created a conflict which led to instruments that should have been questioned not
being questioned, and [leading] over to the financial crisis that we have been suffering
from for the past couple of years.
The other issue besides the business model was the complexity and fast growth of diff
kinds of instruments and ill read from another of the recent articles. "Email. documents
and other messages suggested that executives and analysts at ratings agencies embraced
new business from Wall Street even though they recognized that they couldn't properly
analyze all of the banks' products. And one of the other quotes ends with, "we were so
overwhelmed."
So I want to actually ask, on that issue, the complexity and growth, and I know we're
feeling this with publicly traded corporations and purchases going this way and that way,
and we're trying to figure out what's going on . Are there regulators in the room who feel
like you DO have the analytical firepower you need to assess what is going on with the
entities you regulate in higher education? Those who do feel you have the firepower you
need? I don't think we feel we have the firepower we need.
So the reform back on the financial instrument side of the equation, what they're really
talking about now in Washington on financial reform, one analyst- an academic looking
at what's going on, said it only tinkers with the workings of the ratings agency, it doesn't
end the inherent conflicts of interest, those cont1icts of interest where the ppl who do the
rating are paid for by those who do the ratings. This whole situation w credit agencies,
credit rating agencies, is, as I see it, very similar to the way accrediting agencies work in
this country. The same kind of inherent conflict of interest. Albeit accrediting agencies
are nonprofit and on top of that, what would this crisis look like ifthe banks had actually
been the ones running the credit agencies and were doing a peer review kind of model,
which is the model we have in accreditation, where it is the regulated who are really
looking at each other rather than an outside entity.
So to borrow from Winston Churchill, accreditation, as a part of that triad, in terms of a
way of assessing quality in higher education, is the worst form of accountability except
for all of the others. What Winston Churchill actually said was, democracy is the worst
form of government, except for all the others. So I am bringing up this issue of
accreditation not to say that we should back away from it or change it, 1 actually don't
have a better system for us for assessing quality in higher education. But it is problematic
and we need to remember that as the other two pieces of the triad, as we figure out how
we can do the best job possible in our responsibilities. Federal and state governments
cannot rely on accreditation to insure that consumers and taxpayers are protected t the full
extent that they need to be- all three legs of that three legged stool need to be working
and working well.
There are a number ofthings that we're doing, you've heard about some ofthem-
elevating, monitoring and enforcement, we're working with the inspector general at the
department of education, taking a much closer look at data than ever before to help guide
our selection for program reviews and investigations when necessary by the inspector
general, working with he federal trade commission to join their consumer complaint
system so complaints they get and other agencies that are on their consumer sentinel,
working on the issue of how we can look more at issues of misrepresentation as we do
program reviews and other kinds of monitoring.
A second area besides the monitoring and enforcement is improving consumer
information. We have put graduation rates, retention rates and transfer rates right on the
F AFSA form when students are choosing colleges, the rates are right there as a reminder
to students that they should do some good shopping, look at various kinds of data that
might help them to compare schools. We're also providing them with a more detailed
financial aid estimate in terms of the financial aid that they can get and this is partly to
make sure that people know they can get that aid wherever they go. Sometimes students
think, oh I can get that $12K because the school costs $12K, and I would only get $3k at
a community college that only cost $3K, not realizing hat in fact if they wanted to get
more than what tuition costs at that community college so that they can focus on their
studies instead of working excessive hours, that that is something that they can have
available to them.
And starting this summer, as a result of a regulatory process that has already completed,
schools will have to begin providing placement information and where the have
placement rates, they actually will need to make students aware on their websites of
placement rates they have for programs that they are offering.
Uh coordination and sharing, I head some of the discussion in prior sessions and I look
forward to this af-temoons discussion. Within the federal government, we are working
with the Fed trade commission, the veterans administration around the GI bill , the SEC
because of the involvement of publicly traded institutions, states we have encouraged
involvement in this group and are looking for other ways that we can help. Happy to
discuss that in q&a here as well as this afternoon because we really need to become good
partners if we' re going to do best by taxpayers and students.
And accreditors, there are some new requirements, we're working on sharing some draft
guidance related to all of the requirements for accreditors and again building that triad
and all working together. In fact, I was actually on the internet looking for a three legged
stool to see if [ could bring one for this, and [ noticed one of the three legged stools had a,
not only the three legs, but it had this connecting piece of wood that held the three legs
together, and I thought, well that would be the perfect prop, because that would
demonstrate it's a strong stool if that connecting- that connector is there, making it as
strong as possible.
Also, may of you have heard, reviewing the rules and regulations and where appropriate
revising, in the process of revising those rules. Let me take a I ittle bit of time to tell you
about some of those. We started about a year ago, doing public hearings where we
basically said, we want to know whether we need to in1prove program integrity, are thee
things we need to be doing? Here's a list of some areas, misrepresentation- definition of
credit hour, state authorization, other kinds of things, and we saw input. We did 3 public
hearings, ppl were able to submit items over the internet, through email, and we got a lot
of input about great schools out there, students who were having a good experience,
people who attended the schools, got a job, had a great experience. We also heard from
forn1er students who felt that they were misled, legal aid attorneys who had clients whose
stories were cause for concern.
That was followed up by- we asked for nominations for ppl to serve on committees- the
way this whole process works is that we do our best to work through possible rule
changes with a committee of stakeholders, recommended, nominated by interested parties,
states, various institutions, student organizations, legal aid.tr 3 weeklong sessions,
December, January and February went through each of 14 issues talking about, changes
that might make sense.
One of them, misrepresentation, clarification really against misrepresentation by schools.
High school diploma- one of those things that you take to somebody and think, how
hard can it be to know if a high school dimple is valid?? As you know, not that easy and
many of you are at the state level so you know that the state isn't necessarily declaring
who is good or bad. And the issue of the federal government declaring what is a valid HS
education, for example, gets into areas where the def government isn't supposed to be
declaring such things, so I think a more complicated issue than I think a lot of ppl
expected. We are making - at least in NegReg session - reached some tentative
agreement around the definition.
I would say the most significant thing we are doing is looking at- and I think this is now
a likelihood, when people apply on the F AFSA and it asks for a HS diploma, a list will
actually pop up and they can enter what that high school is, the name of that high school
based on some feral lists we have. It wont necessarily mean that it will be a valid high
school, but it does give us and you the ability to, if for example a suspected diploma mill,
we would be able to see who are, where are the students going who are using this
particular high school as the place they say they get their diploma from. And if we find
that its some particular colleges, that means that it might b encouraging ppl to go and use
and diploma mill. So it will be a useful took for us and you as well. , and that's the most
important change we'll be making there.
Incentive compensation was a major issue, the issue of paid recruiters. A number of
years ago, a number of safe harbors were created and there was a lot of indication that
they were wider loopholes than are appropriate given the wording of the actual law that
prohibits payment of actual compensation based on enrollment. So that's another one that
we are working on.
State authorization, I heard California mentioned and it was a surprise to me when I came
to Washington and asked about California to discover that a legal interpretation of the
Dept of Ed, well if the school is not not authorized, then it is authorized. So this a raised
question that came up in NegReg about what is at least some minimum standard about
what kind of authorization should count in terms of the state role in that Triad.
Satisfactory academic progress is another area taking attendance. What] used to call
R2D2, return to tile 4. and I'm not mentioning all of the issues, but the final one I will
talk some about is Gainful Employment, and this is the one that's been in the news a lot.
It seems that every time I speak somewhere, something thinks I said something new and
calls a stock analyst who then reports it causing the stocks to go up or down or whatever,
and I assure you I am not going to say anything new. If you are a stock analyst or you
know a stock analyst, the answer when they ask you, What did Shireman say? You say
nothing new.
So the statute, the federal law requires that in order for some programs to be eligible for
federal financial aid, they have to lead to gainful employment in a recognized occupation.
This applies to non degree programs at any type of school and it applies to most programs
at for-profit schools, really all except some BA, liberal arts programs through an
exception, a recently enacted exception, that actually begins this July 1
51
but for the most
part, a for-profit institution, in order to be eligible for federal financial aid, has to show
that the program leads to gainful employment or prepares the student for gainful
employment in a recognized occupation.
So a year ago, we began asking the question, what is the definition, what should db e the
definition of gainful employment in a recognized occupation. We had hoped that perhaps
some schools would come forward and say, well when we start a program, here's how we
determine whether or not it complies .. we didn't get that kind of information.
We brought it up in NegReg and made some suggestions for discussion. We suggested,
maybe there should be some relationship to the debt levels that students are taking on and
the expected earnings that they may have from the occupations that you have identified
that you are preparing people for. We also suggested that perhaps a loan repayment rate
approach could be devised where we would be able to see that federal loans are actually
being repaid at a rate that makes sense if ppl were actually gainfully employed.
We looked at the provision and current regulation that currently applies to very short
programs, the 70/70 mle. So 70% completion rate, 70% placement rate, and asked should
something like that be part of the definition of gainful employment? And then for new
programs, we suggested maybe there should be something from an employer, who
employs ppl in the occupations that the program is preparing people for, that at least
asserts that yes, the cuiTiculum, the program that ive seen at this school is designed in a
way where it would prepare people for the jobs that I have in my particular business, so
we suggested that for new programs.
Now in evety other issue in negreg, we got pretty good discussion at the table, sometimes
we actually got consensus from the group on what we should actually, how regulation
should actually be worked out. But for some reason on the gainful employment issue, we
didn't get the kind of discussion that would at least help to guide in a very constructive
way the direction, and to know well, this would be okay with certain kinds of schools but
wouldn't be okay with other kinds of schools.
Instead the reaction from, in particular those who were representing the for-profit
colleges, was you cant do this, you cant define this term, why are you doing this, and that
continued even after the NegReg sessions. We continued to meet, we have gotten
improved input, improved feedback. And where things stand now with whole regulatory
packed, so everything I've just discussed now including the gainful employment, is that
in the next few weeks there will be proposed rule published in the federal register. There
will be a comment period after that proposed rule is published. That will be the
appropriate time to suggest changes or express support for provisions, suggest
alternatives, and then a final mle, our goal would be to publish a final rule by November
I
51
For rules to take effect in general. next year from this July, they need to be published
by November 1
51
So that's where we will be, that's the tin1eline for the rule going
forward.
I wanted to conclude my remarks before going to some Q&A and some discussion with a
piece that Thomas Frank wrote in the WSJ. The title ofthe article is, "Obama and the
Regulat01y Capture," and it is again, back about the financial regulation.
"It was not merely stmctural problems that led certain regulators to nap through the crisis.
The people who filled regulatory jobs in the past administration were asleep at the switch
because they were supposed to be. It was as though they had been hired for their
extraordinary powers of drowsiness.
The reason for that is simple: There are powerful institutions that don't like being
regulated. Regulation sometimes cuts into their profits and interferes with their business.
So they have used the political process to sabotage, redirect, defund, undo or hijack the
regulatory state since the regulatory state was first invented."
So, he follows that up with one more line here, "And it created a situation where banking
regulators posed for pictures with banking lobbyists while putting a chainsaw to a pile of
regulations. Smiles all around. Let the fellows at IndyMac do whatever they want."
So my closing words, is, we should take the photos, we should smile, but lets not shirk
our responsibility for regulating the industry. The schools will make plenty of money and
students and taxpayers will be better off if we do our jobs as best as we can. Thank you
very much.
Host - Some time for questions and answers, and I know gainful employment will come
up, but first I want to just make a comment. I think most of the states, clearly the states in
the room, because the ones that aren't in the room, there might not be a cognizant agency
that actually is involved in the regulation, if it's a registration with the secretary of state
or what-have-you, but fortunately, there is another state within our state that someone can
go get a better rate at or what have you. So I think that's clearly one thing that I was
thinking about. There is an alternative on per se that the school has if they want to be
serving your students, they want to be in your state. They have a set of laws, a little bit
different, but 1 certainly appreciate the parallel, and I think important to add-
Shireman- I was really more talking about the parallel with accrediting and it's not as
much of an issue with states
Host- right so you know, 1 sympathize with that, good parallel. Did the Dept, was there
any discussion when gainful employment issue was being looked at, to say, can somehow
the Dept tap into infom1ation at the Dept of Labor or at the prior [X]. I understand there
may be some challenges there, but we're trying to measure whether or not this individual
is 1, either employed or perhaps employed and had benefited because of the education or
training that they had received and somehow reached their salary or what have you. Was
there any discussion around that area, because it seems like there may be some ways of
doing that?
Shireman -Yes, one of the areas of discussion, so part of the areas we are looking at.;
some of the criticism that has been out there has been about the use of averages or 25lh
percentile of BLS occupational data. And the reason we suggested use of BLS was not as
- it was actually as a way to reduce the work that would need to be done by whoever is
coming up with the income information. So for example, it's going to be the schools
having to figure out what their graduates eamed, we figured why not just have a level
that's an average based on the industry that will eliminate a lot of the school's need to
follow up and look at their own graduates. In other words, if it meets- if the debt I
income ratio is fine given the earnings in the occupation generally, you don't need to
worry about it as a school. Just look at your debt and if it's okay, flne. You only need to
look more detailed if you don't, if it doesn't beat it that way, because maybe your
graduates are earning something more. It was really to try to reduce the number that
would have to go through a process, either using IRS data or a state data system. States
like FL are in a much better position to be able to look individual programs and have this
kind of information using their U[, unemployment insurance, database. So there's a lot
that can be done there, and the discussion at and since negotiated rulemaking has helped
us to think through those issues.
Host- know there's going to be some questions ...
Michael Cooney- Bob, [patt] of your presentation is, do you believe that for profit
institutions are serving their students well?
Shireman -1 think they have to be given their demand, given the number of people that
are going to for-profit institutions, I think it's, I think they are, absolutely. All of them for
all students, I think that's where our role comes in
Cooney- following up on that, then your gainful employment provisions as we
understand it, according to CCA would approximate 300K students from the possibilities
of receiving an education. Is that an unintended consequence or are you all cognizant of
that?
Shireman - there isn't a proposal, and my argument with the- and ive had these
conversations directly with CCA, is work with us about what the definition should be. So
we can get in- what they've done is say, 8% pure labor statistics, these folks are crazy!
But the reality is, thee only way to get to a rule that makes sense, and we need to have a
definition, is to get down in the weeds and start working out the details. And they and
others have come back with, id say some more constructive thoughts in recent weeks, and
those will be considered in the proposed rule, and then they' ll be further input that I hope
we' ll get in a final rule.
[Ken Miller] -Mr. Shireman, im concerned about national averages. Some states are in
much better shape employment-wise than others. I'm fiom Ohio, a county in Ohio that
has 17% unemployment. Question - how can you adjust this rule, this definition, for
those kinds of anomalies between states, and should you?
Shireman -So using actual data from students as was suggested would deal with that
completely. The BLS data actually is available on a state basis so that could potentially
be another option. So this is where we are, we're open to suggestion and ideas about what
would be the best way, or whether it should be various possibilities that a school might
have an option to use. This is that input-seeking process for what would be the right kind
of approach. Again, part ofthe reason we just used the national average was because, to
eliminate a whole lot of work so that fewer needed to look in a more detailed way, and in
some ways I suppose if we had not suggested an approach to reducing work, we wouldn't
have had the BLS stuff on the table and wouldn't have the criticism, but I think it does
make sense, because it was a reasonable, work-reducing item that we put on the table
during negotiated rulemaking.
[Tom CasseJJ(?)] -have you looked at the idea of a value added concept in the
definition?
Gainful to me means you've gained, improved. So if we looked at our students and
where they came to us and what their earni ngs were afterwards, a measurement of change
of gain as the definition?
Shireman - so specific suggestions about that proposal and how to do it, we welcome
that, I mean that's very useful. So in the proposed rule period, that's especially timely.
Make sure, Especially from this point forward, make sure that ideas are considered, that
comment period after we publish the proposed rules. That's the kind of thing that would
be great to work out the details and suggest it during that comment period.
[Keith Boss (?), Northwest Technical Institute) - layers of control can sometimes
create layers of bureaucracy as you well know, you've got to struggle with that. There's
al ready of default rate, and we're very proud of our very low default rate, was it ever
considered that that 's enough control to see that students have the ability to pay offtheir
student loans and why would there need to be more layers of control, more bureaucracy
than that?
Shireman- the discussion at the table around default rate had to do with the fact that the
federal government has made sure that ppl who take out at least federal student loans
aren't put in a situation where it's a choice between food on the table and paying rent and
paying off your student loan. So we have things like forbearance, and if you ask for
forbearance, you could make no payments at all and you're not in default, there's income
based prepayment where you may be making nothing or next to nothing, but if you're on
income-based prepayment, there's very low repayment or maybe nothing at all and
you're not in default. So default rate doesn't measure whether or not people are actually
earning any money.
[Ray]- one of the problems ppl have come to wrestle with this concept is that any
fommla that is derived that results in the elimination of eligibility for a program has the
potential to have a school on the NE corner vs. a school n the SE corner delivering the
same program at the same level of efficiency and quality, but merely because of the
borrowing characteristics of their student population, one school's program could be
declared ineligible while the other school's program continues ... further, as you go
through this and eliminate program eligibility, the students don't go away, they seek out,
in many cases, the same education, in another institution. You could end up with a deluge
in the public institutions because they're not subject to the sanctions of a loss of
eligibility and yet, so it's okay to continue to train that person and for them to continue to
have debt and for state dollars to continue to subsidize their education at a public
institution but meanwhile the resulting product is the same?
Shireman- so what's your suggestion for a definition of gainful employment?
[Ray]- well I don't know- I always thought it was defined in the IRS code, the Social
Security admin ....
Shireman- I need a citation, tell me what we can do, as the US Dept of Ed, to detennine
whether a program is in or out.
[Ray]- I honestly don't know
Shireman- this is the problem. If you cant give me a definition, we're not making any
progress. So anything- I can respond to the details of your question, things like the 90/10
rule - any time you draw a line, there will be people right on one side and people on the
other side of the line that are very similarly situated. But that's what lines are about, so
your suggestions about what would be, and I think asking your colleagues, your own
institution, how do we decide, how do we know our programs are eligible for finical aid,
would help for you to develop some suggestions for us.
Host- see I knew we should have had a workshop on gainful employment.. we've all sat
around trying to figure it out. .. we have some good thinkers in the room ..
Russell - APEI -I have sort of a detailed question im going to pose for you. You
mentioned earlier [xx] with regard to title IV funds, but those fund are not necessatily
earmarked specifically for education costs, or direct education costs, and I think we all
recognize the reality is that many student do in fact realize substantial amount of money
for impotent, incidental spending, a new car, walking around, whatever. My question is,
lets take a scenario where an individual sent to community college for a couple years
borrows $35-40k, then that graduate comes to APEI system, where it will cost you $30k
to get a BA degree from start to finish, we inherit $35-40K worth of debt and add it to the
$30k the student incurs with us, we've got $75-80k of debt. Is that going to count against
us for the purposes of calculating gainful employment?
Shireman -we haven't proposed a rule yet but what we discussed a the table in
negotiated rulemaking, the answer would be no. That's because we proposed using the
median, and using the median means that all those outliers don't pull up the average. It's
the median, schools don't generally have more than half of their students coming in with
huge amounts of debt from someplace else. Now there's also the option in the rule of
saying, it's only debt you incur at the institution, so in the proposed rule in the comment
period will be the time to look back to a particular issue. There's debt/ income ratio in
the rule, and you can provide commentary. Again this is the kind ofthing, the kind of
discussion we needed a year ago and have begun to get recently and I really do appreciate
because they're useful and constructive discussions to have.
Russell - well the reason, ifl can just add another comment to that, I think it's important
to have [other debt] considered because if you take that approach that im only
accountable for my own debt-
Sh.ireman- Right, so to be clear, it's just a proposed rule. Comment period, final rule.
So we've gotten this kind of input, they'll be a proposed rule. If we haven't addressed
this issue and it needs to be addressed further in your view, there will be another
opportunity for you beyond the opportunities you've had in the last several weeks.
Russell - The one thing I would like not to see happen is that for admissions staff to say
we don't want to admit you because you've got debt.
Shireman - understood.
[NA] - can you tell me how this would apply to a public university or state college?
Shireman- well what we discussed at the table, in negotiated rulemaking- and I keep
repeating that because the "this"- we haven't proposed anything, we brought discussion
items to the table during Negotiated rulemaking, but it would mean that for shorter tem1
certificate programs, if more than half of the completers had debt, then the median debt
level of all completers, in other words not just the debt level of those who borrowed, but
the median debt level of all the people who complete the program, that would be the debt
level that would be compared against the expected earnings in the occupation the
program is preparing people for or the set of occupations. And then the discussion earlier
was about well, do you use actual earnings for people who are in the program or do you
use averages or 25th percentile with BLS so there's that question about what the measures
of income should be, so that would be - the concept was that as a standard. And again, at
the table we also had other suggestions for an institution if they didn't meet that
particular measure like graduation placement rates and loan prepayment rate.
LNAJ -so are you going to be asking them to post their placement rates as well/
Shireman- well placement rates have to be posted starting this July anyway, or
placement information. if you have rates, rates need to be posted.
LNA ) -so if they don't do it now, if they don't have placement info now . ..
Shireman - placement information was done as part of a rule done in - Congress
required all schools to post placement information in the Higher Education Opportunity
Act, and that goes into effect under the rule that was adopted last year so that goes into
effect this July.
[John Weir(?)] -from a state regulatory perspective, appreciate the Dept's effort to try
to address the cost issue, and one of the things that we've found at the state level is that
students don't seem to be price conscious when it comes to education. And what do you
think we can do at the state level to help make students better consumers, particularly as
to the cost issue?
Shireman- I think we've seen some improvement in that because of the economic
downturn, are seeing families chop more and really look at, does it really make sense for
me to spend $30-40K a year, what about the state university ... what about other options?
So I think it has improved somewhat, but more needs to be done. There will be, again
because of the Higher Education Opportunity Act, starting I believe this summer, schools
will now post net tuition - I cant remember if it's net tuition or net cost of attendance-
anyway some net figures by income bands so people will actually be able to go and say,
for someone in my fami ly income background, it's more common to pay this price rather
than the other price. And starting in, I think it' s a year and a half, all colleges will have to
have a net cost calculator, where a person can put in their income and other basic
information and get a basic sense of what their net cost would be. So that helps to deal
with some of the issues, we've provided some more tools to help people get more
information and to help them also deal with reality that sometimes a college that looks
really expensive based on its sticker price for a particular family, especially lower income,
may not actually be that expensive. So there's some new tools coming forward. I do
worry that just more information on the internet doesn't really help that much, and that
figuring out who we can provide tools more directly to teachers, counselors, people at
one stop job training centers. You know, our website, when people are actually applying,
finding those moments, those teachable moments when people are actually thinking
through these issues, so suggestions you have about the ways that we can do that, Pell
programs, state programs- we're certainly interested in ideas. It's important and not
easy to figure out how you teach the vast public things that aren't simple.
[Julie]- I guess this is more of a comment, but im somewhat concerned about the
potential for consumer confusion with the net cost calculator and different types of
disclosers about tuition. I know at the state, I'm in Tennessee, we've required that our
enrollment agreements contain the total cost of the program for a very long time. We now
require that the tuition rate be posted on the internet, we've required disclosure of
completion, referral rate, placement rates for a good while. We now do in depth audits of
placement data, so if student looks at our website, it's likely the numbers that they'll see
are going to be very difierent from maybe what they see on the F AFSA or some other
disclosure on the federal side. So how do we address the very likely possibility of
confusion, not knowing different methodologies and standards?
Shireman- I think you're right, there is going to be confusion. And even me, when I
:first started talking about net tuition, I couldn't remember, is it net tuition, is it net cost?
What's included in the definition? And I think there will be confusion, I guess the
positive side of that is, usually when you take a step and it creates those problems, it then
forces us all to sit down and say, how can we improve on what we had come up with
originally? Sort of like, data that's not perfect. Like, we cant post that data, it's not
totally clean. But the way to get people to clean their data is, you post it. And when we
started posting the student debt data that universities had reported to US News & World
Report, and universities would call and say, well that's not our figure and we'd say, well
that's what you reported to US News. And they'd go back, and sure enough, somebody
had reported something wrong, s I think that you're right, there will be some confusion
but that will probably force the issue to some degree. But id also welcome- I mean I'd
love to look at what Tennessee does and take it to some of our folks and ask about what
we can do, A lot of- for some reason Congress in the education space, especially the
higher education space, gets very detailed in the legislation's actual statutory language,
which then makes it very difficult for us in the implementation to do things differently
that what they had prescribed. So we may not have a lot of flexibility in terms of our
implementation but maybe there's some suggestions we could make back to Congress by
pointing out to them these issues for particular states.
[Ray]- there seemed to be a reluctance on the part of Dept officials as well as most
people in the room during the NegReg process to talk about the troublesome aspect of
schools not being able to limit student borrowing,. It seems that we have a statute, a rule
that leaves the amount that students borrow entirely up to them. It's based on fom1ulas,
numbers across[] program lengths are mandated by state laws in many fields of training,
[licensure], all these things are out ofthe control of the institution, they cannot seem to
limit the amount that students borrow except in very specific case-by-case basis, and then
they subject themselves to lawsuits by students claiming they were discriminated against.
Might there be any opportunity to open up and look at some of those rules so that we
might be able to avoid some serious unintended consequences of a formula that could
cause havoc?
Shireman- student loan program and the Pell grant program are entitlements. They are
entitlements for covering tuition, fees, books, supplies, room & board, and to restrict an
entitlement program is a pretty big deal, and that's why it's difficult for us to do in tem1s
of regulations. Now you do have the ability, if you believe someone would default on
their loan, I think that's the ability that the fin aid administrators have ... you're also able
to establish what you feel to be reasonable expenses including in the room & board space.
But restricting loans beyond that would likely be something Congress would need to
consider, and I think part of the reason Congress has had a hard t ime thinking this one
through is higher education has benefitted enormously from Pell grants and student loans,
Pell grants working like an entitlement program on an annual basis, and student loan
programs being an entitlement program. And going down the road of starting to restrict
the entitlement, I think is something to do very cautiously if we want to make sure people
get the need they are eligible for.
END.
(b)(5)
Harold
From: Shireman, Bob
Sent: Thursday, January 07, 2010 1:09PM
To: Jenkins, Harold
Subject: UOP meeting
Importance: High
(b)(5)
From: Julie Shroyer
To: Shireman, Bob
Sent: Thu Jan 07 11:54:23 2010
Subject: Hello and Meeting Request
Dear Bob,
From: Jenkins Harold
To: Marinucci. Fred
Wolff Russell
Woodward, Jennifer
Finley Steve
Sann, RDnald
CC:
Date: l/7/20102:16:50PM
Subject: FW: UOP meeting
I hope the new year is treating you well. Just left you a voice mail message but realize it may be easier to send you a note
via email. I Nonresponsive I
I"'"'"'''"''"
As I indicated in my voice mail, I would also like to submit a fonnal meeting request for Apollo. We are hoping that you
may have time in your schedule sometime over the next few weeks. Here are the specifics:
Meeting Partic1pants:
Greg Cappelli, CEO, Apollo
Terri Bishop, Director/Vice President External Affairs, Apollo
Julie Shroyer, Sr VP, Wheat Government Relations (consultant to Apollo/University ofPhoenix)
Purpose:
To discuss ideas for implementing various student/consumer protections (e.g., related to borrowing practices,
disclosures, etc.) and the technology that Apollo is developing to support compliance and academic quality and
innovation.
When:
Preferably as soon as possible or within the next few weeks.
Thank you so much for your consideration of the request. I look forward to hearing from you and seeing you soon.
Best,
Julie
Julie E. Shroyer
SeniorVP
Wheat Government Relations
1201 S. Eads Street, Suite Two
Arlington, VA 22202
(703) 271-8760
j shroyer@wheatgr.com
(b)(S)
-----Original Message-----
From: Mitchelson, Mary
Sent: Friday, September 03, 2010 3:34PM
To: Holland, Linda; Minor, Robin
Subject: RE: Visit with Bill
(b)(5)
-----Original Message-----
From: Holland, Linda
Sent: Friday, September 03, 2010 2:46PM
To: Minor, Robin; Mitchelson, Mary
Subject: FW: Visit with Bill
(b)(S)
-----Original Message-----
From: Holland, Linda
To.: Yuan. Georgia
CC:
Date: 9/3/2010 3:35:02 PM
Subject: FW: Visit with Bill
From: Joseph D'Amico [mailto:joe.darnico@apollogrp.edu]
Sent: Tuesday, August 24, 2010 8:05PM
To: Holland, Linda
Cc: Gregory Cappelli
Subject: Visit with Bill
Hi Linda .. .I hope you are well. Our Co CEO, Greg Cappelli, who Bill has met, and I would like to set up an
appointment to meet with Bill and review with him some of the things we are doing to address the GAO report and
related matters. If we can get an hour of his time, we would really appreciate it and I know it would be of value to him,
especially with all that is going on in the sector. We will do our best to meet his dates. Thanks and look forward to
seemg you soon.
Warm regards, Joe
Sent from my iPad
This message is private and confidential. If you have received it in error, please notify the sender and remove it from your
system.
(b)(5)
G
-----Original Message-----
From: Holland, Linda
Sent: Friday, September 03, 2010 3:35PM
To: Yuan, Georgia
Subject: FW: Visit with Bill
(b)(5)
-----Original Message-----
From: Mitchelson, Mary
Sent: Friday, September 03, 2010 3:34PM
To: Holland, L i n d a ~ Minor, Robin
Subject: RE: Visit with Bill
(b)(5)
-----Original Message-----
From: Holland, Linda
Sent: Friday, September 03, 2010 2:46 PM
To: Minor, Robin; Mitchelson, Mary
Subject: FW: Visit with Bill
(b)(5)
-----Original Message-----
F.-om: Yuan Georgia
To: Wolfi Russell
CC:
Date: 9/3/2010 5:33:16 PM
Subject: FW: Visit with Bill
From: Joseph D'Amico [ mailto:j oe.damico@apollogrp.edu]
Sent: Tuesday, August 24, 2010 8:05 PM
To: Holland, Linda
Cc: Gregory Cappelli
Subject: Visit with Bill
Hi Linda .. .I hope you are well. Our Co CEO, Greg Cappelli, who Bill has met, and I would like to set up an
appointment to meet with Bill and review with him some of the things we are doing to address the GAO report and
related matters. If we can get an hour of his time, we would really appreciate it and I know it would be of value to him,
especially with all that is going on in the sector. We will do our best to meet his dates. Thanks and look forward to
seemg you soon.
Warm regards, Joe
Sent from my iPad
This message is private and contidential. If you have received it in error, please notify the sender and remove it from your
system.
From: Wolff, Russell
To.: Woodward, Jennifer
CC:
Date: 9/13/2010 4:52:08 PM
Subject: FW: Visit with Bill
(b)(5)
-----Ori gi nat Message-----
From: Holland, Linda
Sent Monday, September 13, 2010 4:36 PM
To: Wolff, Russell
Subject: RE: Visit with Bill
I will let you know a couple of dates, and check your availability.
-----Original Message-----
From: Holland, Linda
Sent Monday, September 13, 2010 4:30 PM
To: Wolff, Russell
Subject: FW: Visit with Bill
(b)(5)
-----Original Message-----
From: Joseph D'Amico [mailto:joe.damico@apollogrp.edu]
Sent Tuesday, August 24, 2010 8:05PM
To: Holland, Linda
Cc: Gregory Cappelli
Subject: Visit with Bill
Hi Linda .. .I hope you are welt. Our Co CEO, Greg Cappelli, who Bill has met, and I would li ke to set up an
appointment to meet with Bill and review with him some of the things we are doing to address the GAO report and
related matters. If we can get an hour of his time, we would really appreciate it and I know it would be of value to him,
especially with all that is going on in the sector. We will do our best to meet his dates. Thanks and look forward to
seeing you soon.
Warm regards, Joe
Sent from my iPad
This message is private and confidential. If you have received it in error, please notify the sender and remove it from your
system.
From: Kanter Martha
To: Yum Georgia
CC: Shireman. Bob
Date: 4/12/2010 1:54:54 PM
Subject: Fwd: Kaplan, DeVry, EDMC Letter to Dep. Sec. Miller
(b){S)
Martha Kanter
Under Secretary
U.S. Department ofEducation
"The future belongs to those who believe in the beauty of their dreams!"
--Eleanor Roosevelt
Begin forwarded message:
From: "AndrewS. Rosen"
To: "Miller, Tony"
Cc: "Kanter, Martha", "Shireman, Bob" , "Fine, Stephanie" , "Rebecca Campoverde"
Subject: Kaplan, DeVry, EDMC Letter to Dep. Sec. Miller
Dear Tony:
Attached, in advance of our meeting on Thursday, is a letter on behalf ofKaplan, DeVry and Education Management
Corp. in response to your request for suggestions regarding the issue of excessive student debt. We look forward to
discussing these and other issues when we meet.
Best regards,
Andy
[ cid :image003 .j pg@O 1 C9E81E.F A05B230]
Andrew S. Rosen
Chairman and CEO
Kaplan, Inc.
6301 Kaplan University Avenue
Fort Lauderdale, Florida 33309
(954) 515-3888
www.kaplan.com
The Honorable Anthony Wilder Miller
Deputy Secretary
U.S. Department of Education
400 Maryland A venue, SW
Washington, DC 20202
Dear Secretary Miller:
April 12, 2010
Thank you for soliciting input on the Department of Education's (ED) proposed Gainful
Employment (GE) regulation at our recent meetings. We are writing on behalf of our
institutions (Kaplan, DeVry, and Education Management Corporation), which together offer
opportunities for over three hundred thousand students to attend college annually. We are
deeply committed to educating and preparing our students for the new jobs of the 21st century,
and to ensuring that our students receive high-quality, results-oriented education, without being
burdened by excessive debt.
We understand and support what you are trying to accomplish. We believe that together we
can flnd a solution that addresses student debt and simultaneously enables the Administration
to achieve its goals of expanding access to quality higher education, particularly among non-
traditional students. We believe both sets of goals are achievable.
We thought it would be most helpful to (a) describe the contribution of the private sector in
achieving the Administration's goals, (b) explain the impact of the latest GE proposal made
public, and (c) offer a constructive alternative to this GE proposal that would address the ED's
concerns without restricting students' access to college opportunities.
Quality Private Sector Colleges Play A Critical Role in Achieving Administration Goals
President Obama has said he wants America to have the highest percentage of college
graduates in the world by 2020. This goal will require educating millions of additional college
students at a cost of many billions of dollars and cannot be met without the participation of
quality private sector colleges like ours. The private sector currently educates some 2.7 million
students a year and has the resources to help alleviate the fmancial burden of achieving the
Administration's goal. Moreover, the private sector attracts more non-traditional students - a
critical requirement to increasing the number of college graduates.
The Honorable Anthony Wilder Miller
Aprill2, 2010
Page2
Not only do private sector colleges attract more non-traditional students, but we also help them
graduate and achieve gainful employment at significantly higher rates. A recent report by The
Parthenon Group, using ED data for public and private two-year and less institutions, shows
that students at private sector colleges graduate at rates roughly 50 percent higher than public
schools. The study further shows that private sector college students achieve higher percentage
wage increases (54% vs. 36%) after completing their education.
1
The Current GE Proposal Would Dramatically Limit Students, College Opportunities
Kaplan, DeVry, and EDMC share the ED's goal of ensuring that students receive a quality
education and enter programs with a full understanding of the costs, without incurring
excessive debt. We would support regulation that appropriately addresses over-borrowing
while enabling high-quality institutions to continue their good work of building capacity and
innovation in higher education.
The GE criteria proposed by the ED at the end of the most recent Negotiated Rulemaking
session attempt to define "gainful employment" by establishing an 8 percent debt-service-to-
income threshold based on median student debt for college graduates. Income would be based
either on the Bureau of Labor Statistics (BLS) 25th percentile wage data, or actual earnings of
college graduates. Loan payments would be based on a 10-year repayment plan.
This proposal as written would have a number of unintended consequences. A recent study by
Mark Kantrowitz, a respected independent authority on financial aid, concludes:
"The 8% debt-service-to-income threshold is so strict that it would preclude for-profit
colleges from offering Bachelor's degree programs. It would also eliminate many
Associate 's degree programs at for-profit colleges. Even non-frofit colleges would find
it difficult to satisfy this standard if they were subjected to it. "
Kantrowitz further found that:
"The proposed use of Bureau of Labor Statistics wage data ... will disproportionately
harm minority and female students. "
3
Kantrowitz also points out that the proposed GE rule tasks institutions with a job without
providing the tools necessary to complete the job:
1
Parthenon Perspectives on Private Sector Post-Secondary Schools, February 24, 2010, by Robert Lytle, Roger
Brinner and Chris Ross; p. 8; Source: NCES BPS 2004-2006.
2
What is Gainful Employment? What Is Affordable Debt?, Mark Kantrowitz, March 1, 201 0, p. 1.
3
Ibid.
The Honorable Anthony Wilder Miller
April 12,2010
Page 3
"The debt-service-to-income threshold effectively establishes borrowing limits based on
field of study and degree programs, but does not give colleges the controls needed to
enforce these limits. Current sub-regulatory guidance precludes colleges from
establishing lower loan limits. "
4
Another study conducted by Charles River Associates reaches similar conclusions, estimating
that 18 percent of private sector programs will be disqualified from participation in Title IV
programs and that this would impact one-third of private sector students. This means that
hundreds of thousands of entering students would be displaced annually from private sector
colleges.
5
By 2020, approximately 5.4 million students who otherwise would be on track to
attend college would be denied access by the proposed GE regulation.
6
Finally, the GE proposal would result in significant job loss among the hundreds of thousands
of faculty members, administrators, and staff who work in the private post-secondary sector,
and in non-degree programs in public sector and independent schools as well.
Students Will Be Protected by Transparent Cost and Debt Information.
We remain concerned that defining "gainful employment" by student debt levels is beyond
Congressional intent. We believe that the necessary data to both defme the problem and
support a sufficient and informed policy have not yet been compiled and analyzed. We are
certain there are numerous consequences of the GE proposal that are not currently
contemplated by the ED.
For these reasons, we propose that student debt concerns be addressed by mandating that all
institutions disclose to students the information students need to make informed decisions prior
to taking on student debt, as well as warn students about programs that fail to meet a minimum
debt-service-to-income ratio under a new student consumer "lemon law." Prospective students
who receive sufficient information at the time of enrollment are in the best position to make an
informed decision regarding whether or not to attend an institution. We believe the information
students need to make decisions concerning the appropriate amount of debt to incur for a given
program should be provided in a disclosure form to students.
The form would include: (a) the cost of the program of study, (b) a reasonable projection of
potential earnings in the students' chosen field upon graduation and throughout the life of their
employment in that field, (c) a reasonable estimate of the debt students typically incur to
complete their program, and (d) students' repayment plan options. A proposed disclosure form
4
Ibid. p. 2.
5
Report on Gainful Employment, Charles Rivers Associates, April2, 2010, prepared by Jonathan Guryan, PhD,
and Matthew Thompson, PhD, p. 38.
6
Executive Summary to Report on Gainful Employment, Charles Rivers Associates, April 2, 2010, prepared by
Jonathan Guryan, PhD, and Matthew Thompson, PhD, p. I.
The Honorable Anthony Wilder Miller
April12, 2010
Page4
is attached as Appendix 1. The accuracy of the information contained in the disclosure form
would be ensured by the misrepresentation prohibition that received tentative agreement at the
last Negotiated Rulemaking session. The proposed misrepresentation prohibition provides,
among other things, that:
If the Secretary determines an institution has engaged in substantial misrepresentation,
the Secretary may revoke or limit that institution's participation in the Title IV
programs.
Misrepresentation is defined as any false, erroneous or misleading statement an
institution makes directly or indirectly to a student, prospective student, or any member
of the public, an accrediting agency, State agency, or the Secretary.
A misleading statement includes any statement that has the capacity, likelihood, or
tendency to deceive or confuse. The omission of information may also be interpreted as
a misrepresentation.
In addition to this disclosure, schools would be required to warn students prior to enrollment of
any program that fails to meet a debt-service-to-income ratio test. The debt-service-to-income
ratio would be based on the approach recently proposed by the ED, with appropriate
modifications discussed below. Institutions offering programs that fail the test would be
required to warn students in appropriate marketing materials, and in a written disclosure signed
by the student prior to enrollment, that (a) the program has failed a debt-service-to-income-
ratio test, and (b) student borrowers enrolling in the program should expect to have difficulty
meeting their repayment obligations upon graduation.
To ensure that the debt-service-to-income ratio is appropriately directed at identifying "outlier"
programs we propose that the ratio currently contained in the GE proposal be adjusted as
follows:
Formula applied to non-degree programs only.
)> Degree programs confer lifetime benefits that don't correlate easily to
specific job codes, such as higher lifetime earnings, higher income growth
rates, greater employability, better career advancement and job stability.
7
In
addition, degree holders tend to change jobs and pursue careers seemingly
unrelated to the degrees, but using the skills they developed in college.
Including degrees in the ratio definition would dramatically undervalue these
programs.
)> By applying the formula only to non-degree programs, both private and
public institutions are impacted in the same manner.
A debt-service-to-income threshold of 15 percent, based on median student debt for
college graduates, and assuming a current unsubsidized Stafford loan interest rate of
6.8% to calculate the annual repayment amount.
'Kantrowitz, pp. 20-21.
The Honorable Anthony Wilder Miller
April 12,2010
Page 5
};;> The 15 percent debt-service-to-income threshold is referenced in the
Kantrowitz study as a well as a recent study published by the College
Board,
8
and is within the range generally used by personal financial
counseling professionals.
Income based either on the BLS 50th percentile wage data, or actual earnings of
graduates if the latter are higher than the BLS 50th percentile.
};;> The 50th percentile of the BLS wage data more accurately reflects the long-
term potential earnings of a graduate. Moreover, there is no reason to
assume that non-degree program graduates, regardless of their backgrounds,
would be unable to achieve average earnings.
Loan payments based on a 20-year repayment plan.
};;> The 20-year loan repayment plan is also referenced in the Kantrowitz study
and supported by the fact that borrowers are permitted to, and do, choose
repayment plans covering a period of up to 25 years.
Exclude prior school debt from the calculation and provide institutions the
regulatory ability to control student borrowing, thereby enabling compliance with
ratio and 90/ 1 0 requirements.
};;> Absent the regulatory ability to control student borrowing, the GE
calculation should be based only on direct cost of education.
Eliminate the ED pre-approval requirement for new programs.
};;> State regulatory bodies and accrediting agencies already require approval of
all new programs.
We also recommend that the ED consider alternative routes to compliance with the debt-
service-to-income ratio test, specifically by establishing: (1) target graduate cohort default rates
(GCDRs) (e.g., 12.5% GCDR on a two-year calculation; 15% on a three-year calculation), (2)
targets for actual post-graduation salaries that include a multiplier of 1.5x to recognize the fact
that lifetime earnings are significantly higher than BLS rates, and (3) thresholds for post-
graduate employment rates.
We believe that the proposal contained in this letter provides an innovative and effective way to
protect students from institutions that over promise and under deliver to students, thus leaving
students with too much debt and not enough return on investment.
8
How Much Debt Is Too Much, Sandy Baum and Saul Schwartz, The College Board, 2006, p. 12.
The Honorable Anthony Wilder Miller
April 12,2010
Page6
We appreciate the opportunity to provide this input and we look forward to sitting down with
you soon to discuss these matters further.
Yours Truly,
Andrew S. Rosen
Chairman and CEO, Kaplan, Inc.
Daniel Hamburger
President and CEO, DeVry Inc.
Todd S. Nelson
CEO, Education Management Corporation
Enclosures
cc: The Honorable Martha J. Kanter
Mr. Robert Shireman
APPENDIX 1
INSTITUTIONAL DISCLOSURES RELATED TO EXPECTED EARNINGS AND DEBT
You have requested information about our _....;.A=c=co=u=n"'-'t=in"""'g.__ ___ program
Program Level: D Associates [g)Bachelors 0 Masters Ocertificate/Diploma
Here are some important disclosures for the award year ending June 30, 2010
During the year ended June 30, 2009 , 75.8 %of students enrolled in this program graduated or
continue to be actively enrolled at the institution while 24.2 % ceased enrollment.
Of the students who graduated, 88.6 % were employed in their field of study, or a related field,
within six months of graduation with an average annual salary of approximately$ 46,300 per year.
This academic program corresponds to the following Standard Occupational Classification (SOC)
codes as reported by the Bureau of Labor Statistics (BLS): 13-2011 . The weighted annual
salaries for these SOC codes at the 25th and 75th percentiles are $ 45,900 and $ 78,210 ,
respectively. For information related to salaries from these and other occupations, please visit
http://www.bls.gov/oes/current/oes_nat.htm.
The cost of this program of study for a student enrolled full-time and with no transfer credits is
$ 62,040 . The average annual tuition increase for the most recently concluded three years was
4.6 %
The average education loan debt of students incurred at this institution and who graduated from this
program during the prior award year was $ 33.100 . This amount includes $ 30,900 of federal
student loan debt and $ 2,200 of institutional loan debt. This does not include any debt incurred
while attending another institution. Additionally, 4.6 %of graduates obtained private student loans
from third parties.
If this average education loan debt was 100% federal loans with an average interest rate of 6.8% and
you chose to repay using a 10 year standard repayment term, the annual total of 12 monthly
payments would be $ 4,571.04 . If you chose to pay using a graduated repayment plan (over 10
years), the total of your first 12 monthly payments would be$ 3,138.60 . For more information
concerning repayment options on federal loans, please visit
https:/ /studentloans.gov/myDirectLoan/index.action.
The latest official Cohort Default Rate (FY07) from the US Department of Education indicates that
1.7 % of graduates in thi s program defaulted on their federal loans.
PLEASE NOTE THAT YOUR ACTUAL EXPERIENCE MAY BE DIFFERENT THAN THE AVERAGES AND
STATISTICS PRESENTED ABOVE.
From: Kanter Martha
To: Weiss, Joanne
CC:
Martin, Carmel
Yuan Georgia
Rose Charlie
Gomez_ Gabriella
Ochoa Eduardo
Miller Tony

Date: 9/2/2010 10:24:24.PM
Subject: FYI: A Message from Dr. John Sperling, Founder- University of Phoenix
September 2, 2010
Attn. Legislative Director
Dear Congress Member,
As founder of the University of Phoenix, I am writing to you and to every Member of Congress concerning recent
actions by the U.S. Department ofEducation and the Senate Committee on Health, Education, Labor and Pensions. The
Department ofEducation, seconded by the HELP Committee, has proposed new rules that will seriously undercut the
ability of the nation to remain globally competitive by undermining a vital sector of our higher education system-for-
profit colleges and universities.
I am writing to you because I am confident that every Member of Congress, whether or not they are a member of a
committee that deals with higher education, would like quality higher education to be available to every American who
seeks to earn a college degree. That is the stated goal of President Obama. However, we will not be able to reach the
President's goals without for-profit colleges. The states do not have the funding needed to increase capacity whereas
private sector colleges like ours can provide the necessary capacity. For-profit colleges already provide flexibility and
access for millions ofunderserved and nontraditional students who could not complete their education in a traditional
institution.
The attached power point has been prepared by NEXUS, a research and policy institute whose primary focus is the for-
profit sector of higher education. Given its commitment to fact based research, not special pleading, the power point
presents a rationale for the need to rethink the reforms proposed by USDOE and the HELP Committee using as an
example the case of the University ofPhoenix, whose massive database on its operations and its academics has been
made available to NEXUS researchers.
As the largest institution in the sector-465,000 students with 90,000 graduates last year-the University ofPhoenix
illustrates the strengths and weaknesses of for-profit colleges and universities. The NEXUS study documents the financial
system that sustains it, the quality of the University' s programs, the innovations it has brought to higher education and
where it has failed to meet regulatory standards and its own code of conduct It also documents the steps the University
has taken to insure future compliance with all regulatory standards.
Perhaps the most important finding of the case study is the fact that for-profit institutions operate at no cost to taxpayers
because the interest students pay on their federal loans plus the taxes paid by the institutions is greater than the Pell
Grants and all of the other state and federal subsidies received by the students and the institutions. Further, the study
shows that not only will the proposed reforms require a mYor increase in Department of Education oversight staff, they
will greatly lower the efficiency and raise the costs of the institutions in the sector-all at the expense of taxpayers.
It would seem wiser to restore the status quo ante when the Department ofEducation was pursuing a reform that would
truly benefit taxpayers, namely to require all institutions of higher education to measure the learning outcomes of their
students and to publish those results on an annual basis. Overseeing the measurement protocols used by institutions in
order to expose cheating on the tests, would be a far more productive use of Department of Education resources than
what is presently contemplated. Only when learning outcomes are measured and published will taxpayers know what
they are getting for their money and the "bad actors" be easily identified-namely the institutions whose students are low
performers.
When you have finished reviewing the power point, we hope you will be persuaded that for-profit colleges and
universities are a vi tal sector of the American higher education system that deserve the support of the Department of
Education and the Congress, along with responsible oversight, rather than a set of regulations that will instead inhibit their
efficiency, their growth, their culture of innovation and, most importantly, their ability to deliver a quality education to
millions of! ow-income Americans now denied access to the education they need to give them a chance to join the
middle class.
The worst of these regulations concerns the definition of"gainful employmenf' and a new regulation that sets the
minimum rates at which students in a program must be repaying on the principal of their student loans. The new definition
of gainful employment is so onerous it would make it impossible for the sector to offer many programs that prepare
students for certification in such occupations as teachers, nurses, counselors and public safety officers. The repayment
percentage requirements, apparently arrived at with insufficient attention to their potential negative consequences, would
have a devastating impact on institutions that enroll low-income students who often require several years in the
workforce before they can begin repaying the principal on their student loans. For example, if these requirements were
applied to Historically Black Colleges and Universities, over 90% of them would have to close their doors.
Representative Robert Andrews (New Jersey - 01) has proposed a definition of gainful employment that would correct
the many faults of the definition proposed by the Department ofEducation. He has written to Education Secretary Arne
Duncan setting forth the negative consequences resulting from the Department's definition while proposing an alternative
which would remove these negative consequences and still gain the same objectives. Many of the institutions in the for-
profit sector support Congressman Andrews' s initiative and we are asking all Members of the Congress to lend their
supp01t as well. I have attached a draft of a letter of support which we hope you wilt use as a model for a letter from you
to Secretary Duncan with copies to Speaker Nancy Pelosi and to Congressman Andrews. The University and every
institution in the sector would very much appreciate your support.
Thank you for your time and consideration of this important issue.
[ cid:imageOO l.gif]
John G. Sperling
Founder
University ofPhoenix
From: Kanter, Martha
Sent: Saturday, April 03, 2010 6:03PM
To: Private- Duncan, Arne
Subject: FW: updated memo
(b)(5)
Martha
Georgia Yuan
Deputy General Counsel
Postsecondary and Regulatory SeiVice
LBJ 6E341
202- 401-6399
From: Kanter Martha
To.: Yuan. Georgia
CC:
Date: 4/3/2010 6:04:36 PM
Subject: FYI: updated memo
Gainful Employment
(b)(S)
1
(b)(5)
1
(b)(5)
1
(b)(5)
~ ~ ~ - ~ . ~ ~ ~ . - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - : 1 - - - - - - ~
(b)(5)
1
(b)(5)
1
(b){S)
1
(b)(5)
1
(b)(5)
1
(b){S)
1
(b)(5)
1
(b)(5)
thanks
From: K vaal James
To: .N.filler, Tony
Weiss, Joanne
Rose. Charlie
Cunningham Peter
Gomez Gabriella
Yuan Georgia
CC: Kanter, Martha
Ochoa, Eduardo
Date: 9/3/2010 6:37:10 PM
Subject: gainful employment
(b)(S)
Richard
-----Original Message-----
From: Scott, Byron
Sent Wednesday, April14, 2010 10:37 AM
To: Nelson, Richard; Hale, Milton
Subject: RE:
From: Nelson, Richard
To.: Finley, Steve
CC: Scott, Byron
Hale, Milton
Date: 4/14/2010 1:40:58 PM
Sub.iect: Gainful Employment Proposal
My goodness, Credit Suisse seems to have a very good source. I wonder who that might be?
(b)(S)
-----Original Message-----
From: Nelson, Richard
Sent: Wednesday, Apri l 14, 2010 12:02 PM
To: Hale, Milton; Scott, Byron
Subject: FW:
Do either of you know anything about the Department's "Gainful Employment" proposal? See the attached PDF file.
-----Original Message-----
From: James, Bob
Sent: Tuesday, April 13, 2010 3:25PM
To: Nelson, Richard
Subject:
Robert H. James
Liaison for Career Institutions of Higher Education US Department ofEducation
FAX: 317-257-2098 Call first
Cell Phone #202-557-5835 D.C.# 202-377-4301 Indianapolis Office 317-257-2098
8527 Quail Hollow Road
Indianapol is. IN 46260-2208
CREDIT SUISSE
Research Analysts
Kelly Flynn, CFA
617 556 5752
13 April 2010
Americas/United States
Equity Research
Education Services (Business & Professional Services) I MARKET WEIGHT
Education Services
=
UPGRADE RATING
keny.nynn@credit-suisse.com Upgrade DeVry and ITT On New Gainful
Patrick Elgrably, CFA E I I . h
312 750 2974 mp oyment ns1g ts
patrick.elgrably@ credit-suisse .com
Adam Shatek, CPA We are upgrading ITT (ticker ESI) and DeVry (ticker DV) to Outperform from
312 750 3317 Neutral due to new insights on the DOE's Gainful Employment stance. We
adam.shatek@credit-suisse.com are raising our ITI price target to $135 from $105 and our DeVry price target to
$75 from $55 as DCF discount rate reductions reflect perceived decreased
Gainful Employment risks. We detail our DCF analysis assumptions for DV and
ESI below. We are restricted on EDMC. All other Neutral ratings are
unchanged (see next page for details}.
We believe DOE's latest GE proposal leaves 8%110 year parameters
unchanged. Following discussions with industry contacts last night, we believe
that the Department of Education on Friday submitted its Gainful Employment
and other program integrity proposed language to the Office Of Management
and Budget (the OMB) for vetting. We believe the goal is to publish the Notice of
Proposed Rule Making by May 15, or June 1 at the latest, and to publish final
regulations by November 1. We also believe the 8% debt-service-to-income
ratio and 10 percent repayment period inputs remain unchanged as of now and
that the "90% of graduates in repayment" exemption remains unchanged.
But, we believe DOE added a 50% completion exemption. Based on our
discussions, we believe one big change in the new draft proposal is the add back
of the "exemption" (that was removed in January after appearing in the initial
draft language) for schools with certain student completion and job placement
rates. We believe the completion rate cut off is now a more generous 50%
(versus 70% included in initial draft) and the placement rate cut-off is 70% (same
as in initial draft language; we believe most companies with placement rates
have rates above 70%).
We think 50% completion rate exemption would help ESI, DV, & EDMC
regulatory positioning the most. We believe the 50% exemption, although not
eliminating Gainful Employment risks, would most significantly improve the
positioning for companies with placement rates at or close to 50% that would,
without an exemption, have potentially seen earnings prospects decimated by a
new Gainful Employment regulation; DeVry, ITT and Education Management fit
this profile. Although none of these companies release completion rates, the
DOE data (which likely understates actual completion rates because it only
includes first-time, full time students and not transfers or part time students) is
close enough to 50% to make us think these companies likely have 50%
completion rates or could achieve them in coming years without decimating
earnings prospects; most recent DOE completion rate data points are -39% for
ESI (ESI also said on recent call that 60% make it through first year), - 31% for
DV and -41 % for EDMC's Art Institute (-56% of company's students). Further,
we believe ESI's valuation, and DeVry's to a lesser extent, have been among
those most negatively impacted by Gainful Employment concerns.
DOE flexibility may also fuel more investor optimism. We also acknowledge
that the 50% completion rate change, if in fact it occurs, could fuel investor
optimism that the DOE could ease its stance further in coming months in
response to more pressure that may arise after the NPRM is posted.
DISCLOSURE APPENDIX CONTAINS IMPORTANT DISCLOSURES, ANALYST CERTIFICATIONS, INFORMATION ON
TRADE ALERTS, ANALYST MODEL PORTFOLIOS AND THE STATUS OF NON-U.S ANALYSTS. U.S. Disclosure:
Credit Suisse does and seeks to do business with companies covered in its research reports. As a result, investors should be
aware that the Firm may have a conflict of interest that could affect the objectivity ot this report. Investors should consider this
report as only a single factor in making their investment decision.
CREDIT SUISSE
Sector Implications/Thesis Update
Countercyclicality and other sector regulatory concerns remain; other ratings
unchanged. We continue to worry that countercyclicality will hurt growth in coming
quarters and we believe that. even if the Gainful Employment proposal is eased somewhat.
the broader regulatory landscape is likely to remain challenging for the foreseeable future;
we expect the DOE to tighten the Incentive Compensation rule and to generally seek to
crack down on schools' recruiting underqualified, under-informed students. Further, our
thesis on other companies under coverage is not changed significantly by our changing
view on Gainful Employment. Our concerns about APEI , APOL, COCO, LINC and UTI are
largely unrelated to Gainful Employment. CPLA, STRA and LOPE trade at premium
valuations already, and we are not confident they would make the cut on the 50%
completion or graduation loan repayment Gainful Employment exemptions. For BPI and
CECO, although shares look cheap, we are also not confident Gainful Employment
exemptions apply, and we believe these companies also face other significant regulatory
risks. We are restricted on EDMC.
Valuation
Our price target changes largely reflect decreases in our discount rate used in the DCF
analyses due to lower perceived Gainful Employment risks, in our view.
Our $135 ESI price target is derived from our DCF analysis. We summarize our DCF
assumptions below:
2010-2020 revenue Compound Annual Growth Rate (CAGR) of 4.9%.
2009 operating margin of 37.9% going to 39.2% by 2020.
A Weighted Average Cost of Capital (WACC) of 16%.
Terminal free cash flow growth of 3%.
Working capital changes and capital expenditures that remain in-line with historical
ratios.
Our $75 DV price target is derived from our DCF analysis. We summarize our DCF
assumptions below:
2010-2020 revenue Compound Annual Growth Rate (CAGR) of 10.7%.
2009 operating margin of 19.7% going to 20.2% by 2020.
A Weighted Average Cost of Capital (WACC) of 14%.
Terminal free cash flow growth of 3%.
Working capital changes and capital expenditures that remain in-line with historical
ratios.
Price Price Rating Target Price
Year'
EPS EPS FY1E
Company ccy 09 Apr 10 Prev. Cur. Prev. Cur. End Ccy Prev. Cur.
DeVry Inc. (DV) US$ 65.06 N 0 55.00 75.00 Jun 09 uss - 3.41
ITI Educational Services (ESI US$ 108.78 N 0 JVI 105.00 135.00 Dec09 US$ - 10.50
13 April 2010
EPS FY2E EPS FY3E
Prev. Cur. Prev. Cur.
- 4.05 - 4.65
- 11.69 - 12.63
o- Outperform, N- Neutral, u- Underperform, R- Restricted
Source: Company data, Credit Suisse estimates.
[VJ = StOCk considered volatile (see DISclosure Appendix).
Education Services 2
CREDIT SUISSE 13 April 2010
Companies Mentioned (Price as of 09 Apr 10)
American Public Education, Inc. (APE!, $46.03, NEUTRAL, TP $41.00)
Apollo Group Inc. (APOL, $63.14, NEUTRAL [V], TP $65.00)
Bridgepoint Education (BPI, $23.60, NEUTRAL [V], TP $18.00)
Capella Education Company (CPLA, $90.90, NEUTRAL, TP $72.00)
Career Education Corp. (CECO, $31.70, NEUTRAL [V], TP $28.00)
Corinthian Colleges, Inc. (COCO, $17.51 , NEUTRAL [V], TP $14.00)
DeVry Inc. (DV, $65.06, OUTPERFORM, TP $75.00)
Education Management Corporation (EDMC, $22.60, RESTRICTED [V])
Grand Canyon Education (LOPE, $25.68, NEUTRAL, TP $21.00)
ITT Educational Services, Inc. (ESI, $108.78, OUTPERFORM [V], TP $135.00)
Lincoln Educational Services (LINC, $25.70, NEUTRAL [V], TP $21.00)
Strayer Education, Inc. (STRA, $238.70, NEUTRAL, TP $195.00)
Universal Technical Institute (UTI, $23.05, NEUTRAL [V], TP $19.00)
Disclosure Appendix
Important Global Disclosures
I, Kelly Flynn, CFA, certify that (1) the views expressed in this report accurately reflect my personal views about all of the subject companies and
securities and (2) no part of my compensation was, is or will be directly or indirectly related to the specific recommendations or views expressed in
this report.
See the Companies Mentioned section for full company names.
3-Year Price, Target Price and Rating Change History Chart for DV
DV Closing Target
Price Price Initiation/
Date {USS) {USS) Ratins Assumetion
4/27/07 34.52 29
6/18/07 35.22 NC
2/21/08 43.86 65 0 X
6/20/08 58.54 N
7/30/08 56.75 R
10/8/08 45.19 52 N
12/5/08 58.37 53
4/21/09 42.12 44
8/14/09 51.89 50
10/28/09 56.13 52
1/27/10 63.32 55
67
62
51
52
47
42
37
32

- aosngPae& TaJQet Pri:e Rating
O:Oulperlo<m; f'I=Ne.rtral; U:Unde<pertotm; R:Restrcled; NR:Nol Rated; NC:NOI Co.eted
3-Year Price, Target Price and Rating Change History Chart for ESI
ESI Closing Target
165
----------------------------11!i-llf-------------------
Price
Date (USS)
4/27/07 97.9
6/18/07 113.52
2/21/08 60.17
2/25/08 54.02
6/20/08 88.4
10/24/08 74.1
2/2109 129.43
4/21/09 101.31
Price Initiation/
(USS) Ratins Assumeti on
110
65
61
81
84
165
105
NC
N
0
N
X
145 -------------------------------------------------
- OosilgPae& Rating
O:Outperlo<m: f'I=Ne.rtral; U=Underpertotm: R=Restri:led: NR=Nol Ra1ed: NC:Nol Co.ered
The analyst(s) responsible for preparing this research report received compensation that is based upon various factors including Credit Suisse's total
revenues, a portion of which are generated by Suisse's investment banking activities.
Analysts' stock ratings are defined as follows:
Outperform (0): The stock's total return is expected to outperform the relevant benchmark* by at least 10-15% (or more, depending on perceived
risk} over the next 12 months.
Neutral (N): The stock's total return is expected to be in line with the relevant benchmark* (range of 1 0-15%} over the next 12 months.
Education Services 3
CREDIT SUISSE 13 April 2010
Underperform (U): The stock's total return is expected to underperform the relevant benchmark* by 10-15% or more over the next 12 months.
*Relevant benchmark by region: As of 2gh May 2009, Australia, New Zealand, U.S. and Canadian ratings are based on (1) a stock's absolute total
return potential to its current share price and (2) the relative attractiveness of a stock's total return potential within an analyst's coverage universe .. ,
with Outperforms representing the most attractive, Neutrals the less attractive, and Underperforms the least attractive investment opportunities.
Some U.S. and Canadian ratings may tall outside the absolute total return ranges defined above, depending on market conditions and industry
factors. For Latin American, Japanese, and non-Japan Asia stocks, ratings are based on a stock's total return relative to the average total return of
the relevant country or regional benchmark; for European stocks, ratings are based on a stock's total return relative to the analyst's coverage
universe ... For Australian and New Zealand stocks a 22% and a 12% threshold replace the 10-15% level in the Outperform and Underperform stock
rating definitions, respectively, subject to analysts' perceived risk. The 22% and 12% thresholds replace the + 1 015% and -1 015% levels in the
Neutral stock rating definition, respectively, subject to analysts' perceived risk.
**An analyst's coverage universe consists of all companies covered by the analyst within the relevant sector.
Restricted (R): In certain circumstances, Credit Suisse policy and/or applicable law and regulations preclude certain types of communications,
including an investment recommendation, during the course of Credit Suisse's engagement in an investment banking transaction and in certain other
circumstances.
Volatility Indicator [V]: A stock is defined as volatile if the stock price has moved up or down by 20% or more in a month in at least 8 of the past 24
months or the analyst expects significant volatility going forward.
Analysts' coverage universe weightings are distinct from analysts' stock ratings and are based on the expected
performance of an analyst's coverage universe* versus the relevant broad market benchmark**:
Overweight: Industry expected to outperform the relevant broad market benchmark over the next 12 months.
Market Weight: Industry expected to perform in-line with the relevant broad market benchmark over the next 12 months.
Underweight: Industry expected to underperform the relevant broad market benchmark over the next 12 months.
*An analyst's coverage universe consists of all companies covered by the analyst within the relevant sector .
.. The broad market benchmark is based on the expected return of the local market index (e.g., the S&P 500 in the U.S.) over the next 12 months.
Credit Suisse's distribution of stock ratings (and banking clients) is:
Global Ratings Distribution
Outperform/Buy 44% (60% banking clients)
Neutral/Hold* 41% (61 % banking clients)
Underperform/Sell* 13% (56% banking clients)
Restricted 2%
For purposes of the NYSE and NASD ratings distribution disclosure requirements, our stock ratings of Outperform, Neutral, and Underpertorm most closely correspond to Buy,
Hold, and Sell, respectively; however, the meanings are not the same, as our stock ratings are determined on a relative basis. (Please refer to definitions above.) An investor's
decision to buy or sell a security should be based on investment objectives, current holdings, and other individual factors.
Credit Suisse's policy is to update research reports as it deems appropriate, based on developments with the subject company, the sector or the
market that may have a material impact on the research views or opinions stated herein.
Credtl Suisse's policy is only to publish investment research that is impartial , independent, clear, fair and not misleading. For more detail please refer to Credtt
Suisse's P o ~ c i e s for Managing Conflicts of Interest in connection with Investment Research:
http1/www.csfb.comtresearch-and-analyticsfdisclaimer/managing_conflicts_disclaimer.html
Credit Suisse does not provide any tax advice. Any statement herein regarding any US federal tax is not intended or written to be used, and cannot
be used, by any taxpayer for the purposes of avoiding any penalties.
See the Companies Mentioned section for full company names.
Price Target: (12 months) for (DV)
Method: Our $75 target price for DV is derived from our Discounted Cash Flow (DCF)based model. Our target price is based on the following
assumptions: 2010-20 revenue CAGR (compound annual growth rate} of 10.7%, 2010 operating margin of 19.7% going to 20.3% by 2020, a WACC
(weighted average cost of capital) of 14% and terminal free cash flow growth of 3%.
Risks: The following factors may affect our projected results and $75target price for DeVry: its heavy reliance on IT-related education which could
hurt results if IT spending declines, an economic recovery which could curt countercyclical post-secondary education companies, changes in the
regulatory and accreditory environments, and the impact of a downturn in the student lending environment which could threaten the magnitude of
federal loans to DeVry's students.
Price Target: (12 months) for (ESI}
Method: Our $135 target price for ESI is derived from our Discounted Cash Flow (DCF)-based model. Our base case DCF has the following
assumptions: 200919 revenue CAGR of 4.9%, operating margins going from 37.9% to 39.2% from 2010-20, a WACC of 16%, and terminal free
cash flow growth of 3%.
Risks: The following factors may affect our projected results for ESI and our $135 price target: its heavy reliance on IT-related education which could
hurt results when IT spending declines, an economic recovery which has the potential to negatively impact countercyclical post-secondary education
services companies, impact of greater regulatory and accrediting agency requirements and the recent downturn in the student lending environment,
which impacts the magnitude of federal loans provided to ITI students.
Please refer to the firm's disclosure website at www.credit-suisse.com/researchdisclosures for the definitions of abbreviations typically used in the
target price method and risk sections.
Education Services 4
CREDIT SUISSE 13 April 2010
See the Companies Mentioned section for full company names.
The subject company (DV, ESI) currently is, or was during the 12-month period preceding the date of distribution of this report, a client of Credit
Suisse.
Credit Suisse provided investment banking services to the subject company (DV, ESI) within the past 12 months.
Credit Suisse has received investment banking related compensation from the subject company (ESI) within the past 12 months.
Credit Suisse expects to receive or intends to seek investment banking related compensation from the subject company (DV, ESI} within the next 3
months.
Important Regional Disclosures
Singapore recipients should contact a Singapore financial adviser for any matters arising from this research report.
An analyst involved in the preparation of this report has visited certain material operations of the subject company (ESI} within the past 12 months.
The analyst may not have visited all material operations of the subject company. The travel expenses of the analyst in connection with such visits
were not paid or reimbursed by the subject company, other than de minimus local travel expenses.
The analyst(s) involved in the preparation of this report have not visited the material operations of the subject company (DV) within the past 12
months.
Restrictions on certain Canadian securities are indicated by the following abbreviations: NVS--Non-Voting shares; RVS--Restricted Voting Shares;
SVS--Subordinate Voting Shares.
Individuals receiving this report from a Canadian investment dealer that is not affiliated with Credit Suisse should be advised that this report may not
contain regulatory disclosures the non-affiliated Canadian investment dealer would be required to make if this were its own report.
For Credit Suisse Securities (Canada}, Inc.'s policies and procedures regarding the dissemination of research, please
http://www.csfb.com/legal_terms/canada_research_policy.shtml.
As of the date of this report, Credit Suisse acts as a market maker or liquidity provider in the equities securities that are the subject of this report.
Principal is not guaranteed in the case of equities because equity prices are variable.
Commission is the commission rate or the amount agreed with a customer when setting up an account or at anytime after that.
CS may have issued a Trade Alert regarding this security. Trade Alerts are short term trading opportunities identified by an analyst on the basis of
market events and catalysts, while stock ratings reflect an analyst's investment recommendations based on expected total return over a 12-month
period relative to the relevant coverage universe. Because Trade Alerts and stock ratings reflect different assumptions and analytical methods, Trade
Alerts may differ directionally from the analyst's stock rating.
The author(s) of this report maintains a CS Model Portfolio that he/she regularly adjusts. The security or securities discussed in this report may be a
component of the CS Model Portfolio and subject to such adjustments (which, given the composition of the CS Model Portfolio as a whole, may differ
from the recommendation in this report, as well as opportunities or strategies identified in Trading Alerts concerning the same The CS
Model Portfolio and important disclosures about it are available at www.credit-suisse.com/ti.
To the extent this is a report authored in whole or in part by a non-U.S. analyst and is made available in the U.S., the following are important
disclosures regarding any non-U.S. analyst contributors:
The non-U.S. research analysts listed below (if any) are not registered/qualified as research analysts with FINRA. The non-U.S. research analysts
listed below may not be associated persons of CSSU and therefore may not be subject to the NASD Rule 2711 and NYSE Rule 472 restrictions on
communications with a subject company, public appearances and trading securities held by a research analyst account.
For Credit Suisse disclosure information on other companies mentioned in this report, please visit the website at www.credit-
suisse.com/researchdisclosures or call+ 1 (877) 291-2683.
Disclaimers continue on next page.
Education Services 5
CREDIT SUISSE
13 April201 0
Americas/United States
Equity Research
This report is not directed to, or intended for distribution to or use by, any person or entity who is a citizen or resident of or located in any locality, state, country or other jurisdiction
where such distribution, publication, or use would be contrary to law or regulation or which would subject Credit Suisse AG, the Swiss bank, or its subsidiaries or affiliates
("CSi to any registration or licensing requirement within such jurisdiction. All material presented in this report, unless specifically indicated otherwise, is under copyright to CS. None of
the material, nor content, nor any copy of it, may be ahered in any way, transmitted to, copied or distributed to any other party, without the prior express written permission of CS. All
trademarks, service marks and logos used in this report are trademarks or service marks or registered trademarks or service marks of CS or affiliates.
The information, tools and material presented in this report are provided to you for information purposes only and are not to be used or considered as an offer or the solicitation of an
offer to sell or to buy or subscribe for securities or other financial instruments. CS may not have taken any steps to ensure that the securities referred to in this report are suitable for
any particular investor. CS will not treat recipients as its customers by virtue of their receiving the report The investments or services contained or referred to in this report may not be
suitable for you and is recommended that you consuh an independent investment advisor if you are in doubt about such investments or investment services. Nothing in this report
investment, legal, accounting or tax advice or a representation that any investment or strategy is or appropriate to your indMdual circumstances or otherwise
constitutes a personal recommendation to you. CS dces not offer advice on the tax consequences of investment and you are advised to contact an Independent tax adviser. Please
note in particular that the bases and levels of taxation may change.
CS believes the information and opinions in the Disclosure Appendix of this report are accurate and complete. Information and opinions presented in the other sections of the report
were obtained or derived from sources CS believes are reliable, but CS makes no representations as to their accuracy or completeness. Additional information is available upon
request. CS accepts no liability for loss arising from the use of the material presented in this report, except that this exdusion of liability does not apply to the extent that liability arises
under specific statutes or regulations applicable to CS. This report is not to be relied upon in substitution for the exercise of independent judgment. CS may have issued, and may in
the future issue, a trading call regarding this security. Trading calls are short term trading opportunities based on market events and catalysts, while stock ratings reflect investment
recommendations based on expected total return over a 12-month period as defined in the disclosure section. Because trading calls and stock ratings reflect different assumptions and
analytical methods, trading calls may differ directionally from the stock rating. In addition, CS may have issued, and may in the future issue, other reports that are inconsistent with, and
reach different conclusions from, the information presented in this report. Those reports reflect the different assumptions, views and analytical methods of the analysts who prepared
them and CS is under no obligation to ensure that such other reports are brought to the attention of any recipient of this report CS is involved in many businesses that relate to
companies mentioned in this report. These businesses indude specialized trading, risk arbitrage, market making, and other proprietary trading.
Past performance should not be taken as an indication or guarantee of future performance, and no representation or warranty, express or implied, is made regarding future
performance. Information, opinions and estimates contained in this report reflect a judgement at its original date of publication by CS and are subject to change without notice. The
price, value of and income from any of the securities or financial instruments mentioned in this report can fall as well as rise. The value of securities and financial instruments is subject
to exchange rate fluctuation that may have a positive or adverse effect on the price or income of such securities or financial instruments. Investors in securities such as ADA's, the
values of which are influenced by currency volatility, effectively assume this risk.
Structured securities are complex instruments, typically involve a hgh degree of risk and are intended for sale only to sophisticated investors who are capable of understanding and
assuming the risks involved. The market value of any structured security may be affected by changes in economic, financial and political factors (induding, but not limited to, spot and
forward interest and exchange rates), time to maturity, market conditions and volatility, and the credit quality of any issuer or reference issuer. Any investor interested in purchasing a
structured product should conduct their own investigation and analysis of the product and consuh with their own professional advisers as to the risks involved in making such a purchase.
Some investments discussed in this report have a high level of volatility. High volatility investments may experience sudden and large falls in their value causing losses when that
investment is realised. Those losses may equal your original investment. Indeed, in the case of some investments the potential losses may exceed the amount of initial investment, in
such circumstances you may be required to pay more money to support those losses. Income yields from investments may fluctuate and, in consequence, initial capital paid to make
the investment may be used as part of that income yield. Some investments may not be readily realisable and it may be difficult to sell or realise those investments, similarly it may
prove difficult for you to obtain reliable information about the value, or risks, to which such an investment is exposed.
This report may provide the addresses of, or contain hyperlinks to, websites. Except to the extent to which the report refers to website material of CS, CS has not reviewed the linked
site and takes no responsibility for the content contained therein. Such address or hyperlink Qnduding addresses or hyperlinks to CS's own website material) is provided solely for your
convenience and information and the content of the linked site does not in any way form part of this document. Accessing such website or following such link through this report or
CS's website shall be at your own risk.
This report is issued and distributed in Europe {except Switzerland) by Credit Suisse Securities (Europe) Limited, One Cabot Square, London E14 4QJ, England, which is regulated in
the United Kingdom by The Financial Services Authority C'FSA"). This report is being distributed in Germany by Credit Suisse Securities (Europe) Limited Niederlassung Frankfurt am
Main regulated by the Bundesanstah fuer Finanzdienstteistungsaufsicht ("BaFin; . This report is being distributed in the United States by Credit Suisse Securities (USA) LLC ; in
Switzerland by Credit Suisse AG; in Canada by Credit Suisse Securities (Canada), Inc .. ; in Brazil by Banco de lnvestimentos Credit Suisse S.A.; in Japan by Credit Suisse
Securities (Japan) Limited, Financial Instrument Firm, Director-General of Kanto Local Finance Bureau (Kinsho) No. 66, a member of Japan Securities Dealers Association, The
Financial Futures Association of Japan, Japan Securities Investment Advisers Association; elsewhere in Asia/Pacific by whichever of the following is the appropriately authorised entity
in the relevant jurisdiction: Credit Suisse (Hong Kong) Limited, Suisse Equities (Australia) Limited , Credit Suisse (Thailand) Limited, Credit Suisse Securities
(Malaysia) Sdn Bhd, Credit Suisse AG, Singapore Branch, Credit Suisse (India) Private Limited, Credit Suisse (Europe) Seoul Branch, Credit Suisse AG,
Taipei Securities Branch, PT Credit Suisse Securities Indonesia, and elsewhere in the world by the relevant authorised affiliate of the above. Research on Taiwanese securities
produced by Credit Suisse AG, Taipei Securities Branch has been prepared by a registered Senior Business Person. Research pro'vided to residents of Malaysia is authorised by the
Head of Research for Suisse Securities (Malaysia) Sdn. Bhd., to whom they should direct any queries on +603 2723 2020.
In jurisdictions where CS is not already registered or licensed to trade in securities, transactions will only be effected in accordance with applicable securities legislation, which will vary
from jurisdiction to jurisdiction and may require that the trade be made in accordance with applicable exemptions from registration or licensing requirements. Non-U.S. customers
wishing to effect a transaction should contact a CS entity in their local jurisdiction unless governing law permits otherwise. U.S. customers wishing to effect a transaction should do so
only by contacting a representative at Credit Suisse Securities {USA) LLC in the U.S.
Please note that this report was originally prepared and issued by CS for distribution to their market professional and instMonal investor customers. Recipients who are not market
professional or institutional investor customers of CS should seek the advice of their independent financial advisor prior to taking any investment decision based on this report or for
any necessary explanation of its contents. This research may relate to investments or services of a person outside of the UK or to other matters which are not regulated by the FSA or
in respect of which the protections of the FSA for private customers and/or the UK compensation scheme may not be available, and further details as to where this may be the case
are available upon request in respect of this report.
Any Nielsen Media Research material contained in this report represents Nielsen Media Research's estimates and does not represent facts. NMR has neither reviewed nor approved
this report and/or any of the statements made herein.
If this report is being distributed by a financial institution other than Credit Suisse AG, or its affiliates, that financial institution is solely responsible for distribution. Clients of that
institution should contact that institution to effect a transaction in the securities mentioned in this report or require further information. This report does not constitute investment advice
by Suisse to the dients of the distributing financial institution, and neither Credit Suisse AG, its affiliates, and their respective officers, directors and employees accept any
liability whatsoever for any direct or consequential loss arising from their use of this report or its content.
Copyright 2010 CREDIT SUISSE AG and/or its affiliates. All rights reserved.
CREDIT SUISSE SECURITIES (USA) LLC
United States of America: + 1 (212) 325-2000
ITT and DV Upgrade.doc
From: Jenkins, Harold
To: Yum Georgia
CC: Marinucci Fred
Date: 4/13/2010 1:47:18 PM
Subject: Gainful employment--lr.: (b-.-: )(7- Sl--------------,
(b)(5)
Harold
-----Original Message-----
From: Finley, Steve
Sent: Tuesday, April 13, 2010 10:21 AM
To: Jenkins, Harold; Marinucci, Fred; Wolff, Russell; Sann, Ronald; Yuan, Georgia
Subject: FW: Signal Hill- Post-Secondary Education: Relief: Gainful Employment Gains Alternative Measure
(b)(5)
Business Services- Education Services
Industry Update
April 13, 2010 Relief: Gainful Employment Gains Alternative Measure
Trace Urdan
415.364.0365
turdan@signalhill.com
Our Call:
As anticipated, Department ofEducation (USDOE) draft regulations went to the Office ofManagement & Budget
(OMB) last Friday 4/9 for review prior to their publication in the Federal Register --likely 5/15. A credible source close
to OMB te11s us that while the 8% median debt/income measure, and the 90% student loan repayment measures appear
to be essential! y unchanged from the tenus presented by the USDOE in January, a third alternative measure has been
added.
This measure would allow programs with a graduation rate of 50% or better and a subsequent job placement (in the
relevant field) of 70% or better to qualify out of the other two measures.
Though the devil will certainly be in the details, including how these items are to be considered for students that are
transferring credits and may be already employed, the new measure effectively removes the significant threat the rules
had created for nationally-accredited degree programs with typically high default rates. The new measure, in fact, seems
to closely resemble rules already imposed by national accrediting bodies. And while we might anticipate that the tenus
could be stricter in USDOE's conception, they should be eminently achievable with minimal disruption for all programs.
According to USDOE analysis, (in process of being reviewed by OMB,) the rules themselves would only affect 6-8% of
all for-profit programs, with culinary, automotive tech, and nursing programs hardest hit. (yVe note that this seems out of
keeping with our knowledge of these programs as represented by publicly-traded schools, but nevertheless appears to
represent a conclusion reached by USDOE.) Our contact indicated that the University of Phoenix
(NASDAQ: APOL; Buy) was not seen as affected at all in the USDOE analysis.
While we expect all post-secondary school stocks to benefit from the news, we believe that the those names that had
been most challenged by the existing tenus are likely to see the biggest benefit as a result of the change. These include
m Educational Services (NYSE: ESI; Buy), Corinthian Colleges
(NASDAQ: COCO; Buy), and Career Education (NASDAQ: CECO; Buy) which recently heralded graduation rates
comfortably above 50%. Though some individual programs may still fail all three tests, we believe these are likely to be
isolated, rare and fixable.
We view this apparent concession as evidence of a desire by Secretary Duncan to defuse what was promising to be an
area of real contention, as evidenced by House Republicans, members of the Congressional Black Caucus, and Senator
Lamar Alexander all calling out the Secretary on this issue. Secretary Duncan needs, in our opinion, a diverse coalition of
supporters to pass his signature issue: K-12 education refonn. Because the Secretary's K-12 proposals do not easily fall
within typical party boundaries, he really does need a bipartisan coalition in both the House and Senate education
committees to pass his version ofESEA (NCLB). No concession at all would have kicked off a loud and publicized
summer of public commentary. Because it is very difficult to argue that a 50% graduation rate and a 70% employment
rate is an onerous burden, we think USDOE has effectively neutered the campaign to ditch the rule entirely.
We still expect industry to challenge the rule in public and likely later in court, but based on our reliable source, we
believe investors need no longer fear that significant revenues could be at risk in the event that the rules are passed.
Please see end of this report for important disclosure(s)
Investment Analysis:
Important Disclosures
Analyst Certification
I, Trace Urdan, hereby certify that all of the views expressed in this research repmt accurately reflect my personal views
about the subject securities or issuers. I also certify that no part of my compensation was, is or will be directly or
indirectly related to the specific recommendations or views expressed in this research report. Signal Hill does not
compensate its equity research analysts based on specific investment banking transactions.
Signal Hill Equity research analysts receive compensation based on several factors, including overall profitability and
revenues of the finn, which include investment banking revenues.
Meaning ofRatings
Signal Hill uses a three-tiered rating system defined as follows:
BUY: We expect this stock to outperform its peers over the next 12 months:
HOLD: We expect this stock to perfom1 in line with its peers over the next 12
months:
SELL: We expect this stock to underperfonn its peers over the next 12 months:
Distribution ofRatings/JB Services
Signal Hill
1B Serv./Past 12 Mos.
Rating Count Percent Count Percent
BUY(BUY) 77 61.6 72 93.5
HOLD(HOLD) 47 37.6 38 80.9
SELL(SELL) 1 0.8 1 100.0
Disclaimer
This report has been prepared using sources we deem to be reliable but we do not guarantee its accuracy and it does
not purport to be complete. This report is published solely for information purposes and is not intended to be used as the
primary basis for making investment decisions, which should reflect the investment objectives and financial situation of the
investor.
The opinions expressed herein are subject to change without notice. This report is not an offer or the solicitation of an
offer to buy or sell securities. Additional information is available upon request.
DRAFT
PRIVILEGED DOCUMENT
ATTORNEY WORK PRODUCT
MEMORANDUM
(b)(5)
6
DRAFT
PRIVILEGED DOCUMENT
ATTORNEY WORK PRODUCT
(b)(5)
6
(b)(S)
DRAFT
PRIVILEGED DOCUMENT
ATTORNEY WORK PRODUCT
6
DRAFT
PRIVILEGED DOCUMENT
ATTORNEY WORK PRODUCT
(b)(5)
6
(b)(5)
DRAFT
PRIVILEGED DOCUMENT
ATTORNEY WORK PRODUCT
6
(b)(5)
DRAFT
PRIVILEGED DOCUMENT
ATTORNEY WORK PRODUCT
6
(b)(5)
DRAFT
PRIVILEGED DOCUMENT
ATTORNEY WORK PRODUCT
6
DRAFT
PRIVILEGED DOCUMENT
ATTORNEY WORK PRODUCT
(b)(5)
6
DRAFT
PRIVILEGED DOCUMENT
ATTORNEY WORK PRODUCT
(b)(5)
6
DRAFT
PRIVILEGED DOCUMENT
ATTORNEY WORK PRODUCT
(b)(5)
6
April 13, 2010
Trace Urdan
turdan@signalhill.com
415.364.0365
Business Services - Education Services
Industry Update
Relief: Gainful Employment Gains Alternative Measure
Our Call:
As anticipated, Department of Education (USDOE) draft regulations went to the Office of
Management & Budget (OMB) last Friday 4/9 for review prior to their publication in the Federal
Register -- likely 5/15. A credible source close to OMB tells us that while the 8% median
debt/income measure, and the 90% student loan repayment measures appear to be essentially
unchanged from the terms presented by the USDOE in January, a third alternative measure has
been added. This measure would allow programs with a graduation rate of 50% or better and a
subsequent job placement (in the relevant field) of 70% or better to qualify out of the other two
measures.
Though the devil will certainly be in the details, including how these items are to be considered
for students that are transferring credits and may be already employed, the new measure
effectively removes the significant threat the rules had created for nationally-accredited degree
programs with typically high default rates. The new measure, in fact, seems to closely resemble
rules already imposed by national accrediting bodies. And while we might anticipate that the
terms could be stricter in USDOE's conception, they should be eminently achievable with
minimal disruption for all programs.
According to USDOE analysis, (in process of being reviewed by OMB,) the rules themselves
would only affect 6-8% of all for-profit programs, with culinary, automotive tech, and nursing
programs hardest hit. (We note that this seems out of keeping with our knowledge of these
programs as represented by publicly-traded schools, but nevertheless appears to represent a
conclusion reached by USDOE.) Our contact indicated that the University of Phoenix (NASDAQ:
APOL; Buy) was not seen as affected at all in the USDOE analysis.
While we expect all post-secondary school stocks to benefit from the news, we believe that the
those names that had been most challenged by the existing terms are likely to see the biggest
benefit as a result of the change. These include ITT Educational Services (NYSE: ESI ; Buy) ,
Corinthian Colleges (NASDAQ: COCO; Buy), and Career Education (NASDAQ: CECO; Buy)
which recently heralded graduation rates comfortably above 50%. Though some individual
programs may still fail all three tests, we believe these are likely to be isolated, rare and fixable.
We view this apparent concession as evidence of a desire by Secretary Duncan to defuse what
was promising to be an area of real contention, as evidenced by House Republicans, members
of the Congressional Black Caucus, and Senator Lamar Alexander all calling out the Secretary
on this issue. Secretary Duncan needs, in our opinion, a diverse coalition of supporters to pass
his signature issue: K-12 education reform. Because the Secretary's K-12 proposals do not
easily fall within typical party boundaries, he really does need a bipartisan coalition in both the
House and Senate education committees to pass his version of ESEA (NCLB) . No concession
at all would have kicked off a loud and publicized summer of public commentary. Because it is
very difficult to argue that a 50% graduation rate and a 70% employment rate is an onerous
burden, we think USDOE has effectively neutered the campaign to ditch the rule entirely.
We still expect industry to challenge the rule in public and likely later in court, but based on our
reli able source, we believe investors need no longer fear that significant revenues could be at
risk in the event that the rules are passed.
Please see important disclosure information on page 2 of this report.
April13, 2010
Important Disclosures
Analyst Certification
I, Trace Urdan, hereby certify that all of the views expressed in this research report accurately reflect my
personal views about the subject securities or issuers. I also certify that no part of my compensation was, is or
will be directly or indirectly related to the specific recommendations or views expressed in this research
report.Signal Hill does not compensate its equity research analysts based on specific investment banking
transactions. Signal Hill Equity research analysts receive compensation based on several factors, including
overall profitability and revenues of the firm, which include investment banking revenues.
Meaning of Ratings
Signal Hill uses a three-tiered rating system defined as follows:
BUY: We expect this stock to outperform its peers over the next 12 months:
HOLD: We expect this stock to perform in line with its peers over the next 12 months:
SELL: We expect this stock to underperform its peers over the next 12 months:
Rating
BUY
HOLD
SELL
Disclaimer
Distribution of Ratings/IS Services
Signal Hill
Count
77
47
1
Percent
61.6
37.6
0.8
18 Serv./Past 12 Mos.
Count
72
38
Percent
93.5
80.9
100.0
This report has been prepared using sources we deem to be reliable but we do not guarantee its accuracy and
it does not purport to be complete. This report is published solely for information purposes and is not intended
to be used as the primary basis for making investment decisions, which should reflect the investment objectives
and financial situation of the investor: The opinions expressed herein are subject to change without notice. This
report is not an offer or the solicitation of an offer to buy or sell securities. Additional information is available
upon request.
Post-Secondary Education 2
From: Shireman, Bob
To: Yuan. Georgia
Manheimer Ann
Arsenault, Leigh
CC:
Date: 4/4/2010 12:07:28 PM
Subject: GE deck-- Confidential DELIBERATIVE PROCESS
Attachments: GE 1 .ppt
(b)( 5)
Robert Shireman
Deputy Undersecretary
U.S. Department ofEducation
(202) 260-0101
(b)(5)
From: K vaal James
To: Bergeron. David
Yuan, Georgia
Martin. Phil
Arsenault L e i ~ h
CC:
Date: 7/20/2010 8:42:52 PM
Subject: GE document control
(b)(S)
(b)(5)
(b)(5)
(b)(5)
(b)(S)
(b)(S)
From: Kanter, Martha
To.: Rose, Charlie
Weiss. Joanne
Cunningham, Peter
Mattin, Cannel
Gomez. Gabriella
Yuan Georgia
CC: Kvaal James
Date: 9/13/2010 10:45:02 PM
Subject: GE documents
Attachments: 090910 lobbying and repay rates 2 ppt
(b)(5)
Martha
From: Kvaal, James
Sent Monday, September 13, 2010 6: 12 PM
To: Miller, Tony
Cc: Kanter, Martha
Subject: FW: GE documents
Tony,
(b)(5)
Thanks
James
DRAFT* PRELIMINARY * PREDECISIONAL
INTERNAL DRAFT
(b)(5)
(b)(5)
lNTERNAL DRAFT
(b)(S)
From: Arsenault Leigh
To: Bergeron. David
Yuan, Georgia
Martin. Phil
Kvaal James
CC:
Date: 7/21/2010 11:21:08 AM
Subject: GE documents, revised
All, thanks for your feedback and comments. Attached are the cleaned-up documents.
Thanks!
Leigh
(b)(5)
1
(b)(5)
1
(b)(S)

(b)(5)
(b)(S)
(b)(5)

From: Yuan Georgia
To: K vaal, James
Weiss. Joanne
CC: Bergeron. David
Rose Charlie
Date: 9/24/2010 5:35:08 PM
Subject: GE- Meetings- Attorney-Client Privileged and Confidential
James and Joanne
(b){5)
Georgia
Georgia Yuan
Deputy General Counsel
Postsecondary and Regulatory Service
LBJ 6E341
202-40] -63 99
(b)(S)
DRAFT 4 *INTERNAL DOCUMENT* DELIBERATIVE PROCESS
CONFIDENTIAL
2
(b)(5)
(b)(5)
2
(b)(5)
Charli e
Sent using BlackBerry
From: Rose Charlie
To.: Yuan. Georgia
CC: Borders. Tia
Date: 3/3112010 5:56:34PM
Subject: Kaplan and devry meeting

You might also like