Professional Documents
Culture Documents
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ.
Not for distribution to the public. Copyright 2014 by Standard & Poors Financial Services LLC (S&P). All rights reserved.
Todays Speakers
Marcel Heinrichs
Director, Business Development
S&P Credit Solutions
S&P Capital IQ
Mark Williams
Executive-in-Residence/Master Lecturer
Finance Department
Boston University School of Management
Alma Chen
Associate Director
Analytics Development
S&P Capital IQ
(Moderator)
Pleaseornote:
The views
and opinions
expressed
by Mr.requires
Williamsthedoprior
not necessarily
reflect
S&P
IQ and itstoaffiliates.
Permission to reprint
distribute
any content
from this
presentation
written approval
of the
S&Popinion
CapitalofIQ.
NotCapital
for distribution
the public.
Topics Of Discussion
II
III
IV
The case for utilizing both market signals and fundamental measures of
credit risk
VI
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Narrow definition: risk that a borrower will default on its issued debt
Wider definition: risk that a business partner cannot fulfil financial obligations
Examples:
Loss of interest payments and principal
Loss in investment
Disruption to cash flows
Increased collection costs
Potential bankruptcy
Need for Regulatory Reporting
Business disruption
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
(INVESTMENT) BANK
CORPORATE
A
INSURER
Underwriting
Credit Analysis
Risk Management
COMMERCIAL
LENDER
Loan Origination
Credit Department
Risk Management
Leveraged Finance
Restructuring
ASSET /INVESTMENT
MANAGER
Idea Generation Pre-Trade
Credit Analysis Pre- and Post-Trade
Portfolio and Performance Risk
Management
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Period of 2001-2013; entire universe of publicly rated companies by Standard & Poors Ratings Services,
that are also listed at a stock market
7316 companies, of which 200 defaulted on their issued debt
Assess all companies with two different kind of models; for defaulters exactly one year prior the actual default.
One model is based on fundamental data, the other is based on stock price volatility as a market signal
generate the following matrix of observed default rates per bucket;
observed default rate = number of defaulters / total number of entities
aaa
aa+ to aaa+ to abbb+ to bbbbb+ to bbb+ to bccc+ or worse
0.00 to
0.01%
(aaa to aa-)
0.00
0.00
0.00
0.00
0.00
0.00
N/A
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
FundamentalsBased Scoring
Model
II
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
II
Problem II: Complex Global Credit Matters And Different Credit Signals
Investing in
emerging markets
14,000
Suppliers
from around
the world
Poor You
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
III
Scoring Models
Scoring models produce a credit score (lower case letter grade such as bbb-), which is then also
mapped to a Probability of Default (PD). However, the primary output and main interest of its users is
the credit score as a qualitative measure of credit risk
Developed on ratings (full scale from AAA to D = default) or similar assessments such as shadow
ratings, credit estimates etc.
Favored by clients with an affinity to ratings, usually with a background as a fundamental credit
risk analyst
Input Data
Ratings and
explanatory factors
State-of-the-art
Modeling Recipe
Output Data
Credit scores in
lower case letters
Cash
EBITDA
Total Assets
Debt/Capital
Proprietary
Algorithm
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
III
PD Models
PD models produce a PD in the first place, which is then also mapped to a credit score. The primary
output and main interest of its users is the PD as a quantitative measure of credit risk
Developed on default data (binary decision: either a company defaulted on its debt repayments in a
particular year or not)
Favored by clients who are not used to ratings as a rank measure or who do not believe in the
relevance of ratings, and often have a quantitative background
Input Data
Default flags and
explanatory factors
State-of-the-art
Modeling Recipe
Output Data
Default probabilities
Cash
EBITDA
Total Assets
Debt/Capital
0 010
Proprietary
Algorithm
0.26% 1.59%
29.64% 0.05%
0.46%
0 0 1 0 1 0 0
10 1 1 0 0
10
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
1.21%
III
Fundamental Data: Any data that is usually collected at periodically, often annually or
quarterly (in rare cases monthly)
Firm-specific financials (Annual report, quarterly financial statements)
Systemic Risk factors such as
(other) Factors that reflect the environment that a company operates in vis-a-vis country risk, industry risk or
sovereign risk
Market Signals: Any data that is usually collected at high frequency, most often daily or even
intra-daily
CDS spreads of companies whose credit risk is traded in the CDS market
Fixed Income spreads of companies that issue debt via bonds or similar instruments
Stock market volatility of public companies
Since ratings are based on fundamental data, anyone with an affinity to ratings tend to
favor models that are based on fundamental data
Anyone that find ratings less relevant tend to favor models that rely heavily (or solely)
on market signals
11
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Longer/Less
volatile
Time Horizon
Shorter/more
volatile
III
12
Bond
spreads
Stock
Price
Volatility
Short- to Mid-Term
Mid- to Long-Term
Public
Ratings
Many
FundamentalsBased Models
For ratings, these are expected to be stable for 3-5 years for
investment grade (IG) and 2-3 years for non-IG companies
Much less volatile results
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
III
Time Horizon
Coverage
Usual
Primary
Measure
Qualitative Judgment
Fundamentals
Scoring or PD Model
PD in %
Then mapped to
credit score
PD in %
Then mapped to
credit score
PD in %
Then mapped to
credit score
Global Coverage
~9k companies
Global Coverage
Unlimited applicability
Global Coverage
Unlimited applicability
Publicly listed
Companies
38k companies
Companies w/
liquid bond market
~6k companies
Companies w/
liquid CDS market
>1k companies
Medium to
long- term metric
Medium to
long-term metric
Medium to
long- term metric
Point-in-Time
metric
Point-in-Time
metric
Point-in-Time
metric
Fundamentals-based quantitative
models expand the rated
universe to any public or private
company around the globe for a
medium- to long-term view of
credit risk
Stock Price
Volatility
Bond Spread
*From Standard & Poors Ratings Services. S&P Capital IQ, as well as its products and services are analytically and editorially separate and independent from other analytical
areas at S&P, including S&P Credit Ratings.
13
CDS Spread
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
IV
30.0%
20.0%
10.0%
0.0%
aaa
aa
CreditModel Score
bbb
bb
ccc &
below
PD Market Signals
In this example, we classified all companies with a Market Signal PD < 9.64%,
or a CreditModel score of b- and above, as healthy companies.
Type I Error:
Number of
defaults in
healthy group /
Total Number of
Defaults
Detected too late:
lose money
because of wrong
acceptance of
business
engagement
Source: Bankruptcy and default data from SP
CreditPro, CreditModel Scores from S&P
Credit Analytics, Market Signals PD from S&P
Capital IQ, from 2001 to 2013.
For illustrative purposes only.
Conclusion:
Market Signals PD Have Lower Type I Errors in the short-term
We can use these as a first cut to shortlist potential defaulters
14
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
IV
Smaller
area below
the red line
than the
blue line in
the shaded
area
60.0%
50.0%
40.0%
30.0%
20.0%
10.0%
0.0%
aaa
aa
CreditModel Score
bbb
bb
ccc &
below
PD Market Signals
In this example, we classified all companies with a Market Signal PD > 9.64%,
or a CreditModel score of ccc+ and below, as unhealthy companies.
Type II Error:
Number of nondefaulters in badcompanies group
/ total number of
healthy
companies;
False alarms: lose
money because of
wrong rejection of
business
engagement
Source: Bankruptcy and default data from SP
CreditPro, CreditModel Scores from S&P Credit
Analytics, Market Signals PD from S&P Capital
IQ, from 2001 to 2013.
For illustrative purposes only.
Conclusion:
CreditModel scores have lower type II errors in the medium- to long-term
First, shortlist potential defaulters using PD Market Signals, then use CreditModel
scores to narrow down the list of potential defaulters
15
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
PD Model Fundamentals
Standard & Poors Ratings
British Petroleum (LSE:BP.)s
share price fell and its CDS spiked
during the oil spill in year 2010
MDS CDS
Scores
Source: S&P Ratings, S&P CreditModel Scores, and PD Market Signals from S&P Capital IQ RatingsDirect, October 2008 October 2013.
Key Developments news from S&P Capital IQs news sources.
16
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
2.27 to 9.64%
(b+ to b-)
>9.64%
(ccc+ or worse)
aaa
0.00
0.00
0.00
0.00
0.00 N/A
aa+ to aa-
0.00
0.00
0.00
0.00
0.00
0.00
0.00
a+ to a-
0.00
0.00
0.00
0.00
0.00
0.00
0.00
bbb+ to bbb-
0.00
0.00
0.03
0.15
0.18
0.75
1.19
bb+ to bb-
0.00
0.00
0.15
0.22
0.62
1.44
4.78
b+ to b-
0.00
0.28
0.46
1.15
3.35
6.07
10.89
0.00
0.00
6.19
11.43
20.00
28.53
ccc+ or worse
N/A
N/A
Double-Red Flag Candidates Around the Globe from Various Sectors as of Sep 30, 2014 (Excerpt)
Company
Industry
Country
CreditModel Score
PD Market Signals
Brazil
cc
53.90% (cc)
Mortgage Finance
Puerto Rico
cc
40.89% (cc)
Airlines
Germany
ccc
22.46% (ccc-)
Indonesia
cc
20.42% (ccc-)
US
ccc-
16.15% (ccc)
Petrobras Argentina SA
Argentina
ccc-
16.02% (ccc)
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Summary I
1. Always remember:
Type 1 errors (accepting bad customers): market signals-based models are superior in
detecting (rapid) credit deterioration &
Issuer Credit
Ratings
18
FundamentalsBased
Credit Scoring
or PD Model
Market SignalsBased PD
Models
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Summary II
Safe
Haven
2.27 to 9.64%
(b+ to b-)
>9.64%
(ccc+ or worse)
aaa
0.00
0.00
0.00
0.00
0.00 N/A
aa+ to aa-
0.00
0.00
0.00
0.00
0.00
0.00
0.00
a+ to a-
0.00
0.00
0.00
0.00
0.00
0.00
0.00
bbb+ to bbb-
0.00
0.00
0.03
0.15
0.18
0.75
1.19
bb+ to bb-
0.00
0.00
0.15
0.22
0.62
1.44
4.78
b+ to b-
0.00
0.28
0.46
1.15
3.35
6.07
10.89
0.00
0.00
6.19
11.43
20.00
28.53
ccc+ or worse
N/A
N/A
Stay
Away
5. For companies with mixed signals, follow the suggested approach in our paper:
http://bit.ly/1zelpbO
19
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
VI
Standard & Poors Ratings Services, S&P Dow Jones Indices and S&P Capital IQ are
engaged with the academic sector in order to continuously provide best-in class data
and analytics both for research and for any of our customers immediate workflows.
Examples of current credit risk projects include:
Analysis of discriminatory power of behavioral data in credit risk with MSc students from Columbia
University; students get credit for this project as part of their curriculum
Project with world-renowned professor of economics from NYU on analysis of ratings momentum
Speaking engagements in credit risk to academics and/or (financial engineering) students from various
universities in the U.S. and the UK
Independent review of our suite of credit risk models by academics from top university in Far East
Well established program of internships in fall with MSc students in financial engineering from University
of Berkeley
Distribution of our data and research articles via WRDS (Wharton Research Data Services)
20
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Frances national railway service SNCF is a GRE and has a standalone rating of BBB-,
but its final rating is AA- (6 notches up!) because of its criticality to Frances
infrastructure system. Frances sovereign rating is AA
Petrobras Argentina gets one notch uplift for assumed support from its parent company
in Brazil
Country Risk
Sovereign Risk
Industry Risk
Economic Risk
Distinguishes risk at issuance (or facility) level, while default risk is assessed at
company-level
4. So much more
21
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Q&A
Marcel Heinrichs
S&P Capital IQ
22
Mark Williams
Boston University School of Management
Alma Chen
S&P Capital IQ
(Moderator)
Pleaseornote:
The views
and opinions
expressed
by Mr.requires
Williamsthedoprior
not necessarily
reflect
S&P
IQ and itstoaffiliates.
Permission to reprint
distribute
any content
from this
presentation
written approval
of the
S&Popinion
CapitalofIQ.
NotCapital
for distribution
the public.
Biographies
Marcel Heinrichs, CFA
Director, Business Development, S&P Credit Solutions, S&P Capital IQ
Marcel is responsible for the market development of credit risk offerings to financial institutions and non-financial corporations in the Americas. In this role, Marcel defines
the roadmap for new offerings of content, tools and analytics, works with marketing and sales teams on activities for branding and sales generation, oversees thought
leadership and interaction with key market influencers including top clients, regulators or associations and paves the path for new markets and client segments. Prior to his
current role, Marcel was global head of the Analytic Development Group (ADG) of S&P Capital IQ, responsible for the analytical innovation, development, maintenance and
ongoing validation of all credit risk models and products. Until 2010, Marcel was based in London and co-leading the services team of S&P Risk Solutions EMEA, the
consultancy business of S&P Capital IQ. Before joining S&P Risk Solutions in 2004, Marcel taught courses in econometrics, financial econometrics, mathematical
economics and macroeconomics at the London School of Economics and consulted various financial institutions on a variety of modeling problems. Marcel is also a
member of the Financial Markets Group, the Research Center in Finance of the London School of Economics. He has a Master degree in economics from the University of
Bonn, Germany and Ecole Nationale de la Statistique et de lAdministration Economique (ENSAE), France.
Mark Williams
Executive-in-Residence/Master Lecturer, Finance Department, Boston University School of Management
Mark is an academic, author, columnist and risk management expert. Prior to joining Boston University he worked as a trust banker, senior trading floor executive and as a
Federal Reserve Bank examiner. Since 2002, he has been on the finance faculty at Boston University specializing in banking, energy and capital markets related matters.
He teaches at the graduate and undergraduate levels. In 2008 he was awarded the Boston University Beckwith Prize for excellence in teaching. Mark frequently appears in
the national media and has been a guest columnist for the Financial Times, New York Times, Reuters.com, Forbes.com, Business Insider, Boston Globe and Foreign Policy
Magazine. In 2010, his book Uncontrolled Risk, detailing the rise and fall of Lehman Brothers and root causes of the financial crisis was published by McGraw Hill.
www.uncontrolledrisk.com. In 2013 he coauthored Longwood Covered Courts and the Rise of American Tennis. This work won a best book award at the New England
Book Show. In 2014 he provided Congressional testimony relating to the risks associated with virtual currencies. In addition to teaching and expert witness work, he
services on several boards including Appleton Partners LLC, a Boston-based, wealth-management company and Standard & Poors Academic Advisory Council. Mark
holds a BSBA in Finance from the University of Delaware and a MBA from Boston University. He is also a founding board member of the Boston Chapter of the Global
Association of Risk Professional, a member of the Boston Analyst Security Association and International Association of Financial Engineers.
Alma Chen
Associate Director, Analytics Development, S&P Capital IQ
Alma is Head of the Analytic Development Group (ADG), Americas, and is currently based in New York. (Formerly, Head of ADG APAC, based in Hong Kong.) Her team is
focusing on analytical development, maintenance and ongoing validation of credit risk models and products, which are used by financial institutions and other creditsensitive entities to measure and manage credit risk, also within regulatory frameworks such as Basel II/III or Solvency II. Her team provides analytical support to existing
clients and Sales during pre-sales support, as well as to Risk Solutions, for ad-hoc assignments in Americas Region and APAC respectively. She has more than 12 years of
experience in the risk analysis and financial modeling. Prior to joining S&P Capital IQ, Alma was a Lead Consultant, who has provided robust and accurate solutions on
credit risk quantitative & expert-judge hybrid models for key components of Expected Loss: Probability of Default, Loss Given Default & Exposure at Default, including
different stages of modeling cycle: development, calibration, performance monitoring, validation and optimization. Before moving to Asia in 2008, Alma worked as an
economist of a U.S.-based company engages in mortgage purchasing, credit guarantee, issuing guaranteed mortgage-related securities and portfolio investment activities,
where Alma accumulated seven years of extensive experience in financial model development, validation and calibration, Alma also conducted economic research and
analysis. Alma holds a Masters Degree in Statistics from Texas A&M University in the United States, and a Bachelors Degree from Tsinghua University in China.
23
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Appendix
24
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Key Weaknesses:
Unable to detect changes in fundamentals between reporting periods
May react too slowly for equity investors and for fixed income investors with
short holding horizons
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Bonds-driven
CDS- driven
Strengths
Covers all publicly listed
companies, including
emerging markets
Weaknesses
May be noisy
Equity prices can react to
non-credit related events
Equity prices can over-react
to news, and exhibit shortterm reversals
Market Derived Signals are represented in lowercase nomenclature to differentiate them from S&P Credit Ratings.
26
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Issuer Credit
Ratings
27
FundamentalsBased
Credit Scoring or
PD Model
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Global coverage includes 246 countries including emerging and frontier markets.
Market Derived Signals are represented in lowercase nomenclature to differentiate them from S&P Credit Ratings.
28
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Managing high-risk
entities:
Adjust exposure terms
(less amount/ higher rate)
Outsource the risk, e.g.
insurance cover
Terminate exposure
Credit Decision
Accept or reject
exposure
Surveillance and
Monitoring
Generating earlywarning indicators
Adjusting reserves
Calculating Value at
Risk (VaR)
29
In-Depth Analysis
Incorporating
entities into a
portfolio
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
www.spcapitaliq.com
Copyright 2014 by Standard & Poors Financial Services LLC, a subsidiary of McGraw Hill Financial Inc. All rights reserved. No content (including ratings, credit-related analyses and data, valuations, model,
software or other application or output therefrom) or any part thereof (Content) may be modified, reverse engineered, reproduced or distributed in any form by any means, or stored in a database or retrieval
system, without the prior written permission of Standard & Poors Financial Services LLC or its affiliates (collectively, S&P). The Content shall not be used for any unlawful or unauthorized purposes. S&P and
any third-party providers, as well as their directors, officers, shareholders, employees or agents (collectively, S&P Parties) do not guarantee the accuracy, completeness, timeliness or availability of the Content.
S&P Parties are not responsible for any errors or omissions (negligent or otherwise), regardless of the cause, for the results obtained from the use of the Content, or for the security or maintenance of any data
input by the user.
The Content is provided on an as is basis. S&P PARTIES DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, ANY WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE OR USE, FREEDOM FROM BUGS, SOFTWARE ERRORS OR DEFECTS, THAT THE CONTENTS FUNCTIONING WILL BE
UNINTERRUPTED OR THAT THE CONTENT WILL OPERATE WITH ANY SOFTWARE OR HARDWARE CONFIGURATION. In no event shall S&P Parties be liable to any party for any direct, indirect,
incidental, exemplary, compensatory, punitive, special or consequential damages, costs, expenses, legal fees, or losses (including, without limitation, lost income or lost profits and opportunity costs or losses
caused by negligence) in connection with any use of the Content even if advised of the possibility of such damages.
Credit-related and other analyses, including ratings, and statements in the Content are statements of opinion as of the date they are expressed and not statements of fact or recommendations to purchase, hold,
or sell any securities or to make any investment decisions. S&P assumes no obligation to update the Content following publication in any form or format. The Content should not be relied on and is not a
substitute for the skill, judgment and experience of the user, its management, employees, advisors and/or clients when making investment and other business decisions. S&Ps opinions and analyses do not
address the suitability of any security. S&P does not act as a fiduciary or an investment advisor except where registered as such. While S&P has obtained information from sources it believes to be reliable,
S&P does not perform an audit and undertakes no duty of due diligence or independent verification of any information it receives.
S&P keeps certain activities of its business units separate from each other in order to preserve the independence and objectivity of their respective activities. As a result, certain business units of S&P may have
information that is not available to other S&P business units. S&P has established policies and procedures to maintain the confidentiality of certain non-public information received in connection with each
analytical process.
S&P provides a wide range of services to, or relating to, many organizations, including issuers or underwriters of securities or obligors, investment advisers, broker-dealers, investment banks, other financial
institutions and financial intermediaries. As a result, S&P may receive fees or other economic benefits from organizations whose securities or services it may recommend, analyze, rate, include in model
portfolios, evaluate, price or otherwise address.
STANDARD & POORS, S&P, CREDITMODEL and GICS are registered trademarks of Standard & Poors Financial Services LLC. Capital IQ is a registered trademark of Capital IQ, Inc. S&P CAPITAL IQ is a
trademark of Standard & Poors Financial Services LLC. MICROSOFT, WORD, POWERPOINT, and EXCEL are registered trademarks or trademarks of Microsoft Corporation in the United States and/or other
30
Permission
to reprint
or may
distribute
contentoffrom
presentation
requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
countries. All other product
or service
names
be theany
property
theirthis
respective
owners.