You are on page 1of 23

The historical background to the current debate

Should commercial
and investment
banking be separated?

Dominic Casserley
Philipp Hrle
James Macdonald

Contents
Summary

Introduction

Part I The rise of universal banking

The United States: Freedom and checks

Germany: A long tradition

10

The United Kingdom: First separation, then integration

11

Part II The historical background to the arguments used in todays debate

15

1 The advantages and drawbacks of universal banks


Are universal banks inherently less stable than specialist commercial banks?

15

Are big universal banks better for the economy?

16

Does universal banking create overmighty subjects?

16

Does universal banking lead to dangerous conflicts of interest?

17

Do universal banks have a competitive edge?

17

2 Moral hazard
Does underwriting the banking system encourage undue risk taking?

18

Under what conditions do banks become too big to fail, and threaten to become too big to bail?

19

3 The economic cost of splitting up the banks


Would splitting up the banks put economic recovery at risk?

20

4 Regulation or separation
Should banks be split up, or simply better regulated?

21

Lessons from history

23

Notes

24

Further reading

25

Should commercial and investment banking be separated?


Summary

Summary
One of the key questions that the UK government has
mandated the Independent Commission on Banking to
answer is whether universal banks should be split into
commercial and investment banks, have their scope of
operations narrowed, or have their scale reduced.
This question is not new. It has been discussed on
many different occasions in many countries, whether
in response to a banking crisis or not. Although
todays banks and financial markets have developed
tremendously and become far more complex since
the question was last aired, the current debate cannot
easily be understood without a knowledge of the
history that lies behind it.
The rise of universal banking in all major Western
markets is the first strand in this history, and can best
be illustrated by the contrasting experiences of the
United States, Germany, and the United Kingdom.
Whereas the large German banks have been universal
banks for a long time, the separation of commercial
and investment banking by law was a reality in the US
for much of the twentieth century. In the UK, universal
banking developed relatively recently, in the 1980s,
although it was not prohibited by law prior to that.

A look at the historical background of the


arguments used in todays debate is the second
strand of the analysis presented here. The dangers and
advantages of universal banks have come under close
scrutiny on several previous occasions, most notably
in the United States in the 1930s. Some of todays
arguments for or against universal banking were used
on those occasions; others were not. The historical
arguments in favour of universal banking resemble
those used today, whereas the historical arguments
against universal banking focused largely on the
avoidance of conflicts of interest and the restriction of
power. Moral hazard arguments played a limited role,
and systemic risk barely figured in the debate.
Whenever the separation of commercial and
investment banking has been discussed in the wake
of a banking crisis, the debate has focused, as it does
today, on solving the issues that the crisis has created.
All such debates have been highly political, and their
outcomes have been driven more by the strength of
politicians opinions on the immediate crisis than by
deep analysis of the long-term causes.

Should commercial and investment banking be separated?


Introduction

Introduction
The severity of the 2008 financial crisis has inevitably
prompted calls for strong measures to prevent any
recurrence. Blame for the crisis has been laid at many
doors: that of bankers, for their greed and hubris;
regulators, for their complacency; economists, for their
mistaken theories; politicians, for their encouragement
of subprime mortgage lending and a general increase in
leverage; and even Asian consumers, for their allegedly
excessive savings.

The new United Kingdom government has set up the


Independent Commission on Banking, chaired by
Sir John Vickers, to look at the reform of the banking
industry with the objectives of reducing systemic risk
. . . mitigating moral hazard . . . [and] reducing both the
likelihood and impact of firm failure. Its remit includes
structural and non-structural measures of reform
including the complex issue of separating retail and
investment banking functions.2

Another possible culprit is the universal banking model


that has come to dominate North American and European
banking over the past twenty years, and has become the
leading model across most of the rest of the world. Today
it is accepted wisdom that a globalized world economy
needs global banks capable of handling every type of
financial service as well as traditional lending. Many
banks balance sheets have increased substantially in
recent years and have started to incorporate types of
risk unimagined in earlier decades. When the banking
system faced collapse in the autumn of 2008, the scale of
the risks that governments assumed in order to stabilize
it shocked politicians and voters alike. So it is hardly
surprising that the reforms proposed to prevent future
crises include several that would involve splitting up
banks, either to narrow the scope of their operations or
simply to reduce their size.

The issues involved in reforming the banking system are


numerous and complex. The financial world has changed
dramatically over the past thirty years in the wake of
deregulation. However, the current debate cannot easily
be understood without a knowledge of the history that lies
behind it.

In the United States two former governors of the Federal


Reserve have thrown their weight behind such ideas.
Paul Volcker has argued that banks should no longer
be allowed to undertake proprietary trading or to invest
in hedge funds. His successor, Alan Greenspan, has
argued that the biggest banks need to be reduced in size
because If they are too big to fail, they are too big.1 Some
of Volckers ideas were taken up by the US government
and incorporated into the DoddFrank Act signed into law
in July 2010.

The first part of this paper summarizes the historical


evolution of the debate with reference to three countries.
The United States has the longest and richest history of
debates about universal banking, as well as experience
with forced separation by law. Germany, an exponent of
the continental European tradition of universal banks,
continued on this path even after severe banking crises.
The UK has had a long history of separated banks, and
although there were no legal barriers to universal banking,
it was not until the deregulation of the 1980s that this
model took hold.
The second part of this paper analyzes to what extent
todays arguments for and against universal banking were
used historically.

Should commercial and investment banking be separated?


Part I
The rise of universal banking

Part I
The rise of universal banking
The United States: Freedom and checks
In no other country has the debate over whether
commercial and investment banks should be separated
raged more fiercely or for so long.
The history of banking in America can be understood
only in the context of a continuing tension between
two opposing forces. One was the need for a powerful
financial sector that could fund the requirements of
a rapidly growing economy; the other was a deeprooted popular opposition to big banks, and to large
concentrations of financial power in general. This tension
goes back to the earliest years of the republic. It fed
into the National Bank Acts of 1863 and 1864, which
established the principle of the separation of commercial
and investment banking; it was an essential feature of
the GlassSteagall Act of 1933, which set this principle in
stone; and it is still in evidence in current political debates
on banking reform.

Early days
The First Bank of the United States was founded in 1791
by Alexander Hamilton, the first US Secretary of the
Treasury. It was modelled on the Bank of England, the
thinking being that if the new republic wanted to emulate
the economic success of Great Britain, it also needed to
emulate its financial system. However, popular opposition
to concentrations of financial power extended to the very
idea of a central bank, and in 1811 the disproportionate
representation of the smaller frontier states in the Senate
led to its abolition. The Second Bank of the United
States was founded in 1816, but it too fell foul of western
opposition, and its charter was not renewed when it
expired in 1836.
Thereafter the United States operated without a central
bank, and the only form of paper money to augment the
supply of coins was a motley assortment of notes issued by
approximately 7,000 state-chartered banks. Some of these
banks were sound, but many were insubstantial singlebranch unit banks whose notes circulated at a discount.
When the Civil War broke out in 1861, it was immediately
apparent that it could not be financed without a national
currency of some kind. The alternative, as Senator John
Sherman put it, would have been to depend on the
inflated currency of all the local banks in the United States;
banks over which you have no control, and which you
cannot regulate or govern in the slightest degree.3
The Lincoln administrations initial solution was to issue
paper money itself greenbacks, as they became
known. However, this was considered a stop-gap
emergency measure, and in 186364 the government
passed acts authorizing the creation of nationally
chartered banks. These banks would be permitted to

issue bank notes of uniform quality and appearance


which they would have to back with US government
bonds. The immediate purpose of this requirement was
to help finance the war. However, if government bonds
declined in value, bank solvency could be threatened.
So in order to ensure that the banks remained sound,
the legislation required them to hold 25 percent of their
deposits in cash, and restricted their lending to shortterm, self-liquidating business loans. State-chartered
banks continued to exist, but their notes were taxed out of
existence in the interests of national monetary uniformity.

Crises and panics


For the first time, America had a workable national
banking system. However, the system still harboured
deep-seated tensions, both financial and political, which
led to what might be termed the crisis of the big banks
from 1913 to 1933.
First of all, the absence of a central bank meant that the
biggest banks, which were located in New York, had
to act as an informal central banking system under the
leadership of John Pierpont Morgan, the pre-eminent
banker of the day. A series of banking crises culminating
in the devastating panic of 1907, when call-money rates
hit 100 percent and the stock market almost collapsed
tested this informal system to, or perhaps beyond, its
limit. Although Morgan and his banking allies succeeded
in calming the panic, it became clear that a central bank
was essential in a developed economy. The attempt
to manage without one had led to the concentration of
financial power in a few private banks, a situation that
popular opinion regarded as just as great a threat to
democratic liberties as any central bank.
In a parallel development, the big banks, again with
J. P. Morgan at the forefront, were playing a central role in
the consolidation of American business into giant trusts.
Many of these trusts were deliberately designed to end
destructive price wars and reduce competition, a process
that inevitably involved the banks in the capital markets,
notwithstanding the nominal provisions of the National
Bank Act. In 1890, the Sherman Antitrust Act started a
political counter-reaction to the growth of such business
monopolies, which culminated in the breaking up of
the Standard Oil and American Tobacco trusts in 1911.
Attention then turned to the so-called money trust a
small group of Wall Street bankers believed to lie behind
a web of interlocking shareholdings and directorships
that controlled the bulk of the American economy. In
1913 Democratic Congressman Arsne Pujo set up a
committee to investigate it.
The committee summoned leading bankers to testify,
and grilled them about the extent of their control over
corporate America. The hearings attracted extensive

news coverage and coincided with the publication of


Louis Brandeiss highly influential book Other Peoples
Money and How the Bankers Use It. Brandeis claimed
that the bankers controlled assets equivalent to $22 billion
at a time when the entire US GDP amounted to only $40
billion. Moreover, he identified the merging of investment
and commercial banking as one of the means by which
the bankers had increased their power over the economy,
and stated that the only legitimate sphere of the banking
business was the making of temporary loans to
business concerns. 4
Although the Pujo committee recommended that banks
be confined to commercial banking, it did not lead to the
splitting up of the banks. In fact, soon after it reported,
its advice became irrelevant: the capital markets more or
less closed to private borrowers for the duration of World
War I, and the banks were needed to sell Liberty Bonds
to finance the war. However, the report did help to ensure
the establishment of the Federal Reserve System in 1913,
which finally ended the central banking role of the big
national banks (despite the view of conspiracy theorists
that it was all a plot to enhance these banks power). The
report also led to the passage of the Clayton Antitrust Act
of 1914, which banned directors from holding positions in
competing companies.
During the boom years of the 1920s, inhibitions about
merging commercial and investment banking faded.
Although the Comptroller of the Currency (the official
regulator of the national banks) initially opposed the
setting up of securities businesses by banks, by 1920 he
had changed his mind. A decade later, there were 114
affiliated securities companies of national banks which
between them sold over 60 percent of all bond issues. It
seemed that the era of universal banking had arrived.
But it did not last for long. The financial crisis of the early
1930s soon put the process into reverse. The stock
market fell by 90 percent from its 1929 peak, a third of all
US banks failed, and the majority of international bond
issues defaulted. Public anger was inevitably directed at
bankers, or banksters, as they were now known.
The debate about the causes of the Great Depression
has raged ever since the early 1930s. At the time, most
people blamed the Wall Street Crash of 1929, which
seemed to mark a sharp transition between the fat
years of the 1920s and the lean years of the 1930s.
Since then, opinion has changed. Most economists now
believe that the crash need never have developed into the
depression. Some blame the banking crises of 193033
for transforming a normal business recession into the
worst depression in modern history. When they look at
the US banking industry, they identify the small local unit
banks as the fatal weakness that made the outcome so

much worse in America than it was in Canada or the UK,


with their well-established branch banks.
Similar views were held in the 1930s by the advocates
of the large banks, who hoped to use the crisis to break
down the barriers that prevented them from establishing
statewide if not nationwide branch networks. The
advocates of the small banks countered by arguing that
the unit banks that failed in large numbers were not the
cause of the problem, but its victims. Blame should, they
maintained, be laid at the door of the securities activities
of the large money-centre banks for blowing up the stockmarket bubble and setting off the crisis.
Other commentators have taken a different view. Many
economists now believe that the deflationary spiral was
triggered by failures in central banking practice that
allowed the money supply to shrink and permitted banks
to fail in the absence of a lender of last resort. Others
point to the collapse of trade as the world retreated into
protectionism in the wake of the disastrous Smoot
Hawley Tariff Act of 1930.

Separation
In the period immediately after the depression, the
question of what had caused it was central to enacting
reforms intended to prevent future crises.
In 1932, Carter Glass, the most influential member of
the Senate banking committee, introduced a bill to
separate commercial and investment banking. He and
his supporters reasoned that allowing the banks to enter
the securities business had created an overproduction
of securities5 that had inexorably led to the crash.
At the same time, the holding of volatile securities on
banks books had weakened their balance sheets and
contributed to a loss in confidence in the banking system.
Moreover, selling securities to their customers had given
rise to serious conflicts of interest.
The bill was opposed by the Hoover administration and
by the Federal Reserve of New York, on the grounds that
regulation was a sufficient solution to any problems that
had occurred and that separation would cause further
disruption to an already fragile financial system. Support
in Congress was mixed, and the bill might have died had it
not been for the confluence of three factors.
The first was the bills adoption by Franklin Roosevelt
in his presidential campaign. He declared, Investment
banking is a legitimate business. Commercial banking is
another, wholly separate business. Their consolidation
and mingling is contrary to public opinion.6 The second
factor was the Senate investigation into banking practice
led by Ferdinand Pecora in early 1933, which uncovered
a series of unsavoury insider deals, conflicts of interest,
and tax-avoidance tactics. The revelations outraged

Should commercial and investment banking be separated?


Part I
The rise of universal banking

the public and so embarrassed the heads of the two


largest New York banks that they closed down their
securities businesses. The third factor was an alliance
of convenience between Carter Glass in the Senate and
the chairman of the House Committee on Banking and
Currency, Henry Steagall. An advocate of the small unit
banks, Steagall sought to forestall any attempt to allow
nationwide branch banking, while establishing a national
deposit insurance system that would allow the unit banks
to compete with the larger banks.
The result of this alliance was the GlassSteagall Act
of 1933, which forced banks to close down or spin off
their securities businesses and established the Federal
Deposit Insurance Corporation (FDIC). In spite of critics
misgivings, the separation of banking activities was
achieved relatively simply. Banks had to choose whether
they wished to accept deposits or deal in securities; they
could no longer do both. By and large, the commercial
banks got out of the securities business, and the
investment banks stopped accepting deposits.
The most difficult decision was that faced by the private
banking partnership of J. P. Morgan & Company, which
was neither a national nor a state-chartered bank, had
never openly sought deposits, and was not the original
focus of the legislation. Even though its business
practices were largely beyond reproach, it was included
because the Pecora hearings shone an uncomfortable
light on the extraordinary reach of its business and
political influence, and on the negligible income tax
paid by its partners. In the end the partners decided to
become a commercial bank, while allowing a number of
their colleagues to set up a separate investment banking
business under the name of Morgan Stanley.
An objection voiced in the House of Representatives to
the GlassSteagall Act in 1933 had noted that:
The boom in 191920, ending in a crash that marked the
beginning of the period of high bank mortality, was one of
commodity prices. At its peak, brokers loans amounted
to only $1,750,000,000, as compared with about
$8,500,000,000 in 1929. Nothing in the bill will prevent a
recurrence of this situation.7
This was a prescient observation. Whatever the virtues
of the GlassSteagall Act, it did nothing to avert the
commodities booms and busts of 197374 and 197980.
Nor did it prevent the savings and loan (S&L) crisis of
the 1980s, in which a third of these small specialist
institutions failed.

Full circle
From 1933 to the late 1970s, the GlassSteagall Act
remained largely unchallenged. However, it is not hard
to see why the commercial banks started to push for its

repeal in the 1980s. Profits from traditional lending were


declining as creditworthy corporations funded themselves
in the commercial paper market. Meanwhile the investment
banks were expanding their scope and seeing their profits
soar. Access to investment banking activities would
provide commercial banks with sources of non-interest
income as an alternative to seeking capital-intensive
interest income from ever riskier sources (although some
of these banks did that too). Universal banking was
becoming the norm in an increasingly globalized financial
world, especially once the UK put an end to its tradition of
specialized financial institutions in the mid-1980s.
The US banks had a number of eminent supporters, most
notably Alan Greenspan, who argued strongly in favour of
deregulation. In addition, revisionist academic accounts
started to appear that attributed the destabilization of the
banking system in the 1930s not to securities activities,
but to the small local banks without securities businesses
that failed in their thousands while the large national banks
survived. Support for this analysis seemed to be provided
by the savings and loan crisis of the 1980s and 1990s. More
than 700 S&L associations failed, demonstrating once
again the frailty of a system based on single-branch banks.
The passage of the GlassSteagall Act took just over a
year; its repeal arguably took twenty. Bankers Trust made
the first inroad in 1978 by starting to sell commercial
paper. Despite being sued for breach of the Act by the
Securities Industry Association, it was eventually allowed
to set up an affiliate that was permitted to generate up
to 5 percent of its total revenues through underwriting.
In 1988 the Federal Reserve Board gave bank affiliates
permission to underwrite commercial paper, mortgagebacked securities, and municipal revenue bonds with
a limit of 10 percent of total revenues. In 1990 this
concession was extended to corporate bonds and
shares. In 1995 an attempt at legislative repeal failed,
but in 1996 the FRB expanded the acceptable level of
securities business to 25 percent of total revenues. By
1999, when the GrammLeachBliley Act repealed
the provisions of GlassSteagall, the return of universal
banking had become inevitable.
By 2008, the large US banks under the supervision of the
Federal Reserve were all universal banks of one kind or
another. Over the same period, former non-deposit-taking
broker-dealers under the supervision of the Securities
Exchange Commission (in particular Morgan Stanley,
Goldman Sachs, Merrill Lynch, Bear Stearns, and Lehman)
had expanded their balance sheets substantially to
become sizeable lenders funded through the securitization
of assets and the wholesale funding market. They were also
among the weakest parts of the system, as demonstrated
by the bail-out of Bear Stearns and the bankruptcy of
Lehman in 2008.

10

The US debate in the past couple of years about the


separation of commercial and investment banking has
been less concerned with conflicts of interest than was the
case in the 1930s. Nor has it regarded universal banking
per se as a risk to financial stability, because all types of
banks failed: pure investment banks, specialized retail
banks, and universal banks. Rather, those who argued for
separation were mainly concerned about the moral hazard
that would arise if banks were able to fund themselves
cheaply thanks to an implicit government guarantee and
then use those deposits to invest in risky assets.

Germany: A long tradition


A counterpoint to the United States can be found in
Germany, a country where universal banking has grown
up organically and seldom been challenged. As in most
other continental European countries and most other
parts of the world large banks have traditionally been
universal banks.

Funding industry
When Friedrich Krupp wanted to build his first factory
in 1811, he had to turn to his mother and siblings for a
loan. However, by the 1840s German private banks were
helping to finance business start-ups in exchange for
board representation. In the 1850s they were joined by
the first joint-stock banks, which were able to deploy the
greater amounts of capital needed for railway investment.
After unification in 1871, the pace of industrialization
accelerated dramatically. The boom of the early 1870s led
to the formation of 183 joint-stock banks, while the crash
of 1873 prompted successive waves of consolidation
that led to the rise of the big Berlin banks. Without any
restrictions on branching, the largest banks were able to
assume a dominant role in the national economy. By 1913
the three largest German companies were banks.
While the biggest German banks were larger in relation to
the size of the national economy than anything found in
America, they played no central banking function. In 1876
Germanys central bank, the Reichsbank, had been set
up with a monopoly on issuing notes. Unlike the Bank of
England, it had an extensive branch network that enabled
it to provide not just payments facilities but also shortterm business loans. As a result, the big banks tended to
concentrate on providing capital for industrial expansion.
In common with the United States, there was a tendency
for German companies to form monopolies or cartels so
as to avoid what was perceived as fruitless competition.
By the early twentieth century the German economy was
characterized by networks of dominant corporations
financed by a small group of large banks that exercised
influence through shareholdings and directorships.
For the most part these banks focused on their large

corporate clients, and it was not until the 1960s that


they entered retail banking. In the meantime, private
customers were served by the savings and loan and
cooperative sector.
The German equivalent of Americas Louis Brandeis
was the Marxist economist Rudolf Hilferding, who
published his Das Finanzkapital in 1910. He argued that
the concentration of business into cartels through bank
finance was the ultimate development of capitalism:
As capital itself at the highest stage of its development
becomes finance capital, so the magnate of capital, the
finance capitalist, increasingly concentrates his control
over the whole national capital by means of his domination
of bank capital.8
This analysis led to calls for curbs on the power of banks in
some quarters. From a socialist perspective, though, this
was a moot point, since the concentration of economic
power in the hands of a few capitalist magnates was
regarded as the result of the fatal inherent contradictions
of capitalism and perceived as leading naturally to the
concentration of economic power under the dictatorship
of the proletariat.

WWI and its aftermath


World War I not only put a stop to questions of banking
reform, it also transformed Germanys economic and
financial position. The hyperinflation that followed in its
wake drastically weakened the banks, reducing their
capital to less than a third of its pre-war level in real terms.
Ironically, one of the few assets that saved the banks from
total insolvency was their holding of industrial shares.
After the restoration of monetary stability in 1924, the
banks no longer resembled the powerhouses they had
been before the war. Their leverage was considerably
greater, and they, like the whole German economy, were
dependent on inflows of foreign capital.
The kind of universal banking conducted by the banks
required low leverage and a very stable deposit base so
that longer-term loans and industrial securities could
safely be held on the books. With their higher leverage and
dependence on foreign deposits, the banks were in no
position to withstand the banking crisis that started with
Credit-Anstalt in Vienna in May 1931 and soon spread to
Germany. The Darmstdter Bank declared bankruptcy,
and the state was forced to partly nationalize the other
big banks, acquiring 91 percent of the shares of Dresdner
Bank, 70 percent of Commerzbank, and 35 percent of
Deutsche Bank.
The authorities reaction to the crisis reflected the
traditional German suspicion of excessive competition.
They blamed the weakness of the banks on the risks
they took in the late 1920s to make up the capital lost as

Should commercial and investment banking be separated?


Part I
The rise of universal banking

a result of hyperinflation. The solution was the creation


of a government-controlled banking cartel with limits on
interest rates and restrictions on opening new branches.
The banks were reprivatized in 1936, but under the Nazi
regime they remained servants of the state.

Break-up and restoration


The arrival of the American occupation in 1945 brought a
new perspective. Cartels were regarded with suspicion,
and the closely linked German networks of companies
and banks were viewed as the economic backbone of
a pernicious nationalistic military machine that should
be reformed along decentralized democratic lines. The
three big Berlin banks were broken up into ten constituent
parts, one for each of the new Lnder in the federal
republic. This drastic cutting down to size of the big banks
along geographic lines may have explained why they were
not legally required to give up universal banking. In Japan,
by contrast, a version of the GlassSteagall Act was
imposed under the American occupation.
The advent of the Cold War soon necessitated the
rebuilding of West Germany as an effective industrial
power, while thoughts of remodelling its economy on
American lines receded. The ten subdivisions of the
big banks were restored to three in 1952, and then in
1957 they were allowed to reconstitute themselves as
nationwide universal banks. Although not as dominant
as they had been before 1914, they still retained the old
practices of shareholdings and interlocking directorships.
By the 1970s they were being criticized by left and right
alike: the left because of excessive concentration of
capitalist power, the right because of the inhibition of
free-market competition. In 1975 the Social Democratic
Party published a programme calling for the abolition
of universal banking and greater government control
of credit allocation. Meanwhile the right called for bank
shareholdings in non-financial corporations to be limited
to 5 percent.
The Gessler Commission was set up in 1974 to investigate
the banking system in the light of such criticisms. After
extensive delays it eventually produced its report in 1979,
concluding that
The universal banking system has proved its worth.
. . . deficiencies of the current banking system are not
sufficient to necessitate a change of system. . . . A
transition to a system based on separation might be
able to eliminate the kinds of conflict of interest which
exist within the universal banking system. However,
major structural change of this nature would have such
detrimental effects that it can ultimately not be justified.9
The only reform proposed was a limit on shareholdings
in non-financial companies of 25 percent. A sceptical
press suggested that the commission had been merely

11

a stonewalling exercise a suggestion seemingly


supported by the fact that even its modest proposals
were not put into effect.
Discussions about the influence banks exerted on
corporations through minority shareholdings and
directorships continued through the 1980s and 1990s.
By the late 1990s most banks had started to divest their
corporate shareholdings and reduce their directorships,
partly in response to pressure from their investors and
partly so that they could boost their capital with the gains
from divestitures. Today German banks no longer hold
significant corporate shareholdings, and the number of
their directorships continues to decline.
Not even the 2008 financial crisis succeeded in igniting a
debate in Germany about the separation of commercial
and investment banking. With the exception of
Commerzbank, the major German bank casualties (Hypo
Real Estate, WestLB, BayernLB, HSH Nordbank, and IKB)
had little or no retail banking operations or retail deposits.
Their highly leveraged investments in securitized and
other financial instruments were funded not by deposits
but by short-term commercial paper and the interbank
funding market.

The United Kingdom:


First separation, then integration
The defining event in the early history of British banking
was the founding of the Bank of England in 1694. In
exchange for providing finance for the government, the
bank was granted a monopoly in joint-stock banking
throughout the country. Moreover, after 1708 no other
bank was allowed to have more than six partners if it
wanted to issue notes. This meant that the Bank of
England gradually assumed a dominant position in the
economy, while other banks were underdeveloped.

Power at the centre


Until the nineteenth century, the UKs only financial
institutions were the Bank of England, with a solitary
branch in the capital; small country banks with fewer
than six partners that offered banking facilities to the
areas of the country beyond its reach and issued notes;
and London-based merchant banks that focused
primarily on trade finance and the placement of
government bonds.
Because of their small size the country banks were
inherently fragile, and there were repeated banking crises
in the late eighteenth and early nineteenth centuries,
leading to a very severe crisis in 1825 during which,
according to one close observer, the country was within
twenty-four hours of barter.10 To stave off the crisis, the
Bank of England undertook for the first time what would

12

now be understood as the role of lender of last resort.


As its governor Jeremiah Harman stated, We made
advances to an immense amount and we were not on
some occasions over-nice. Seeing the dreadful state in
which the public was, we rendered every assistance in
our power.11
In the wake of the crisis, joint-stock banking was allowed
so that banks could raise more capital. However, the
shareholders of the new joint-stock banks still had
unlimited liability, and although this was supposed to
make their management more prudent and provide better
security for depositors, it ultimately reduced the amount
of capital that they could raise. It was only after a further
severe banking crisis in 1857 that a law was introduced to
allow limited-liability banking for new banks. Generalized
limited liability for all joint-stock banks was established
only in 1878 after the failure of the unlimited-liability City of
Glasgow Bank.
As The Economist commented, the risks of unlimited
liability within a corporate structure was driving wealthy
investors away so that an almost incredible number of
bank shareholders were spinsters and widows, . . .
clergymen . . . and others whose occupations do not
appear to have enabled them to have accumulated much
wealth. The paper concluded that the limited liability
of the wealthy may be expected to prove as good if not
better security than the unlimited liability of the poor.12

Consolidation
The advent of limited-liability banking was followed by
a wave of consolidation, so that by the early twentieth
century Britain was dominated by a small number
of nationwide banks. However, unlike their German
counterparts, the large British banks confined themselves
to commercial banking even though there was no law
requiring them to do so.
The likely explanation for this division of labour is that Britain
had had plenty of time to develop efficient capital markets
with specialist investment banks, so there was no need for
commercial banks to get involved in securities activities. At
the same time, because of the relatively late development
of limited-liability banking, an increasingly wealthy society
was able to provide more than enough profitable business
for retail banking. By comparison, Germany came much
later to the Industrial Revolution and found it needed a lot
of capital to catch up with Britain. Since its capital markets
were undeveloped, it needed universal banks.
By the First World War, some qualms were emerging at
the excessive concentration of banking in Britain, which
now had the worlds biggest banks. In 1918 the Colwyn
Committee recommended that any further consolidation
be avoided. However, the issue did not excite the passion
aroused in America or Germany, most likely because the

absence of universal banking meant that British banks had


never exercised the control over industry that the American
and German banks were accused of maintaining.
Once it had consolidated into a system of big banks
with nationwide branches, the British banking system
became impressively stable. Moreover, its focus on
short-term self-liquidating business loans allowed it to
operate with leverage of 10:1 in 1913 (compared with 4:1
in the United States and 3:1 in Germany) without undue
risk. Unlike their American and German counterparts,
the British banks emerged from the crisis of the early
1930s virtually unscathed, a point that was not lost on
American lawmakers.

Deregulation and expansion


The separation of investment and commercial banking in
Britain had always been a matter of convention rather than
law. During the 1960s and 1970s the clearing banks started
to provide a wider variety of loans than before, moving into
consumer finance, mortgages, and medium-term business
loans. They also made their first steps into investment
banking when Midland bought a 25 percent equity stake
in Samuel Montagu and National Westminster set up a
merchant-banking subsidiary. The rise of the Eurodollar
market in London heralded the arrival of numerous foreign
banks and introduced the practice of longer-term loans
funded on a revolving basis.
By the 1980s the biggest barrier to the creation of fully
integrated banks was posed by the internal rules of the
London Stock Exchange. These required members to
operate as partnerships specializing either as brokers or
as market makers, and prevented outsiders from owning
a significant financial interest in member firms. It was the
breaking down of these rules in response to a government
investigation into restrictive practices and price fixing that
opened the door to fully integrated universal banking.
By the eve of what would become known as Big Bang
on 27 October 1986, the four big clearing banks had
positioned themselves to become fully integrated banks,
and had between them invested close to 1 billion in
securities businesses at a time when the capital of the
average stockbroker or merchant bank could be measured
in tens of millions. These figures were a foretaste of the
massively increased scale on which globalized universal
banking was to operate in the coming years.
In the years up to the financial crisis of 2008, British banks
enhanced their standing among banks internationally.
Commercial banking was a highly profitable business
in the UK, and from time to time it gave rise to concerns
about the level of competition, especially in retail banking.
HSBC and Standard Chartered Bank continued to expand
internationally in line with their roots in emerging-market
banking. Barclays successfully built an investment bank,

Should commercial and investment banking be separated?


Part I
The rise of universal banking

and RBS became one of the largest banks in the world


through a series of acquisitions and rapid expansion into
leveraged lending to corporations and private equity
firms. By the time of the financial crisis all of them had
become universal banks, albeit with very different mixes
of commercial and investment banking activities. While
Barclays, HSBC, and Standard Chartered weathered the
crisis without government support, RBS, Lloyds (largely as
a result of its purchase of HBOS), and smaller banks relying
on the securitization market for funding (most notably
Northern Rock) needed substantial government funds and
guarantees. The UK government had to inject billions in
capital into the industry, and the Bank of England (and the
European Central Bank) had to provide significant funding
to a number of these banks to keep the industry afloat. The
sheer size of the banks, and the resultant bail-out costs for
the UK, raised real concerns.
The post-crisis debate on the merits of separating
commercial and investment banking has been more
animated in the UK than elsewhere. While moral hazard
is seen as the central issue, as it is in the US, those who
advocate separation also argue that it could make a
material contribution to the stability of the financial system
if implemented in conjunction with other measures such as
substantially higher capital requirements.
***
The idea of separating commercial and investment banking
by law was not seriously considered by either the UK or
Germany for most of their history. In the UK, a system of
large commercial banks developed gradually, and by the
time these institutions were large enough to enter into
investment banking, that slot was already occupied by
specialized firms organized as partnerships. Before the Big
Bang in the 1980s the rules of the London Stock Exchange
played a role in maintaining this structure, although they
concerned only a subset of investment banking activities.
Germanys system was one in which the large banks
defined themselves primarily as banks for bigger
companies, a role that naturally included related
commercial and investment banking activities. These large
banks were late entrants into retail banking and even today
play only a minor part in a market segment dominated by
savings banks and cooperative banks.
In the US, two key factors shaped the history of banking: a
deep suspicion of financial power and a political preference
for small local banks.
Both now and in the past, many different arguments have
been put forward for the separation of commercial and
investment banking. Part II revisits todays arguments in
light of the historical experience.

13

Should commercial and investment banking be separated?


Part II The historical background to the arguments used in todays debate

Part II The historical background to


the arguments used in todays debate
This section sets some of the main arguments being aired
in the current debate about universal banking in their
historical context. Not all the questions being discussed
today have long histories. Conversely, some arguments
that have been important in the past are playing little part
in the current debate. However, an understanding of how
key issues have been regarded at different points in time
and in different countries will provide food for thought and
help put todays arguments in a broader perspective.

1 The advantages and drawbacks of


universal banks
Are universal banks inherently less stable than
specialist commercial banks?
One argument for separation advanced today is
that having non-loan assets on bank balance sheets
is generally more risky than having only commercial
bank assets.
This argument was used in the United States in the early
1930s. For instance, Senator Bulkley, one of the major
backers of the GlassSteagall Act, noted that:
The English banks of deposit have kept themselves
strictly clear of the investment-security business,
while the German banks, on the other hand, have not
hesitated to make substantial investment of their own
funds in promotions and refinancing with a view to
public distribution at such time as might be convenient.
In banking literature there are arguments both ways. It
seems, however, that the English banking situation has
been maintained in a more satisfactory manner than the
German, and [this] should lead us to prefer the English
practice, under which commercial banking is strictly
segregated from the origination and underwriting of
capital issues.13

valued according to estimates of ultimate losses. In many


cases banks were forced to dump their bonds on the
market in their search for liquidity, taking heavy losses that
contributed to their insolvency. Had there been a more
effective central banking response to provide liquidity, the
problem of forced selling might have been averted, but
there is little doubt that the move away from short-term
business loans had weakened banks balance sheets.
Moreover, it can be argued that securities activities
could pose a further risk to stability by damaging a
banks reputation through contagion. If a bank sells a
security that falls in value, public perception of the bank
is likely to be affected. This point was made in a study of
bank security affiliates by William Nelson Peach in the
late 1930s:
There are certain dangers arising from these
relationships and it is not possible to avoid these by
legislation. If affiliates sell securities in the name of the
parent bank, then the good will of the affiliate and the
parent bank rise and fall together.14
A similar point was made in 1986 by the governor of the
Bank of England, worried about some of the unintended
consequences of Big Bang:
Banks may be vulnerable because they have built up
large exposures to securities businesses. . . .Those with
subsidiaries engaged in securities business will also feel
a practical obligation . . . to their securities subsidiaries
far in excess of the amounts of the facilities granted. Such
consideration may mean that a bank within a financial
conglomerate may be particularly exposed to contagion
and a loss of confidence.15

What was the background to these remarks? The first


dent in the provisions of the 1864 National Bank Act came
when banks were allowed to set up bond departments.
During the 1920s banks had sharply increased their bond
holdings, and their portfolios started to include a higher
percentage of lower-grade corporate and municipal
bonds. During the early 1930s there was a flight to quality,
and spreads on BAA-rated bonds widened by over 300
basis points.

Evidence for the argument against separation can also


be drawn from the experience of the United States in the
1930s. The banks that failed in their droves were not in
fact the large national banks with their securities affiliates,
but the small unit banks. Of the 9,000 banks that failed
between 1930 and 1933, very few had securities affiliates.
When asked by Senator Glass about the causes of bank
failures, the Comptroller of the Currency replied that 90
percent of the banks were in small rural communities,
and that he knew of no instance where the shrinkage in
value of collateral or bank investments as far as national
banks are concerned, has been responsible for any bank
failure, or very, very few of them.16

Holdings of marketable bonds threatened banks


stability because they were forced to mark these bonds
to market, whereas traditional loan portfolios could be

Not surprisingly, perhaps, the big banks argued that the


cause of the crisis lay in state laws that prohibited branch
banking, and in the fragmentation of the regulatory

15

16

system. Thomas Lamont of J. P. Morgan argued for two


vital changes:
The first is to bring all the commercial banks, small as well
as large, under the single aegis of the Federal Reserve
System. The second is to establish sensible provisions for
regional branch-banking. . . . Such reforms . . . ought to
bring the country some measure of banking stability.17
Similar arguments were put forward by the Comptroller of
the Currency.
Yet thanks to the entrenched power base of the small
banks in Congress, with Henry Steagall as chairman of
the House Banking Committee, the small banks emerged
from the crisis with their position enhanced and with
deposit insurance long desired as a way to compete
with the big banks enshrined in law. It took the savings
and loan crisis of the 1980s, in which a third of S&Ls
failed, to demonstrate yet again the vulnerability of a
system based on unit banks. In addition, the crisis finally
broke down the barriers that prevented cross-regional
and national branch banking and took the heat out of the
argument about the risks of integrated universal banking,
paving the way for the eventual repeal of GlassSteagall.

Are big universal banks better for the economy?


One argument against separation is that a
sophisticated global economy needs large integrated
banks that are able to offer a full range of banking
services. It follows that splitting up banks would harm
economic growth.
This argument was first advanced in the debates of
the 1930s. As Alan Pope, a member of the Investment
Bankers Association, testified to the Senate: I do not
believe that this country could have developed industrially
to the extent that it has since the war without the
assistance of bank [security] affiliates.18
A number of studies in the lead-up to the repeal of the
GlassSteagall Act suggested reasons why universal
banks foster economic growth. These studies argued
that such banks take a longer-term interest in the
companies that they finance, and, because they are often
shareholders, there are fewer conflicts of interest between
borrower and lender.
It has also been argued that universal banks can offer
economies of scope that enable them to provide
investment banking services more cheaply than specialist

firms. A study by economist Charles Calomiris of the


underwriting fees charged by German banks before 1914
showed that they averaged 3 to 5 percent, compared with
up to 20 percent charged by American investment banks.
In other words, not only did universal banking bring gains,
but those gains accrued to customers.19
On the other hand, numerous studies have suggested
that the economies of scale in banking occur at a size well
below that actually existing in the market.

Does universal banking create overmighty


subjects?
One argument for separation aired in the current public
debate is that banks have too strong a lobbying influence
and have become too powerful the old Wall Street
versus Main Street tension.
This issue has a clear precedent in the concerns about
a money trust in the period before the First World War.
The first lines of Louis Brandeiss Other Peoples Money
quoted president-to-be Woodrow Wilson in 1911:
The great monopoly in this country is the money
monopoly. So long as that exists, our old variety and
freedom and individual energy of development are out of
the question. A great industrial nation is controlled by its
system of credit. Our system of credit is concentrated.
The growth of the nation, therefore, and all our activities
are in the hands of a few men, who, even if their actions
be honest and intended for the public interest, are
necessarily concentrated upon the great undertakings in
which their own money is involved and who, necessarily,
by every reason of their own limitations, chill and check
and destroy genuine economic freedom. This is the
greatest question of all; and to this, statesmen must
address themselves with an earnest determination to
serve the long future and the true liberties of men.20
In 1933, Ferdinand Pecora took up the same theme when
he described the 126 directorships held by the partners
of J. P. Morgan in 89 of the largest US companies as
incomparably the greatest reach of power in private
hands in our entire history.21
Reading the introduction to Simon Johnson and James
Kwaks recent book 13 Bankers evokes a sense of dj vu:
The Wall Street banks are the new American oligarchy
a group that gains political power because of its
economic power, and then uses that political power for

Should commercial and investment banking be separated?


Part II The historical background to the arguments used in todays debate

its own benefit. . . .This is not the first time that a powerful
economic elite has risen to political prominence. In the late
nineteenth century, the giant industrial trusts many of
them financed by banker and industrialist J. P. Morgan
dominated the US economy with the support of their allies
in Washington, until President Theodore Roosevelt first
used the antitrust laws to break them up.22
The counter-argument is that claims regarding the
accumulation of power by banks are exaggerated, at least
nowadays, when banks seldom exercise influence through
shareholdings or directorships in a systematic way.

Does universal banking lead to dangerous


conflicts of interest?
One argument for separation is that universal banks are
more likely to experience conflicts of interest.
The conflicts of interest that were exposed by the Pecora
hearings in the US Senate in 1933 were one of the
main reasons for the passage of the GlassSteagall Act.
Banks had sold securities to their customers without
disclosing their own interests in the transaction, as
when National City sold its entire holdings in Anaconda
Copper to its customers as soon as the price of copper
dropped, while continuing to recommend the stock as a
sound investment.
Conflicts of interest undermined confidence in banks in
general, a factor that was felt to have contributed to the
panics of the early 1930s. Senator George Jones argued
that confidence in the system would return if banks were
prohibited from carrying on that sort of business which
puts them on the opposite side of transactions from their
own customers.23 Similarly, when the Comptroller of
the Currency allowed banks to sponsor mutual funds in
1966, his decision was overruled by the Supreme Court
on the grounds that it infringed the terms of the 1933 Act:
not so much because it posed a risk to bank solvency, but
because it might undermine public confidence.
This issue was also addressed by the Gessler
Commission in Germany in the 1970s, which identified
several situations in which universal banks could find
themselves in a conflict of interest. These included
the temptation to sell securities to repay loans on their
own books; the possible threat of credit rationing to
sell products and services; and the temptation to place
unsold securities in trust accounts that they themselves
ran. The commission concluded, however, that although
potential conflicts existed, a bank . . . .cannot exploit

them without ruining its reputation in the long run.24


Much of the German press regarded this conclusion as
rather lame.
More recent studies, such as George Benstons 1990
book The Separation of Commercial and Investment
Banking, have concluded that the abuses identified by the
Pecora hearings were overstated, and that most of them
were not committed by banks.

Do universal banks have a competitive edge?


One argument against separation is that universal
banks have a competitive edge that enables them to
provide real benefits to the economy.
Bankers seem to believe in these benefits, and politicians
worry that they might put their country at a disadvantage
if they were to adopt a unilateral policy of splitting up
banks. In the early twentieth century, American banks
came under pressure to circumvent the restrictions
of the National Bank Act because of competition from
unregulated trust companies able to offer a wider variety
of services. As a study put it in 1909:
It is a distinct convenience to most people to have all
of their financial business attended to under one roof.
The trust company will not only care for their banking
business, but will also receive valuables for safekeeping,
care for property, manage their estates temporarily or
permanently for them, make investments for them, [and]
give financial and legal advice.25
Another concern was that commercial banks traditional
business was in decline because of the development
of the securities markets, which enabled companies to
bypass the banks and fund themselves directly. A study
in 1931 observed that:
if economic progress continues to be associated with the
growing importance of larger companies having access
to stock and bond markets, there is a strong probability
that the commercial loan will decline relatively to other
bank assets.26
The same trends were behind the pressures for American
banks to overturn the GlassSteagall Act in the 1980s.
More and more companies were able to finance
themselves in the commercial paper market, leaving
banks to choose between enduring shrinking profits or
raising their levels of risk. It was hardly surprising that
they sought non-interest sources of revenue, especially

17

18

when competing in an international economy against


universal banks that had none of their restrictions. As Alan
Greenspan put it in 1990:

what is then commonly pledged and easily convertible


the alarm of the solvent merchants and bankers will be
stayed.29

In an environment of global competition, rapid financial


innovation, and technological change, bankers
understandably feel that the old portfolio and affiliate rules
and the constraints on permissible activities of affiliates
are no longer meaningful and likely to result in a shrinking
banking system.27

In the United States, there was almost no history of


a lender of last resort in the nineteenth century, for the
simple reason that the country operated without a central
bank for most of that period. Instead, it was the United
States that first attempted to develop systems of deposit
insurance, precisely because this suited the small local
banks that so disliked the idea of a dominant central bank.

The counter-argument to this view of the benefits of


integration was that many of them could be replicated
by having customers carefully select from a range of
providers, and that, as noted earlier, any cost
benefits occur at a scale well below that of the large
integrated banks.

2 Moral hazard
Does underwriting the banking system encourage
undue risk taking?
One argument for separation is that it is not appropriate
for universal banks to make risky bets that are funded at
least partly by insured deposits and implicit government
guarantees. This is the central moral hazard argument. If
you believe that someone will bail you out, you can take
high risks because you are protected from failure.
The moral hazard argument consists of two strands: the
underwriting of the banking system through a lender of
last resort, deposit insurance schemes, or both; and the
undue risk taking that may result from it.
The history of the lender of last resort goes back to the
banking panic in England in 1825, when the Bank of
England had to provide liquidity, as its governor said, by
every possible means and in modes that we had never
adopted before.28 However, the thrusting of the bank
into this new role did not prevent further crises, which
recurred with monotonous regularity every decade until
1866, when the bank had to use all available means to
offset the dash for liquidity. In the wake of this event,
Walter Bagehot, editor of The Economist, set out the
principles underlying the role of lender of last resort in his
book Lombard Street: A description of the money market.
According to Bagehot, the bank should declare itself
ready to lend freely on all acceptable forms of security
during any future panic:
If it is known that the Bank of England is freely advancing
on what in ordinary times is reckoned a good security on

The first insurance scheme was set up in New York State


in 1829. Because it was voluntary, it ran into the problem
of adverse selection, and when the panic of 1837 arrived,
it quickly used up all its reserves and was forced to refund
smaller and smaller percentages of the deposits it was
supposed to protect.
Towards the end of the century some states, such as
Oklahoma, set up compulsory schemes, attracting
protest from the sounder banks, which felt they would
end up paying for the fecklessness of their unsound
competitors. Yet even these schemes ran into trouble
because of inadequate supervision of member banks,
some of which chose to speculate with the deposits that
were now easier to raise.
In the 1920s state insurance schemes suffered a
succession of failures, fuelling opposition to Henry
Steagalls attempts to get compulsory nationwide
insurance grafted on to Carter Glasss bill for the
separation of commercial and investment banking.
Objections came not just from the big banks but from the
Comptroller of the Currency (who argued that it would
put a premium on incompetency and irresponsibility),30
from the government, from the Federal Reserve, and from
Glass himself. However, political horse trading prevailed,
and deposit insurance became part of the reform
package of 1933.
Concerns about undue risk taking occurring as a result
of an implicit government guarantee or deposit insurance
scheme also have a long history.
Bagehots views on the lender of last resort sparked a
debate that has resounded ever since. If the banking
system knows it will be supported in a crisis, is it bound
to take ever greater risks? Bagehots opponent in this
debate was a former governor of the Bank of England
called Thomson Hankey, who argued that competitive
pressures from those banks that took risks in the
knowledge that they would be supported in a crisis would

Should commercial and investment banking be separated?


Part II The historical background to the arguments used in todays debate

force all banks down the same path. Until Bagehots


doctrine was repudiated, he said, the difficulty of
pursuing any sound principle of banking in London will
always be very great.31 Bagehot replied that moral hazard
would not occur because the Bank of England would
charge a high rate of interest that would discourage banks
from using its facilities.
After the 1866 crisis, the British banking system
stabilized and there were no further general panics until
the First World War. Some see this as a vindication of
Bagehots doctrine; others see a different tale with a
different moral.
Between the commercial banks and the Bank of England,
there was a tier of specialist banking partnerships
referred to as discount houses, which provided
liquidity for local banks by discounting their bills. If the
discount houses needed liquidity themselves, they could
rediscount the bills at the Bank of England. Until 1857
the bank was obliged to provide an almost unlimited
window for such discounts, and it argued that the crisis
of that year had been caused in part by the refusal of the
discount houses to keep adequate reserves on hand.
In 1858, therefore, the bank stated that it would make
money available only according to normal seasonal
needs, and that it was up to the discount houses to
maintain their own emergency reserves.
This prompted Overend, Gurney & Company, the largest
of the discount houses, to organize a mini-run on the Bank
of England by way of protest. When Overend, Gurney
itself got into trouble in 1866, the bank deliberately let it fail
before providing liquidity to stave off the ensuing general
panic. It was this salutary lesson that brought stability in
its wake.
Concerns about undue risk taking had also been voiced
by a commentator who opposed the introduction of
limited-liability banking in England in response to the
panic of 1857:
America . . . originated the panic that is sweeping over
the entire mercantile world. . . .She has added to its
virulence by her limited liability and credit system. . . . I
ventured the remark two years ago, that in the United
States, banks paid a better rate of profit than trading and
mercantile companies. Banking, some of us think, ought
to be a safe, prudent calling, and not a high rate of profitpaying business.32
These concerns were also central in the 1930s
Congressional debates in the US. Senator Bulkley argued

that the move of the commercial banks into the securities


business had been motivated by an element of greed33
and that the resulting excessive competition for profits
had contributed to the stock market bubble:
Can there be any doubt that under such pressure
of competition there was an overproduction of
capital securities? . . . Is it not fair to attribute the vast
development of loans on collateral security at least in
part to the necessity for keeping up a market condition
capable of absorbing capital issues?34
Some bankers sympathized with the sentiment that
seeking to maximize profits was dangerous and
inappropriate in a business that should put safety first.
A. P. Frierson of the East Tennessee National Bank stated
in Congress that:
We are operating with other peoples money, and I do
not think that we ought to be permitted to amass huge
earnings to speculate with in any way, or to pay huge
dividends through the earnings of banks.35
An argument against separation is that the moral
hazard issue has little to do with universal banks as
such. It has much more to do with large banks in the
case of implicit government guarantees, and banks of all
sizes and business models where deposit insurance is
concerned. There is no clear historical evidence that large
banks or banks benefiting from deposit insurance engage
more in undue risk taking.

Under what conditions do banks become too big


to fail, and threaten to become too big to bail?
Whether the failure of a particular bank would constitute a
threat to the whole financial system is virtually impossible
to ascertain unless you actually let it fail. The question of
whether a banks size determines its systemic relevance is
a controversial one.
The term too big to fail goes back to the rescue of
Continental Illinois from imminent insolvency in the US in
1984. However, this was not the first time a government
had bailed out a bank that was important to the whole
banking system.
During the European banking crisis of 1931, Austria
was forced to take over Credit-Anstalt, its largest bank,
for fear of what would happen to the national economy if
it were allowed to fail. Germany followed suit by injecting
capital into its three biggest banks, Dresdner Bank,
Commerzbank and Deutsche Bank. A similar chain of

19

20

events occurred in Italy, where the three largest banks were


nationalized in 1933.
None of these banks seemed as yet to be too big to bail,
but Credit-Anstalt came close. It had become the largest
bank in Austria as a result of being forced to take over a
number of failing institutions in the 1920s. When it faced
collapse, the government had to guarantee 1.2 billion
schillings of liabilities at a time when the countrys total
budget was 1.8 billion schillings. In a way, Austria was
lucky that Credit-Anstalt was perceived internationally to
be systemically important, because foreign creditors felt
obliged to accept some of the costs of recapitalization.
Eventual losses of which two-thirds were borne by
the Austrian government amounted to over 1 billion
schillings, against a pre-crisis capital of only 165 million.
The banking crisis of the 1930s seems not to have
prompted a debate in any of the countries affected about
the financial dangers of allowing banks to become too big
to fail. Given the political and economic turmoil of the next
fifteen years, perhaps this is not surprising. It also reflects
the fact that in Austria and Italy, the big banks remained
nationalized for decades, while in Germany they were split
up after the war for political reasons.
In America, it was the small banks, not the large, that
failed during the 1930s. The exception was the Bank of
United States, the twenty-eighth biggest in the country. Its
failure in October 1930 is blamed for the rapid escalation
of a nascent banking crisis into a full-blown panic. In
the face of mounting bank failures, the Reconstruction
Finance Corporation was established in January 1932 to
help prop up the system. By 1935 it had injected capital
into over 6,000 banks, mostly small ones.
When the authorities looked at the impending insolvency
of Continental Illinois in May 1984, they may have had
the consequences of the collapse of the Bank of United
States in mind. At that time Continental Illinois was the
seventh-largest bank in the country. Rather than let it fail,
the FDIC injected $4.5 billion and took over 80 percent
of the shares. Continental had funded its headlong
expansion almost entirely in the wholesale market, with
the result that federally insured deposits amounted to only
10 percent of its total liabilities. Yet the FDIC made sure
that all depositors and bondholders were repaid in full.
In the wake of the crisis, the Comptroller of the Currency
suggested that there were eleven banks in the country
that were too big to be allowed to fail. Perhaps he hoped
to make depositors at the remaining banks understand

that they would not be entitled to such privileged


treatment. However, it seemed that the nominal terms
of FDIC insurance were becoming moot in any case.
During the S&L crisis some years later, 99.7 percent of
all depositors were repaid in full even though this meant
repaying deposits that were not insured, in banks that
were too small to be a systemic risk.
The most plausible reason, other than regulators
natural reluctance to be associated with painful losses
to depositors, is the fear that the S&Ls, although small
individually, posed a collective threat to the financial
stability of the banking system. If that is the case, then the
question of too big to fail may need to be viewed in the
light of the banking industry as a whole, and not just its
largest institutions.

3 The economic cost of splitting up the banks


Would splitting up the banks put economic
recovery at risk?
One argument against separation is its possible impact
on economic recovery, although this seems to feature as
a relatively minor issue in the current debate.
Recent discussions on the unintended consequences of
regulatory reform on economic recovery have focused
primarily on the impact of higher capital and stricter
liquidity requirements.
In 193233 many voices argued against the Glass
Steagall Act on the grounds that it would damage the
already fragile financial situation. The Federal Reserve
Bank of New York argued that forcing banks out of the
securities business was unwise. American business
derived two or three times as much money from capital
markets as it did from commercial bank lending, and the
proposed law would disturb the mechanism of the capital
market, the free functioning of which is now so important
to a recovery from existing business conditions.36 The
banking community and others argued that the law would
be deflationary. As a newspaper editorial entered into the
congressional record put it:
. . . just when we are emerging from an atmosphere of
hysteria and fear . . . would hardly seem to be a propitious
time for enacting new and far-reaching provisions which in
their very nature are excessively deflationary. 37
As it turned out, the separation of commercial and
investment banking proceeded fairly smoothly. That
was least partly because banks operated their securities

Should commercial and investment banking be separated?


Part II The historical background to the arguments used in todays debate

underwriting business through separate companies.


Some banks including the two biggest in the country,
Chase National Bank and National City Bank had got
out of the securities business even before the legislation
was signed into law. The rest either closed their affiliates
or spun them off. The process was complete by 1935. In
the meantime, the recovery, which had got under way in
mid-1933, continued to accelerate. The economy grew by
9 percent in 1934 and 10 percent in 1935.
On the other hand, the amount of capital available for
investment banking shrank dramatically. In 1929 the
capital of the securities affiliates of National City and
Chase National alone amounted to over $220 million.
Ten years later the total capital of the eight largest
investment banks was a mere $75 million. At first sight this
would suggest that the GlassSteagall Act had adverse
consequences for the economy.
However, much of the capital available in 1929 had
disappeared even before the passage of the Act, mostly
through losses incurred in the bear market. In any case,
the demand for securities placement was minimal in the
1930s, even when growth resumed, because industry
was sitting on so much spare capacity. Moreover, one of
the arguments of the proponents of the GlassSteagall
Act was that the intrusion of so much commercial bank
capital into the investment banking business in the
late 1920s had been surplus to the economys actual
requirements, and had only led to the overdevelopment
of the capital market, which has brought upon us such
disastrous consequences.38

4 Regulation or separation
Should banks be split up, or simply
better regulated?
One argument against separation is that better
supervision is preferable to drastic structural measures.
However, debate rages over whether it is possible to
achieve truly effective supervision.
The question of supervision versus separation was
discussed as frequently in the 1930s as it is today. The
Federal Reserve of New York stated that:
the broad question to be determined is the extent to
which the capital market should be divorced from the
banking system and removed from all supervision, or
whether its relations with the banking system should be
maintained and placed under appropriate supervision.39

The New York Fed came down on the side of regulation as


an adequate and less damaging alternative to separation.
However, Eugene Meyer, chairman of the Federal Reserve
Board in Washington, was less certain. He was not
opposed to separation in principle, and argued that the
issue was sufficiently complex that it should be deferred
for three years so that the possible consequences could
be studied.
The regulation of investment and commercial banking
was in any case part of the response to the crisis.
The Securities Acts of 1933 and 1934 set out rules of
disclosure for securities offerings and established the
Securities Exchange Commission to enforce them.
The Federal Reserve was given the power to set margin
requirements. Commercial banking regulation was
extended by the creation of the FDIC, which brought all
banks under federal supervision for the first time.
The current regulatory alternative to splitting up banks
is the Basel accords. From their inception in the 1980s,
these focused on setting capital requirements as a way of
dealing with the growing inherent risk and volatility of the
assets held by banks as they move away from traditional
commercial banking.
Earlier regulatory regimes had mostly tended to focus on
liquidity rather than capital, as seen in the reserve ratios
established by the National Bank Act of 1864, and after
1913 by the Federal Reserve. The nineteenth-century
debate about capital adequacy in the United States,
and even more in Britain, focused on the question of the
liability of shareholders beyond their nominal investments.
Even though America was quick to adopt limited liability
as the standard form of incorporation, it maintained a
special double-liability regime for bank shareholders
until the 1930s. Banks were seen as intrinsically different
from other corporations because of the money deposited
with them by the public a sentiment foreshadowed
by the nineteenth-century debate in Britain about the
moral hazard of limited-liability banking. Even when
limited liability was definitively established in 1879, bank
shareholders continued to have full liability for any bank
notes still in circulation, and might face specified further
calls on capital in the event of liquidation.

21

22

Radical alternatives
The economists Laurence Kotlikoff and Christophe Charnley
proposed in 2009 that the banking industry should be transformed
into limited-purpose banking.40 In a reversal of the GlassSteagall
Act, which forced banks to abandon investment banking and
revert to traditional commercial banking, Kotlikoffs proposal would
effectively turn all banks into investment banks. Their function would
be to create and administer mutual funds in order to provide finance
for the economy in a variety of ways from which the investor would be
at liberty to choose. This proposal is reminiscent of Lowell Bryans
suggestion in his 1988 book Breaking Up the Bank41 that structured
securitized credit was the future of banking, and would provide
cheaper and more efficient financing to the economy with less
systemic risk.
The common theme in Kotlikoff and Bryan is the transformation
of banks from deposit takers into unleveraged intermediaries. As
a result, private investors would assume all the inherent risks and
rewards in lending and the systemic threat to the financial system
posed by the fragility of highly leveraged banks would no longer exist.
This idea has a partial ancestor in Irving Fishers 1935 book 100%
Money. Like Bryan and Kotlikoff, Fisher was worried about the risk
inherent in the fractional deposit system, which increases deposits
as money is re-lent by the banks. He had two objections. First, the
system posed the risk of banking panics because of the inevitable
mismatch in maturity between banks demand deposits and their
loan portfolios. Second, it amplified booms and busts by expanding
the broad money supply (namely cash and demand deposits) in
good times as deposits were created, and reducing it in bad times as
deposits were withdrawn.
Fishers solution was to have all demand deposits backed 100
percent by money, a step that would be achieved through a form of
quantitative easing. The central bank would print money up to an
amount that represented the reasonable requirements of the broad
money supply, which he estimated at one-third of GDP. It would then
buy assets such as government securities from the banks until they
held sufficient money to back all their demand deposits with cash.
A useful by-product of Fishers system would be that governments
would be able to borrow up to 33 percent of GDP free of interest.
Deposit insurance, which Fisher blamed for increasing risk, would no
longer be necessary.
Because demand deposits would be backed by cash, only savings
deposits would be available for making loans. However, since the
governments borrowing needs would be reduced, the amount of
money available for private-sector lending would be proportionately
increased.
Needless to say, Fishers plan was never put into effect.

Should commercial and investment banking be separated?


Lessons from history

23

Lessons from history


History does not provide a clear answer to the question
whether commercial and investment banking should be
separated. For every reason in favour of separation, there
is an argument against.
In the United States, attempts to separate commercial
and investment banking activities by law did not succeed
in preventing financial crises. It is true that there were
none between the 1930s and 1970s a very long time
by todays or pre-1930 standards. Some commentators
argue that the GlassSteagall Act should be credited for
this period of calm. Others point to a range of possible
contributory factors, including regulatory reforms of
securities markets and banking practice in the 1930s,
post-war controls on international capital flows, and the
historically low levels of private-sector debt that prevailed
from the 1940s into the post-war era. In any case, the
reforms of the 1930s did not stop crises striking again
four decades later, when they were still in place.
If the events of the twentieth century can offer no
conclusive lessons, past political debates may yield
some insights.
First, each of these debates has focused on solving the
current crisis and determining whether it would have
happened if different regulation had been in place.
The challenge of preventing future crises, possibly of a
different nature, has seldom received much attention.
Second, the outcome of the debates has hinged less on
a deep factual analysis of the underlying issues than on
the passion and conviction of individual politicians not
senior members of the executive, but MPs or senators
like Glass, Steagall, and Pecora in 1930s America.

Third, the political climate is of utmost importance for


governments seeking to push through difficult reforms,
as Germanys experience in the 1970s attests. By the
time the Gessler Commission concluded five years
of deliberation by publishing its report on the banking
system, the government had become exhausted by the
debate about employee and shareholder representation
on supervisory boards, the coalition between the SPD
and FDP had weakened, and there were divisions over
the stationing of US cruise missiles in Germany. The
political establishment had lost the appetite to undertake
a bold reform of the banking system.
History provides one more lesson. It would be unwise
to have high expectations of any process of structural
reform. The likelihood that it will solve the underlying
issues once and for all appears to be slender.

Dominic Casserley and Philipp Hrle are directors in


McKinseys London office. James Macdonald is a
financial historian and the author of A Free Nation Deep in
Debt: The financial roots of democracy.
The authors would like to acknowledge the contribution
of Charles Roxburgh.

24

Notes
1 Alan Greenspan addressing the Council of Foreign Relations, 15 October 2009.
2 Independent Commission on Banking Terms of Reference.
3 Addressing the US Senate in January 1861. Quoted in R. E. Sharkey, Money, Class, and Party, Johns Hopkins, 1959, p. 44.
4 Louis D. Brandeis, Other Peoples Money and How the Bankers Use It, F. A. Stokes, 1914, reprinted A. M. Kelley, 1971, p. 19.
5 Sen. Robert Bulkley, 10 May 1932, US Congressional Record, volume 75, part 9, p. 9911.
6 US Congressional Record, volume 77, part 4, p. 3956.
7 US Congressional Record, volume 77, part 4, p. 3956.
8 Rudolf Hilferding, Finance Capital, 1910, chapter 14 (from www.marxistsfr.org/archive/).
9 Andreas Busch, Banking Regulation and Globalization, Oxford University Press, 2008, p. 110.
10 Quoted in W. T. C. King, History of the London Discount Market, Cass, 1936, p. 37.
11 House of Commons Committee, British Parliamentary Papers, 1832, volume VI, p. 154.
12 The Economist, 25 October 1879, quoted in T. E. Gregory, British Banking Statutes and Reports, Oxford University Press,
1929, p. 299.
13 Sen. Robert Bulkley, 10 May 1932, US Congressional Record, volume 75, part 9, p. 9911.
14 W. Nelson Peach, The Security Affiliates of National Banks, Johns Hopkins, 1941, p. 175.
15 Robin Leigh Pemberton in October 1986, quoted in Richard Dale, International Banking Deregulation, Blackwell, 1992, p. 108.
16 Quoted in George Benston, The Separation of Commercial and Investment Banking: The GlassSteagall Act revisited and
reconsidered, City University, London, 1990, p. 37.
17 Thomas Lamont, Primary steps for banking reform, Proceedings of the Academy of Political Science, volume 15, number 2,
January 1933.
18 Quoted in Vincent Carosso, Investment Banking in America, Harvard University Press, 1970, p. 369.
19 Charles Calomiris, Universal Banking and the Financing of Industrial Development, World Bank working paper 1533, 1995.
20 Louis D. Brandeis, Other Peoples Money and How the Bankers Use It, F. A. Stokes, 1914, reprinted A. M. Kelley, 1971, p. 1.
21 Quoted in Ron Chernow, The House of Morgan, Simon & Schuster, 1990, p. 367.
22 Simon Johnson and James Kwak, 13 Bankers: The Wall Street Takeover and the Next Financial Meltdown, Random House,
2010, p. 6.
23 US Senate, 10 May 1932, US Congressional Record, volume 75, part 9, p. 9915.
24 Quoted in George Benston, The Separation of Commercial and Investment Banking: The GlassSteagall Act revisited and
reconsidered, City University, London, 1990, p. 209.
25 Quoted in Richard Dale, International Banking Deregulation, Blackwell, 1992, p. 22.
26 Lauchlin Currie, The decline of the commercial loan, Quarterly Journal of Economics, 45, 1931, p. 709. Quoted in Richard
Dale, International Banking Deregulation, Blackwell, 1992, p. 22.
27 Alan Greenspan, Subsidies and power in commercial banking, p. 5, in Proceedings of the 24th Annual Conference on Bank
Structure and Competition, Federal Reserve Bank of Chicago, 1990.
28 House of Commons Committee, British Parliamentary Papers 1832, volume VI, p. 154.
29 Walter Bagehot, Lombard Street: A description of the money market, 1873, 14th edition, John Murray, 1915, p. 188.
30 Henry Pole, Comptroller of the Currency, testifying in the House of Representatives, 14 March 1932.
31 Thomson Hankey, The Principles of Banking, Its Utility and Economy; with remarks on the working and management of the
Bank of England, second edition, Effingham Wilson, London, 1873, p. 30.
32 Forrest H. Capie, ed., History of Banking, volume IV, William Pickering , London, 1993, p. 368.
33 Sen, Robert Bulkley, 10 May 1932, US Congressional Record, volume 75, part 9, p. 9911.
34 Sen, Robert Bulkley, 10 May 1932, US Congressional Record, volume 75, part 9, p. 9911.
35 Hearings before the Subcommittee of the Committee on Banking and Currency, US House of Representatives, 23 March
1932, p. 59.
36 Letter from Federal Reserve Bank of New York to the Committee on Banking and Currency on 7 April 1932. Hearings before
the Committee on Banking and Currency, United States Senate, 2830 March 1932, p. 501.
37 Article in Fort Worth Star-Telegram,1 May 1932; US Congressional Record, volume 75, part 9, p. 9916.
38 Sen. Robert Bulkley, 10 May 1932, US Congressional Record, volume 75, part 9, p. 9913.
39 Hearings before the Committee on Banking and Currency, United States Senate, 2830 March, p. 501.
40 Christophe Charnley and Lawrence Kotlikoff, Limited purpose banking, The American Interest Online, MayJune 2009.
41 Lowell L. Bryan, Breaking Up the Bank: Rethinking an industry under siege, Dow JonesIrwin, 1988.

25

Further reading
Theo Balderston, German banking between the wars, Business History Review, 1991
George Benston, The Separation of Commercial and Investment Banking, City University, London, 1990
Andreas Busch, Banking Regulation and Globalization, Oxford University Press, 2008
Charles Calomiris, US Bank Deregulation in Historical Perspective, Cambridge University Press, 2000
Vincent Carosso, Investment Banking in America, Harvard University Press, 1970
Michael Collins, Money and Banking in the UK: A history, Routledge, 1990
Richard Dale, International Banking Deregulation, Blackwell, 1992
Irving Fisher, 100% Money, Adelphi, 1935
Richard Grossman, The shoe that didnt drop: Explaining banking stability during the Great Depression, Journal of
Economic History, volume 54, number 3, September 1994
Daniel Verdier, Universal Banking and Bank Failures Between the Wars, European University Institute, 1997
Ingo Walter, ed., Deregulating Wall Street, Wiley, 1985
Eugene White, Before the GlassSteagall Act, Explorations in Economic History, January 1986

Designed by London Graphics


Copyright McKinsey & Company
www.mckinsey.com

You might also like