Role of Commercial Banks in U.S. Credit Markets by Katherine Samolyk*
How important a role do commercial banks play in funding
nonfinancial borrowing? Ten years after
the end of the industrys most significant crisis since the Great Depression,
does banking remain a major player in financing the nations economic activity? This paper examines the evolving role that
commercial banks play in U.S. credit markets.
The available data reveal several consistent patterns over
the past two decades. First, there has
been a permanent increase in the overall borrowing capacity in credit
marketsin other words, an increase in the credit market pie associated with
the functioning of the economy. This
increase was associated with a decline in the share of domestic nonfinancial
borrowing that is directly funded by commercial banks. When debt growth leveled
off in the early 1990s, so did commercial banks share of this credit-market
pie. Banks smaller share of the credit-market pie reflects a dramatic shift in
the way loans to households and businesses are being financed. Specifically, asset securitization (the
pooling of loans and their funding by the issue of securities) has allowed
loans that used to be funded by traditional intermediaries, including banks, to
be funded in securities markets.
The data also reveal, however, that commercial banks still
play a significant role in funding business borrowers: we estimate that the
share of nonfinancial business borrowing that commercial banks fund on their
balance sheets has not declined notably in five decades. Nevertheless, there has been a clear shift in
how banks lenda shift from shorter-term lending not secured by real estate to
loans collateralized by business real estate.
This shift may reflect banks continuing comparative advantage in real
estate lending, a form of lending less well suited to the standardization
necessary for asset securitization.
With respect to borrowing by households, in contrast, we
find that the securitization of home mortgages andmore recentlyof consumer
credit has reduced the extent to which these types of loans are directly funded
by commercial banks (and savings institutions).
This finding is consistent with the broadening of household-sector
credit markets over time; longer-term increases in borrowing by households have
generally not been associated with greater intermediation through banks. The securitization trend, however, has had a
more severe effect on savings institutions than on commercial banks.
At the same time, the commoditization of credit marketsthat
is, the standardization, unbundling, and repackaging of payments and risks
associated with credit flowsmakes it harder to measure the importance of banks
as well as other intermediaries in providing credit-related services. Balance-sheet data on who is funding loans
can be a poor proxy for who is providing the financial services associated with
the credit flows. Commercial banks,
particularly larger institutions, provide significant services in originating,
servicing, and enhancing the liquidity and quality of credit that is ultimately
funded elsewhere. Hence, market-share measures
based on balance-sheet data are likely to understate the importance of banks to
a greater extent than even a decade ago.
The provision of financial services is, however, reflected in bank
earnings. And indeed, when one looks at
income-based measures of market share, one does not see any evidence of a
secular decline in the importance of commercial banking.
Thus, the conclusion of this study is that although the role
of commercial banks in U.S. credit markets has certainly evolved, banks remain
a critical part of the modern flow of funds that has broadened the availability
of credit in the U.S. economy.
Banks have historically been viewed as playing a special
role in financial markets for two reasons.
One is that they perform a critical role in facilitating payments.1 The other is that they have long played an
important, although arguably less exclusive, role in channeling credit to
households and businesses. Commercial
banks, as well as other intermediaries, provide services in screening and
monitoring borrowers; and by developing expertise as well as diversifying
across many borrowers, banks reduce the costs of supplying credit. Thus, in their role as lenders, banks are
often not merely buying someones debt; rather, they are providing significant
financial services associated with extending credit to their customers.2 And to the extent that investors want to hold
bank liabilities, banks can fund borrowers directly.
In the early 1990s, as the U.S. banking industry emerged
from its most significant crisis since the Great Depression, policy makers were
asking whether the importance of banks in financing economic activity had
become permanently diminished.3
Now, ten years later, the share of domestic debt funded on
commercial-bank balance sheets stands at just over 20 percent, down from 30
percent three decades ago. Commercial
bank loans now account for only 60 percent of short-term borrowing by
nonfinancial businesses, compared with 75 percent in the mid-1970s.4
Even now, therefore, when profitability and other measures
of performance indicate that banking has rebounded from the crisis, the role of
banks in U.S. credit markets remains under scrutiny. Other types of financial intermediaries and financial
instruments appear to have become more important in channeling funds to
businesses and households. Stories about
competition from other segments of the financial-services industry con tinue
to be reported in the popular press. For
example, according to a fairly recent report in the Wall Street Journal,
The financial services arm of General Electric Co. [GE
Capital] illustrates how nontraditional lenders are taking over from banks as
suppliers of credit to big slices of the U.S. economy. . . . Twenty years ago, banks and thrifts supplied
40% of the economys credit. . . . Today it is down to 19%. Housing financiers Fannie Mae and Freddie Mac
own about as many residential mortgages as all commercial banks combined.5
This paper assesses the evolving role of commercial banks in
U.S. credit markets during the past decade.
We use available data to quantify the importance of banks as credit
providersthat is, their market sharetaking a historical perspective in
assessing credit-market trends. Not
surprisingly, we find that the importance of banks depends on the markets one
chooses to consider and on how one measures banking services. However, some consistent patterns
emerge. From a historical perspective,
we now see that the debt buildup of the 1980s was actually a permanent increase
in the volume of debt associated with economic activity in the United
States. In other words, the
credit-market pie to be divided up among financial-service providers is now
substantially larger than it was 20 years ago.
And although overall the
provision of credit by banks has kept pace with the growth of the
economy, the capacity of the broader financial sector has grown by much
more. Accordingly, the share of our
economys debt that commercial banks fund directly has fallen relative to the
growth of the credit-market pie, reaching its low point in 1993 and then
An important dimension of these trends that is not always
emphasized is the dramatic change in the way credit flows in our economy are
being funded. Traditionally,
intermediaries funded portfolios of loans (and bonds) by issuing very different
types of liabilities (mainly deposits and insurance and pension liabilities) to
investors. But the growth of
credit-market activity in our economy during the past two decades has been
associated with the rise of intermediation in the form of asset securitization,
referring to the pooling of loans and their funding by the issue of
securities. Asset securitization
reflects a fundamental transformation of loan markets, particularly those where
households borrow. Home-mortgage and
consumer-credit markets have become commoditized, in the sense that these loan
products have become more standardized commodities, allowing the attendant
credit-related services to be unbundled, repackaged, and provided by a variety
of financial-service providers.
Moreover, standardization extends beyond the terms of the loan contracts
to the underwriting and pricing process, in which characteristics of the
borrower are increasingly linked to the use of statistical models in extending
and pricing credit.
The commoditization of credit often generates more layers of
intermediation between investors and the borrowers who ultimately receive the
funds. Intermediation funded by issues
of securities is often re-intermediated (for example, through mutual funds,
insurance companies, or pension funds).
The layering makes it harder to quantify the importance of banks (as
well as other intermediaries) in channeling credit from savers to borrowers
because it makes it more difficult to identify who is ultimately funding
certain types of loans. And quantifying
the value-added of the additional layers of intermediation is difficult as
Nonetheless, according to some fairly standard measures, we
find that commercial banks still play a significant role in channeling
credit. With respect to business
lending, we find that not only are banks important for small business
borrowers, but they also remain remarkably important for all business
borrowers: we estimate that the share of nonfinancial-sector business borrowing
that commercial banks fund directly has not declined notably in five decades. There has, however, been a dramatic shift in
how banks lend, a shift from shorter-term lending not secured by real estate to
loans collateralized by business real estate.
This shift may reflect banks continuing comparative advantage in real
estate lendinga form of lending less well suited to the standardization
necessary for asset securitization.
With respect to borrowing by households, we find that the
securitization of home mortgages andmore recentlyof consumer credit has
reduced the extent to which these types of loans are directly funded by
commercial banks (and savings institutions).
This finding is consistent with the broadening of household-sector
credit markets over time; longer-term increases in the debt capacity of the
household sector have not tended to be associated with greater intermediation
through banks. The securitization trend,
however, has had a more severe effect on savings institutions than on
The evolution of U.S. credit markets and the changing role
of commercial banks suggest that on-balance-sheet market-share measures
understate the importance of banks to a greater extent than even a decade
ago. Commercial banks, particularly
larger institutions, often provide important credit-related services to
borrowers that are ultimately funded elsewhere, but the provision of these
services is reflected in bank earnings.
Indeed, when one looks at income-based measures of bank market share,
one does not see evidence of a secular decline in commercial banking. Thus, although the importance of banks
depends on how one defines banking, from a variety of perspectives the
commercial banking industry remains far from extinct as a force in credit
The next five sections of the paper discuss the changing
nature of credit-market flows and the implications of the changes for using
balance-sheet data to measure bank market share; an overview of the historical
trends that culminated in the apparent decline of commercial banking during the
1980s and early 1990s; what researchers had to say about this apparent decline;
credit-market trends from the early 1990s to the present; and alternatives to
balance-sheet-based measures of bank market share. A final section summarizes our findings and
their implications for the future role of commercial banks in U.S. credit
Credit Market Concepts and
To examine trends in the role of commercial banks in U.S
credit markets, much of this paper uses 50 years of quarterly data from the
Federal Reserve Boards Flow of Funds Accounts (FFA). These accounts provide a detailed and
comprehensive picture of quarterly credit flows and balance-sheet outstandings
across various sectors of the U.S. economy since the early 1950s.6 They include a wealth of detail on specific
types of financial institutions and financial instruments; hence, they allow
one to study the evolution of the financial-services industry over time. However, with these data, ones findings
depend on the choice of what to measure.
Hence, we begin by providing a conceptual framework for thinking about
how to measure the role of commercial banks in U.S. credit markets.
Standard academic textbooks on banking often include a
diagram showing how credit markets traditionally workedthat is, how funds from
primary investors (those having accumulated wealth, i.e., savers, lenders) are
channeled to primary borrowers (those who need external finance to fund their
expenditures).7 As figure 1
indicates, primary borrowers include households seeking mortgage or consumer
loans; federal, state, and local governments financing their outstanding debt;
and nonfinancial businesses borrowing to finance their business activities
(larger publicly traded corporations also obtain external finance in equity
markets). These borrowers are classified
in the FFA as nonfinancial sectors.
Primary investors technically consist of the same groups, but it is
ultimately private individualsthat is, the household sectorthat accumulate
wealth (save) and need to invest it.
The financial sector, which facilitates external finance,
tends to be conceptually divided into direct credit markets, where investors
directly buy and hold securities issued by businesses or governments, and
indirect credit markets, where intermediaries pool the funds of many
investors to fund a pool of borrowers (see figure 1). Direct finance involves a brokerage function
but does not require intermediation per se (for example, when an investor buys
a U.S. Treasury security, even from a bank, the transaction does not involve
intermediation). In contrast, a key
feature of indirect finance is that it involves the funding of financial assets
by issuing to investors indirect claims on these assets. These indirect claims can have very different
characteristics (in terms of promised payments, liquidity, and default risks)
from the assets that they are funding.
The process by which a pool of financial assets can be funded by issuing
claims having different payment streams is referred to as asset transformation
(Gurley and Shaw ; Tobin ).
The nature of credit markets 50 years ago helps to explain
some important conventions in the FFA.
Specifically, the accounts were designed to measure the flow of credit
to nonfinancial-sector borrowers and the flows link to economic activity. To this end, the FFA defined the set of
credit-market instruments to include the types of claims that
nonfinancial-sector borrowers use to obtain financing in formal credit
markets. These include loans from
intermediaries as well as bonds and short-term paper issued in securities
markets.8 The traditional
indirect liabilities issued by intermediaries (deposits, and claims on
insurance and pension funds) are not credit-market instruments because
nonfinancial borrowers do not issue these types of claims. As we discuss throughout this study,
financial intermediaries can, and increasingly do, raise funds by issuing
credit-market debtmost often securitiesin their role as financial
middlemen. In the latter case,
credit-market debt issued by the financial sector is used to fund other
credit-market debt on the intermediaries balance sheets.
The distinction between total debt and nonfinancial-sector
debt was not as important 50 years ago.
As summarized in table 1, credit markets were somewhat simpler then:
commercial banks funded their lending by issuing checking and savings accounts;
savings institutions were largely home mortgage lenders that issued saving
accounts; insurance companies issued insurance polices and defined-benefit
pension-plan contracts, funding future payments on these contracts by investing
the premiums in securities and commercial mortgages. The financial sector did not raise funds by
issuing credit-market debt to a great extent; in the early 1950s, only 2.5
percent of total credit-market debt was issued by financial-sector firms.
In that world, intermediation between a borrower and a
lender generally involved one middleman and tended to involve a high degree of
asset transformation. Notably,
commercial banks funded relatively illiquid, unmarketable loans by issuing
extremely liquid demandable deposits. To
a large extent, the high degree of asset transformation reflected the
relatively high costs of processing and tracking information about financial
Asset securitization as a funding mode did not begin until
the 1970s, when federally sponsored agencies began to pool home mortgages and
issue mortgage-backed securities. Asset
securitization by the private sector did not become significant until the
mid-1980s. And although mutual funds
have a 60-year history, until the 1980s they accounted for only small shares of
the financial assets held by investors.
Until then, investors who wanted to hold stocks and bonds tended to hold
A prominent theme of this paper is that advances in the
application of information technologies in the financial-services industry have
dramatically changed both the nature of the asset transformation taking place
in U.S. credit markets and the types of indirect liabilities that are being
used to fund nonfinancial borrowers. In
recent decades, the volume of credit-market debtspecifically, marketable
securitiesissued by financial firms has grown dramatically. Currently, a third of total outstanding
credit-market debt is now issued by financial intermediaries (see figure 7),
and asset securitization accounts for a large share of this debt. Thus, as lower costs make it increasingly
feasible to standardize, unbundle, and repackage credit flows and risks, loans that
used to be funded by traditional lenders are increasingly being funded in
securities markets. Moreover, the
asset-backed securities are often bought by other intermediaries to be held in
their portfolios. Thus, unlike
the traditional flows of credit as diagrammed in figure 1, credit flows to
nonfinancial borrowers in U.S. credit markets increasingly involve more
complicated layers of intermediation between nonfinancial savers and
nonfinancial borrowers (figure 2).
When financial intermediaries hold the claims issued by other financial intermediaries,
an extra layer of intermediation is created.
For example, when a mutual-fund portfolio includes commercial paper or
bonds issued by a finance company or asset-backed securities issued to fund
consumer loans, there are two layers of financial intermediation between the
consumer who is borrowing and the mutual-fund investor.9 It is certainly possible for there to be more
than two layers of intermediation.
The increasing complexity of credit-market flows raises
methodological issues about how to measure bank market share. One very basic issue is simply that looking
at total credit-market debt increasingly overstates the amount of borrowing
associated with economic activity because a growing share of this total debt
comprises claims issued by financial intermediaries just to fund other debt.
In this regard, the focus on nonfinancial borrowing is
useful because it allows us to characterize the role of banks in facilitating
the flow of credit to the economy and to avoid double-counting debt issued
purely in the context of intermediation.
But even with this focus, the growing issuance of securities by
financial firms has made measurement issues more prominent: source data for the
FFA do not generally allow one to ascertain the extent to which corporate bonds
or commercial paper are issued by nonfinancial firms as opposed to financial
firms. Thus, in measuring funds advanced to nonfinancial businesses by banks,
mutual funds, and other holders of corporate debt, we estimate the shares that
are nonfinancial issues.10
Credit-Market Trends through
the Early 1990s
To understand the dramatic transformation of U.S. credit
markets, it is necessary to look at historical trends leading up to the
banking-sector problems of the 1980s and early 1990s and the apparent decline
in the importance of commercial banks as credit providers.
From the 1950s through the early 1980s, domestic
nonfinancial borrowing (by households, nonfinancial businesses, and
governments) grew roughly at the same rate as economic activity (measured in
terms of economic outputGross Domestic Product, or GDP). Indeed, the ratio of debt owed by domestic
nonfinancial sectors to GDP was remarkably stableso stable that it became a
stylized fact used by economists in analyzing macroeconomic issues such as
the effects of federal deficits (Friedman , Friedman ).11 But although total nonfinancial debt grew
roughly at the same pace as overall economic activity, borrowing by particular
nonfinancial sectors did not grow at the same rate: the share of borrowing by
households and nonfinancial businesses grew faster as the share of debt owed by
the federal government (accumulated during WWII) declined.
During this time, the number of commercial banks in the
United States was growing; thus, the industry continued to be made up of a
large number of banks that tended to be very geographically localized (partly
because of branching restrictions).
Banks also faced public policies that restricted entry, oversaw mergers,
and regulated permissible activities.12 On the liability side, commercial banks were
limited in terms of the types of liabilities they could issue and the rates
they could pay depositors. They were
generally relegated to the business of making (primarily) business loans and
providing transaction accounts (or close substitutes) in fairly localized
areas. They were also an important
funding source for the federal government.
Thus, for investment banking and insurance services, individuals and
corporations had to go to other financial-service providers. The phenomenon of the bank holding company
was a response to restrictions on the scale and scope of banking. A larger banking organization could be formed
if banks were held as affiliates, and if nonbank financial firms were held as
affiliates, the holding company could expand the scope of its activities to
encompass certain permissible lines of financial services. Of course, as holding companies evolved, they
too fell under regulatory scrutiny.13
The interplay that always exists among policy, regulation,
and financial-market trends was evident during this three-decade period,
particularly with respect to interest rates on deposit accounts. Rates on these accounts were regulated, but
in 1962 the marketable large certificate of deposit (CD) was created to
circumvent interest-rate ceilings and enable banks to pay market rates to
attract funds. On the asset side of the
balance sheet, after credit crunches in the late 1960s threatened the
availability of bank credit to commercial firms, the commercial-paper market
became considerably more active (Judd ); in effect, banks were making
fewer loans to prime corporate clients.14
Through the mid-1970s, commercial banks continued to be
special both in their role as lenders and as a transmission mechanism for the
implementation of monetary policy (Friedman , Fama , Wojnilower
). Of all the financial
intermediaries issuing claims to raise funds from investors, commercial banks
were the only ones allowed to issue demand deposits that could be used as a
direct means of payment, although demand deposits could pay no explicit
interest.15 Meanwhile, for
most businesses, the costs of direct financethat is, the raising of money by
issuing and placing bonds or commercial paperwere prohibitive enough that
their most attractive source of funds remained commercial banks. And of course commercial banks, as well as
savings institutions, were afforded federal deposit insurance. Hence, despite regulatory restrictions, periodic
credit crunches, and economic downturns, the U.S. commercial banking industry
performed quite well in the three decades following WWII. And although commercial banks share of
nonfinancial-sector debt dipped slightly as the war-related federal debt was
drawn down,16 it rebounded as borrowing by households and businesses
increased during the 1960s and early 1970s (see figure 3).
With the mid-1970s came a severe recession paired with high
inflation; however, relatively few banks failed. The number of commercial banks (and banking
organizations) was still increasing, although at a slower pace than banking
assets. Thus, although there were more
banks, banks were also, on average, getting larger as the industry established
more branches (Savage , Rhoades , Amel and Jacowski ). Banks were also becoming increasingly
complex in terms of their off-balance-sheet activities (such as issuing
standby letters of credit that promise to pay in the event of nonpayment of a
third party), which caught the attention of policy makers and researchers at
the time because of their implications for bank safety and soundness
(Lloyd-Davies ; Wolkowitz et al. ; Goldberg and Lloyd-Davies
; Benveniste and Berger ).
At the end of the 1970s, the pace of financial-market change
escalated significantly (Simpson ; Berger, Kashyap, and Scalise
). High nominal interest rates,
ceilings on the interest that could be paid on deposits, and better information
processing made the formation of money-market mutual funds a cost-effective
proposition (Mack ).17
These funds added to the competition associated with the creation of NOW
(Negotiable Order of Withdrawal) accounts by savings institutions in the
Ultimately, deregulation was implemented in the 1980s to
allow banks to compete more effectively: interest-rate ceilings were raised
(and were later eliminated), and commercial banks (and thrifts) were allowed to
offer a wider range of deposit accounts to attract depositors. But in the meantime, evolving financial
technologies were permanently altering the way financial markets channeled
capital to investment opportunities in the U.S. economy. Technical innovations in information processing
reduced the costs associated with financial transactions, and the result was a
proliferation of new products and new providers of financial services, as well
as the growth of existing ones. In
particular, asset securitization became an increasingly important means of
funding loans that had been traditionally funded by banks.18
As noted above, the origins of asset securitization can be
traced to the pooling and funding of mortgages by the government-sponsored
agencies involved in the secondary mortgage market. But by the late 1980s, securitizations of
loans by private asset-backed-securities (ABS) issuers had become a viable
means of funding other types of loans, such as consumer loans.
On the liability side, financial-sector development in the 1980s
also increased the competition that banks faced (Simpson ). Depository-institution deregulation allowed
savings institutions to issue the same types of deposits as banks. But more significantly, a growing mutual-fund
industry in tandem with the regulatory shift toward defined-contribution
pension plans served to channel the funds of smaller investors into direct debt
(and equity) markets. Not surprisingly,
it has been argued that the mutual-fund industry helped to reduce the role of
depositories in credit markets (see Mack  and Fortune , for
The evolution of financial-market technologies on both sides of
the balance sheet contributed to a dramatic increase in credit flows to
nonfinancial businesses and households, even while the federal government was
running large deficits (figure 4). After
three decades of relative stability, nonfinancial-sector borrowing increased
sharply as a ratio to GDP, from about 1.3 in 1981 to more than 1.8 by
1989. Financial intermediationincluding
a growing volume of securitized assetsincreased in tandem with the economys
appetite for debt. From the perspective
of researchers and policy makers at the time, the debt buildup was of great
concern, particularly the question of whether it was a debt bubble that was
going to burst in an economically detrimental fashion (Federal Reserve Bank of
Kansas City ).19 In
addition, the transformation of the asset menu available to investors through
banks and other intermediaries disrupted the historical relationships between
monetary aggregates and nominal output that the Federal Reserve Board used in
conducting monetary policy.20
Commercial banks, once the dominant type of financial
intermediary, did not appear to share in the proliferation of financial-sector
activity during the 1980s. The national
expansion was accompanied by regional economic downturns (related to troubled
industries, including oil and farming) severe enough to take down local banks
(FDIC ). By the early 1990s the
condition of the industry was marked by crisis, failures, and consolidation;
this was an industry under siege by competitors. Banking-sector problems continued as
real-estate markets collapsed on both coasts, taking their toll on exposed
institutions. And even as the industry
returned to a healthier state, the consolidation trend did not appear to be
In addition, the importance of commercial banks measured in
terms of credit flows seemed to be declining.
Between 1974 and 1994, the share of domestic nonfinancial-sector debt
that was advanced by U.S. commercial banks declined from 30 percent to just
over 20 percent (see figure 5). Savings
institutionsmost like banks in terms of their funding (deposits), regulations,
and decentralized industry structurefaced similar issues and appeared to be
faring even worse.
The Declining Role of
A host of studies assessing the evolving role of banking
were published in the wake of the banking crisis of the 1980s and early
1990s. These papers were written in the
context of what had become a decade-long consolidation trend, an even
longer-term decline in bank market-share measures, and concerns about a credit
or capital crunch.22 Not
surprisingly, opinions about the declining role of commercial banking
One view was that changes in the financial sectorevidenced
by the increasing competition from nonbank financial-service firmsreflected a
decreasing need for banks. From this
perspective, consolidation could be viewed a response to excess capacity in the
banking industry.23 Others
argued that the evidence did not support either the popular claims that large
banking firms were more efficient than smaller firms or the notion that the
industry was consolidating to eliminate excess capacity. Rather it was suggested that public policies
rather than performance gains were encouraging banks to merge.24 More sanguine observers argued that banking
was a battered but viable industry that needed industry consolidation and
regulatory reform if it was to adapt to the evolving financial environment. In this environment, such observers argued,
larger banks with broader banking powers would be able to compete by providing
more services at lower costs and by spreading the costs of new banking
technologies over more customers. In
addition, as banks became larger and expanded geographically, the geographic
scope of their activities would make them less vulnerable to the localized
economic problems that had plagued banks during the 1980s and early 1990s.25
Others research argued that when bank balance-sheet data
were looked at in isolation, they understated the share of financial services
provided by banks in the broader financial sector. Boyd and Gertler (1994b) conducted perhaps
the most extensive examination in this regard, documenting a host of alternatives
to standard measures of balance-sheet market share. These alternatives quantified activity in the
banking sector relative to activity in the broader financial sector or in the
entire economy. The term activity is
purposely general because Boyd and Gertler quantified banking-sector activity
(and the activity of other financial-service providers) in numerous ways; they
used measures that adjusted credit flows to reflect off-balance-sheet
activities as well as measures of profitability, employment, and compensation.
Boyd and Gertler argued that a careful reading of the
evidence did not support the view that banking was in decline. Although on-balance-sheet assets held by
commercial banks had declined as a share of total assets held by intermediaries,
they noted that this measure ignored the substantial growth in banks
off-balance-sheet activities, in offshore lending by foreign banks, and in the
size of the financial-intermediation sector.
They found that when measures of bank assets were adjusted for these
considerations, the measures showed no clear evidence of long-term
decline. Neither did an alternative
value-added measure, constructed with data from the national income
accounts. As Boyd and Gertler concluded,
At most, banking may have suffered a slight loss of market share lately. But this loss is a temporary response to a
series of adverse shocks rather than the start of a permanent decline. Thus, by defining banking more broadly to
include financial services that do not appear on bank balance sheets, the data
did not indicate an industry in decline.
Finally, others argued that banks were still important to
certain borrowersparticularly households and businesses that continued to rely
on banks for credit.26
Samolyk (1994b) analyzed bank market share from this perspective,
distinguishing between bank lending and other asset holdings (such as
securities holdings) and arguing that lending involves more intermediation
services than holding securities does.
Using FFA data to look at the markets where households and businesses
borrow, that study found shifts in how banks were funding private borrowers,
but the overall decline in market share was less than might have been
expected. As business lenders, banks
were facing increased competition from finance companies and direct credit
markets;27 the broadening of the commercial-paper market provided an
alternative to banks as a funding source.
However, as of the early 1990s, the securitization of business loans had
not really taken hold yet, and the share of business mortgages funded by banks
was actually increasing. Meanwhile, the
share of home mortgages and consumer credit that banks were funding was similar
to the share they had funded in the early 1960s. Moreover, although asset securitization was
becoming a more dominant way to fund household-sector borrowing, during the
1980s asset-backed lending grew more at the expense of savings institutions and
finance companies than of commercial banks.
During the 1990s, survey data obtained from households and
businesses also became important sources of information about the markets in
which banks competed as lenders. These
data were particularly useful because they yielded disaggregated pictures of
the financial services used by households and by businesses. For example, data from the triennial Survey
of Consumer Finances (SCF) were used to study the nature of rising
household-sector debt ratios during the 1980s and early 1990s.28 Kennickell, Starr-McCluer, and Sunden (1997)
found little evidence of a serious rise in debt payment problems even though
more families had debt, and more of it.29 On the other side of the household balance
sheet, the share of families who owned equities, and the amount of their
holdings, were also rising. The FFA
data, too, indicated rising debt burdens and equity holdings in the household
sector, but the SCF data were important because they indicated that aggregate
increases were associated with the use of these financial instruments by a
broader range of households (as opposed to increased usage by previously active
vein of research during the 1990s examined whether the services provided by
bankssuch as lendingare different from those provided by other
financial-service firms in ways that do not appear on a balance sheet. Using data on individual loans, Carey, Post,
and Sharpe (1996) compared corporate lending by banks with corporate lending by
finance companies.31 Although
their evidence suggested that both of these intermediaries were special in
solving informational problems, the two types of institutions did not make the
same types of loans. Although banks and
finance companies competed across the spectrum of borrower risk, finance
companies tended to serve observably riskier borrowers, especially highly leveraged
Passmore and Laderman (1998) investigated whether there were
differences between savings associations and commercial banks that would result
in reduced lending to traditional mortgage borrowers if the savings-association
charter were eliminated. Their empirical
tests did not indicate significant differences between savings associations and
commercial banks, suggesting that elimination of the savings-association
charter would not impair home mortgage credit availability.
A final vein of research that gained prominence in the 1990s
examined whether the consolidation of the banking industry into large
organizations adversely affected the availability of credit to small
businesses.32 This literature
did not directly yield evidence about bank market share vis-à-vis the nonbank
competition, but it raised the important question (somewhat overlooked in many
bank market-share analyses) of whether banks might be willingly reducing the
services they supplied to certain customers, such as small-business borrowers. If they were (or are), one would hope that
other financial-service suppliers would step forward to meet the credit needs
of these customers.
This discussion of some of the research of the 1990s
indicates that by looking at particular markets where banks are thought to play
a special role for lenders as well as by looking beyond the extent to which
banks are funding loans on their balance sheets, researchers were able to find
evidence that the decline in the share of total nonfinancial-sector debt funded
by banks could be misrepresenting the importance of banks in U.S. credit
markets. The next two sections examine
more recent credit-market trends from both of these perspectives to better
illuminate the evolving role of banks in the twenty-first century.
Trends: Who Is Funding Whom?
There is no doubt that the share of nonfinancial-sector debt
directly funded by commercial banks declined during the 1980s. More than a decade after that decline, it has
become clear that the debt buildup of the 1980s was actually a secular increase
in the volume of nonfinancial borrowing associated with economic activity in
the U.S. economy, which can be thought of as a permanent increase in the
economys financial capacity (figure 6).
Moreover, this increase in financial capacity was not associated with
intermediation funded by banks; hence, banks share of the pie had
declined. However, as the debt capacity
of the economys nonfinancial sector stabilized in the 1990s, so did the market
share of commercial banks. During the
past decade, the banking sector has rebounded to record profits, and although
consolidation has continued, it is occurring in the context of a healthy
industry.33 Here we look at
how the players and the instruments used to fund nonfinancial borrowers in U.S.
credit markets have evolved during the past decade.
Changed Players and Funding
The types of credit market instruments (loans and
securities) issued by nonfinancial borrowers to obtain funds in formal credit
markets have not changed as much as the types of instruments used to fund these
credit flows (table 2). Households still
obtain credit primarily in the form of home mortgages and consumer loans
(although the former now include home equity lines of credit, which were an
innovation of the 1980s). But now
asset-backed securitiesissued by both private asset-backed-securities (ABS)
issuers and federally related mortgage poolshave become an important funding
mode. And although nonfinancial
businesses still obtain credit primarily in the form of (a) loans
collateralized by business real estate (business mortgages), (b) other
(nonmortgage) loans from intermediaries, and (c) corporate securities, business
loans are also being securitized, and larger amounts of corporate securities
are funded by the issuance of mutual-fund shares. The appendix discusses changes in the
composition of investors portfolios and the way in which these changes relate
to changes in the funding of credit-market debt.
All of these changes are reflected in the growing extent to
which the commoditization of credit markets has allowed borrowing by businesses
and households to be funded in direct credit markets by securities issues.34 Roughly one-third of total outstanding
credit-market debt is now issued by the financial sector to fund other
credit-market debt (figure 7). And
whereas during the 1980s the growth of securitization largely reflected
mortgage funding through federally related mortgage pools, during the past
decade, securitization by private ABS issuers has expanded rapidly. FFA data estimate that now almost half of
outstanding corporate bonds have been issued by financial firms that fund other
credit-market debt, with private ABS issuers accounting for a fourth of the
corporate bond market (figure 8). The
commercial-paper market has always been dominated by financial-sector issues;35
during the past decade, however, private ABS issuers have become the dominant
issuers of commercial paper. More than
half of outstanding commercial paper (roughly two-thirds of financial issues)
is now funding securitized pools of loansincluding loans originated by banks
So who is funding whom?
The funding of loans through private securities markets and the
additional layers involved in modern credit flows have made it more difficult
for researchers to track the flow of funds between primary lenders and primary
borrowers. However, we use the FFA to examine
the extent to which loans to nonfinancial businesses and households are being
directly funded by commercial banks and other intermediaries.
Nonfinancial Business-Sector Credit
Borrowing by nonfinancial businesses can be divided into
three markets, each of which has historically accounted for roughly a third
of outstanding nonfinancial-sector business debt: corporate bonds, shorter-term
nonmortgage loans and commercial paper, and loans secured by business real
estate (business mortgages). Commercial
banks have tended to hold only small amounts of corporate bonds, so here we
focus on banks role in funding shorter-term nonmortgage business borrowing and
Shorter-term business borrowing (depicted in figure 10) is a
very heterogeneous credit market. It
includes all nonmortgage loans to nonfinancial businessesfrom vehicle or
equipment loans to business credit lines.
It also includes the very liquid commercial-paper issues that fund only
the largest corporations. Trends in the
composition of shorter-term business borrowing are also most often cited as
evidence of the declining importance of commercial banking (for example, by
Herring and Santomermo ). The
share of shorter-term nonfinancial-sector business credit funded directly by
banks declined from more than 75 percent in the early 1970s to just over 50
percent in the early 1990s (it has stabilized during the past decade). Meanwhile the share funded by finance
companies has steadily increased, now accounting for 20 percent of shorter-term
nonfinancial business-sector credit. ABS
issuers have made inroads in funding nonmortgage business loans, although they
still account for only 6 percent of this market. Interestingly, commercial paper, one of the
widely cited alternatives to bank borrowing, accounts only for roughly 7
percent of this short-term business credit market.
Trends in the business mortgage marketdefined to include
loans secured by business real estate, including commercial, multifamily
residential, and agricultural propertiesare depicted in figure 11. Commercial banks now directly fund more than
a third of outstanding business mortgages, up from 20 percent two decades ago
(and that was before the banking crisis).
Private ABS issuers, which did not exist 20 years ago, are now the
second-leading business-mortgage funding mode, accounting for 15 percent of the
market.36 Meanwhile, direct
funding by life insurance companies and savings institutions has declined
Figure 12 depicts commercial bank holdings of the three
types of business borrowing (combined) as a share of total outstanding
nonfinancial business-sector debt.38
The figure also relates this ratio to the growth of nonfinancial
business borrowing over time (measured relative to GDP). As the figure indicates, we estimate that
commercial banks fund roughly a third of nonfinancial business-sector
debt. And somewhat surprisinglygiven
discussions about the declining importance of banking for U.S. businessesthis
market share has not exhibited a downward trend during the past several
decades. But what we do find is a
notable shift in the type of business loans being extended by banks, from
shorter-term nonmortgage business loans to loan collateralized by business real
estate. Thus if one looks only at
nonmortgage bank lending, one sees a decline in bank market share, seemingly
related to the growth of nonfinancial business-sector debt. However, looking only at this decline is to
ignore the other markets where banks fund nonfinancial-business borrowers.
These business-sector trends are broadly consistent with
more recent evidence offered in the Federal Reserve Boards Report to the
Congress on the Availability of Credit to Small Businesses (2002).39 This report analyzes small-business financing
trends using a wide range of data sources and concludes that the patterns of
credit use evident in small-business survey data do not indicate a decline in
the importance of commercial banks (see also Bitler, Robb, and Wolken ). Commercial banks remain the leading source of
credit to small businesses that borrow and the most common source of credit
products of all types.40 The
report also discusses trends in asset securitization but notes that the
securitization of small-business loans has been modest, and it appears unlikely
that the securitization of small-business loans will increase significantly in
the near term. Thus far, the data do not
indicate that asset securitization has yet to become a dominant funding mode
for businesses, undoubtedly because business lending is less conducive to
standardization than other types of lending.
Home-mortgage debt has long been the primary type of
borrowing for households, and its share of total household-sector debt has
risen since the elimination in 1986 of tax deductions for interest paid on
nonmortgage credit.41 By the
early 1990s the secondary mortgage market had already made enormous inroads
into the funding of home mortgages, and the past decade has seen further
increases in the market share held by federally related mortgage pools,
government-sponsored enterprises (GSEs), and private issuers of asset-backed
securities (see figure 13). GSEs and
federally related mortgage pools now fund close to half of outstanding
home-mortgage debt, up from 35 percent a decade ago and from a mere 10 percent
in 1983. Commercial banks holdings of
home-mortgage debt have been remarkably stable, roughly equal (at 18 percent
now) to the level they were 20 years ago.
Clearly, this is the market that manifests the rise and fall of savings
institutions, whose share of the home-mortgage market has declined from more
than 50 percent 20 years ago to only 13 percent today. Some of this decline in market share (and the
stability of commercial bankings share) reflects the absorption of savings
institutions into the commercial-banking sector through mergers and charter
conversions. Life insurance companies,
which had significant home-mortgage holdings in the 1950s and 1960s, directly
fund almost no home mortgages today.42
Of course, commercial banks, savings institutions, and
insurance companies canand dofund the home-mortgage market indirectly when
they invest in the securities issued in the context of secondary market
activity. However, we net these indirect
holdings out of our market-share measures to avoid overstating the flow of
credit to home-mortgage borrowers.
In terms of consumer
credit, commercial bankings share of funding has not been so stable (figure
14). From the 1950s through the 1970s,
an institutionalization of the consumer-credit market took place, referring to
the increasing extent to which consumer credit was funded through
intermediaries (depository institutions and finance companies) rather than
directly by nonfinancial corporations (e.g., manufacturing and retail
firms). In its infancy, asset
securitization by private ABS issuers represented a shiftrather than an
increasein the intermediation of consumer credit. In the late 1980s and early 1990s, the shift
came at the expense of savings institutions and finance companies rather than
commercial banks or credit unions.
Indeed, as recently as 1994, close to half of outstanding consumer
credit was directly funded by commercial banks, and analysts speculated about
the long-run role of asset securitization as a funding mode for consumer
credit. A decade later, the speculations
are answered. The funding of consumer
credit through financial intermediation stands at an all-time high of 97
percent, and securitized pools now finance a third of outstanding consumer
credit. Commercial bank holdings of
consumer credit have declined to roughly a third of the market. Finance companies, savings institutions, and
credit unions account for the remainder.
In the evolving consumer-credit market, credit unions appear to have
fared the best among traditional intermediaries in terms of maintaining market
What then do the FFA data indicate about trends in the overall
importance of commercial banks in household-sector credit markets? Figure 15 relates commercial bankings market
share of home-mortgage and consumer debt to the overall growth of these types
of credit markets (the latter measured relative to GDP). Five decades of FFA data indicate that
commercial bankings share of home-mortgage and consumer credit has tended to
trend downward when borrowing capacity in these markets has been expanding
(again, measured relative to GDP). Thus
(as with broader nonfinancial-sector debt) although commercial bank funding of
home mortgages and consumer credit has grown, the overall flow of credit to
households through these markets has expanded by much more.
And Banks Competition?
Our analysis of the markets where households and businesses borrow
does not seem to validate the dire predictions suggested by some analyses. Although we certainly find that commercial
banks on-balance-sheet market share is lower than it was 20 years ago, the
decline we are measuring in the role of banks seems to be smaller than the
declines advanced by others. Here we
reconcile our findings with the findings of those who suggest a more serious
decline in the importance of banks; we then look at the competition faced by
We find less in the way of a decline than other researchers for
two reasons. First, when we examine the
role of commercial banks in channeling credit to nonfinancial-sector borrowers,
we net out credit-market debt issued to fund more debt. Netting out financial-sector debt (figure
16) yields generally stable market
shares since the early 1990s, while market-share measures that are based on
total debt show further declines through 2002 (figure 17).43 This indicates that the growing volume of
credit-market debt being issued by financial firms does not entirely reflect
funds being issued to displace lending by other financial-sector players. Indeed as we shall see, some financial-sector
issues of credit-market debt (notably those of Federal Home Loan Banks) are
channeled as sources of funding to other financial firmsincluding banks.
A second reason we find less of a decline in banks market
share than other researchers do is that we focus on commercial banking rather
than on banking in the sense of all depository institutions. Savings institutions historically have been
quite different from commercial banks and certainly have had distinctly
different experiences in the nations evolving financial environment.44
What, then, can we say about the overall market-share trends
for the competition? Figure 18
illustrates the share of nonfinancial-sector debt directly funded by sectors
commonly viewed as the strongest competition for banks in the new financial
world.45 Finance companies,
GSEs, and asset-backed-securities issuers largely fund their intermediation by
issuing securities in direct credit markets.
Mutual funds issue mutual-fund shares that may be held directly by
individuals or indirectly as assets by defined-contribution pension plans. The picture displays some intriguing results.
Not so surprisingly, we find that significant competition
has indeed come from asset securitization, both federally related and
private. The evolution of home-mortgage
financing in the direction of securitization suggests that large segments of
the mortgage market are better suited to funding by the issue of long-term
debt. Certainly this funding mode
reduces the interest-rate risks associated with funding long-term mortgages by
issuing deposits. As we argue in the
appendix, mortgage securitization is better suited to attracting long-term
investment funds associated with the accumulation of pension wealth. At the same time, the evolution of
home-mortgage funding has basically reduced the role of S&Ls (and insurance
companies) in funding home mortgages.
A somewhat more curious source of competition for commercial
banking appears to be securitization by private ABS issuers. This mode of
funding has affected the nonmortgage markets where households and businesses
borrowmost particularly, consumer-credit markets. However, as we discuss in the next section,
there is some question as to whether ABS issuers are competitors of banks or
merely an alternative mode of funding for banks (particularly large ones).
A third sector that is growing its market share as a funder
of the nonfinancial sector is the mutual-fund industry. Of course, since mutual funds hold securities
rather than loans, their growth represents less direct competition than the
growth of federally related mortgage pools and ABS issuers for the types of
lending that make banks specialloans to households and businesses. Indeed, as we show in the appendix, the growth
of mutual funds reflects a shift on the part of investors from holding
securities directly toward holding them through mutual funds to achieve
diversification of risks.
In recent decades two other sectorsones that are often
brought up in discussions of the growing competition faced by bankshave not
measurably increased the share of nonfinancial-sector debt they fund. These sectors are finance companies and
GSEs. Finance companies may also be
securitizing some of the more standardized types of loans they make to
households and businesses. GSEs share
of total credit-market debt has risen (see figure 7), but as noted
earlier, much of this debt funds intermediariesincluding commercial banks and
savings institutionsrather than nonfinancial borrowers.
Alternative Measures of
Bank Market Share
As discussed above, a decade ago, differences in what
constitutes banking services led to different assessments of the prospect for
banks. Researchers who tended to define
banking services in terms of what one sees on bank balance sheets (deposit
taking, lending, and investments in securities) tended to be more pessimistic
about the future of banking.46
Alternatively, researchers who tended to look beyond traditional banking
activitiesat an extreme, broadly defining banking as including the measuring,
managing, and accepting of riskargued that banks were not becoming less
important.47 From the latter
perspective, new services provided by bankswhether the selling of mutual-fund
shares to investors or the origination, sale, and servicing of loans funded by
securitizationsare merely banking in different forms.
In this section we broaden our perspective and ask how else
we might measure the importance of banks in the U.S. financial sector. The growth in our economys debt capacity
that has been funded through direct credit markets rather than through
traditional intermediation does not mean that intermediariesparticularly
banksdo not provide important services that facilitate funding in securities
markets. Here, therefore, we revisit the
notion that looking beyond what is measured on bank balance sheets may yield a
different view of the evolving role of commercial banks in facilitating credit
As credit-related services have become unbundled, layers of
transactions have been added to the intermediation process, and each layer
(albeit just a piece of the overall services associated with a given flow of
funds) adds value. Banks now provide
services in originating, servicing, or enhancing the creditworthiness of credit
flows that end up being funded elsewhere.
But even though the asset is not booked on a banks balance sheet, the
provision of any and all credit-related services should be reflected in the
income of the providers. Accordingly,
here we examine income-related measures of bank activity, which should reflect
the flow of services provided over time (since income is generated by the
production of goods and services in our economy). We examine data on income and profitability. We also look at estimates of output,
employment, and annual compensation in the banking sector compared with other
sectors in the economy.
Income and Profitability
The unbundling of credit-related services (as well as the
concomitant provision of off-balance-sheet financial services that generate
income) suggests that income-based measures of market share may in fact be
superior to balance-sheet-based constructs.
Ideally one would like to measure the income flows associated with the
provision of particular types of services (origination, servicing, packaging
and funding, credit enhancing) in particular types of credit markets (the
home-mortgage, consumer-credit, or business-credit markets). Unfortunately, comprehensive income-based
equivalents of the FFA do not exist.
Hence we must piece together evidence about banks provision of
credit-related services both on and off their balance sheets and must infer the
meaning of such evidence for the evolving importance of commercial banking in
the U.S. financial sector.
In both household- and business-sector credit markets, the
off-balance-sheet roles of commercial banks are increasingly important. In home-mortgage and consumer-credit markets,
bank off-balance-sheet activities tend to be related to the loan-securitization
process. More than half of home
mortgages and an increasing share of consumer debt are funded through
asset-backed securities, and commercial banks (particularly large ones) play
growing off-balance-sheet roles in these markets.
The Survey of Consumer Finances (SCF)which tends to
indicate where households obtain credit, not necessarily where the credit is
fundeddoes not indicate a decline in the share of debt that households
reported obtaining from commercial banks since 1989 (table 3).48 This contrasts with the market-share trends
in funding we found using FFA data.
In terms of consumer credit, the participation of commercial
banks (particularly large ones) in the securitization process tends to involve
more than loan origination. For
credit-card securitizations, large commercial banks originate, service, and
monitor the accounts. Thus, they have
the relationship with the borrower.
Through a legally separate special-purpose-entity they channel their
receivables into a package that can be funded by investorsincluding mutual and
pension fundsthat are willing to hold asset-backed securities. The originating institution manages the
assets being securitized to maintain the credit quality of the pool and often
holds a tranche to further enhance the pools quality. Finally, the bank that is sponsoring the pool
generally receives any residual income earned on these assets in the pool,
beyond what is promised to the investors buying the ABS-issuers
securities. What credit-related services
is the bank not performing in this process?
One could, therefore, argue that credit-card securitizations should be
included dollar for dollar when the share of commercial banks in
consumer-credit markets is measured, since this process is effectively a means
of funding the loans by the issuance of secured debt (rather than deposits).
As for business credit markets, although securitization
plays less of a role than it does in household-sector credit markets,
commercial banks have long played a role in providing the liquidity and credit
facilities that support the placement of debt in direct credit
marketsincluding debt issued by financial firms. This activity was highlighted by researchers
a decade ago.49 Thus income
for services not reflected on banks balance sheets extends beyond income
connected with loans sold into securitized loan pools.
Noninterest income as a share of earnings has received
considerable attention from analysts during the past two decades, for the share
of net operating revenue from noninterest income has more than doubled since
1980 (figure 19). Until recently, however, it has been difficult to identify
the extent to which the growth in noninterest income has been related to such
off-balance-sheet activities as asset securitization. Before 2001, bank Call Reports asked for
detail only on three categories of income: service charges on deposit accounts,
fiduciary (trust) income, and revenues from trading operations. All other noninterest income was reported in
two residual categories: Other Fee Income and All Other Noninterest
Income. Thus, although the relative
growth of bank noninterest income was driven by these two residual categories,
it was impossible to discern the nature of the activities associated with this
growth in income. Since the beginning of
2001, however, commercial banks (and savings institutions) have reported
greater detail about noninterest income.50
As summarized in table 4, these relatively new data indicate
some interesting facts about the noninterest revenue and the nature of bank
About 34 percent of noninterest income comes from what can be thought of
as traditional banking activities. The
traditional sources include deposit account fees, trust activities, and asset
sales not associated with securitization.
The large number of institutions reporting and the relatively low
concentrations of income earned by the five largest income earners in these
categories suggest that these sources of income are used fairly broadly.
Roughly 15 percent of noninterest income in 2001 came from
sources formerly associated with nonbank firms. The activities not generally
thought of as traditional banking include trading, investment banking (fees and
commissions from investment banking, advisory, brokerage, and underwriting
services), and insurance services. Of
these, income from trading activities is concentrated among a relatively small
number of institutions, but a wide range of banks earn at least some income by
providing investment banking and insurance services.
In terms of noninterest income associated with the
commoditization of credit, about 18 percent of noninterest income reported by
banks in 2001 reflected fees for servicing assets funded elsewhere and
securitization income (net gains on sales of securitized assets plus
nonservicing fees). As Waldrop (2002)
pointed out, The data show that securitization income (net gains on sales of
securitized assets plus non-servicing fees), at $16.4 billion for the year,
represented the largest amount of any of the new categories. The next-highest category was servicing fees,
at $11.6 billion.
Also included in the residual category of All Other
Noninterest income is income from unconsolidated subsidiaries, data processing
services, ATM usage fees charged to depositors from other institutions, and
income from other services (notably the provision of liquidity and credit
facilities). This residual category is
still the largest component of total noninterest income. At $51.3 billion in 2001, it represented 33
percent of commercial banks noninterest income.
Income is obviously closely related to profitability, and in
this regard banking has been holding its own.
The data on profits, like the data presented below on output,
employment, and compensation, are from the Bureau of Economic Analysis (BEA),
which constructs estimates of these measures for broad sectors of the U.S.
economy, including financial sectors.
One limitation of the BEA data series is that the classification of
credit agencies was changed in 1987.
Through 1987, commercial banking was specifically broken out as a
component of credit agencies; other lenders (such as savings institutions and
finance companies) were aggregated simply as other credit agencies. With the elimination of many of the differences
between commercial banks and savings institutions, industrial classifications
for credit agencies were redefined as depository institutions and other
credit agencies. Thus, we cannot
directly observe what is happening to commercial bankings share of economic
activity, but we can draw some inferences based on what we observe for all
depository institutions and on the relative shrinkage of the savings
institution industry since 1987. A
second limitation of these data is that they do not contain the same level of
detail as the FFA. Hence we cannot look
at trends for banking vis-à-vis particular types of other financial-service
Corporate profits for finance and insurance industries have
been rising as a share of total corporate profits, particularly since the
mid-1980s. Although the data after 1987
are for all depository institutions, the broad trends do not suggest an
industry in decline (see figure 20).
While banks were returning to record-setting earnings in the 1990s, so
were other financial-service providers (hence the decline in depository
institutions share of finance and insurance corporate profits), but the
profitability of the banking sector has outpaced that of other financial
sectors during the past few years.
To deal with the lack of detail in the BEA data on the performance
of other financial-service providers, FDIC analysts have compiled and tracked
profitability data available for publicly traded U.S. financial
corporations. Because large
conglomerates are involved, classifying financial enterprises into a single
financial-service category is not always easy.
In addition, like the BEA data, this information classifies banking to
include both commercial banks and savings institutions. These estimates yield some very interesting
patterns (figure 21).
First, the category of credit providers accounts for roughly
three-fourths of the net income of financial corporations. And although this finding no doubt reflects
declines in stock market valuations, this share of profits is comparable to the
share in 1984before the growth of U.S. financial capacity and the attendant
decline in the share of nonfinancial-sector debt directly funded by banks. Second, profitability reflects the amount of
intermediation services provided, not just the volume of funds brokered to
investors. From this perspective, it is
not surprising that government-sponsored enterprises have a relatively small
share of profits compared with the volume of credit they channel, both directly
and through asset securitization.
Finally and most pertinent to the point of this paper, the banking
industry (here, however, defined to include savings institutions as well as
commercial banks) appears to have maintained its market share quite well in
terms of profitability, rebounding from the problems encountered during the
1980s and early 1990s.
Output, Employment, and Compensation
The other economic-activity-based measures of bank market
share that may tell us something about the importance of commercial banking in
U.S. credit markets include output, employment, and compensation. These are useful because they indicate the
resources allocated to the provision of services by banks compared with the
resources used in the production of goods and services by other sectors.
Figure 22 illustrates the contribution of the finance and
insurance industries to GDP (it also illustrates credit agencies share of the
total output of the finance and insurance sectors). Consistent with the growth of credit-market
debt in the U.S. economy, financial-service firms account for a growing share
of aggregate output (except for an increase during the debt buildup of the
1980s, credit agencies share of financial- and insurance-sector output has
been remarkably stable). During the past
decade, the estimated output of depository institutions has been growing more
slowly than the estimated output of other credit providers (i.e., depository
institutions share of the GDP of finance and insurance industries has been
declining), but this trend may reflect the continuing contraction of the savings-institution
Figure 23 depicts employment trends measured in terms of
full-time-equivalent (FTE) employment.
Until the mid-1980s, employment in the finance and insurance industries
grew as a share of total employment in the U.S. economy; since then, employment
growth in finance and insurance industries has lagged employment growth in
other industries. This pattern probably
reflects the application of computer technologies in financial-service
industries, technologies that have increased the productivity per worker and
therefore reduced the person hours needed to produce a given level of financial
services. Commercial banking is an
important driver in these trends. Until
the 1980s, commercial bankings share of total employment in our economy was
rising (as was employment by other credit agencies, which included savings
institutions). The share of total
employment in commercial banking flattened out in the early 1980s, when the use
of ATMs became widespread, and the data for all depository institutions
indicate a long-term decline in these institutions FTE employment share during
the past 15 years. Although employment
growth for insurance industries has also been slowing, bankings declining
share of FTE employment has been more pronounced.
Similarly, data on total compensation (see figure 24)
indicate that although compensation in the finance and insurance industries has
been steadily increasing as a share of total compensation paid in the U.S.
economy, this increase has not been fueled by the growth of compensation in the
banking sector. Since 1987, compensation
paid by depository institutions has declined as a share of the total
financial-sector pie. Nevertheless,
because both employment and compensation trends reflect dramatic changes in the
technologies used to deliver financial services, they are likely to overstate
declines in the contribution of credit providers to economic activity.
This paper assesses concerns that banks are becoming less
important in U.S. credit markets, using available data to quantify the
importance of commercial banks as credit providersto quantify their market
share. Certainly as the debt capacity
of the U.S. economy expanded in the 1980s, the share of nonfinancial-sector
debt that was directly funded by banks declined. This decline was associated with a dramatic
increase in the extent to which lending to households and businesses became
securitizedthat is, standardized, pooled, and funded by the issue of securities. The shift away from traditional
intermediation towards asset securitization reflects not only changing credit
technologies but also the activity of government-sponsored enterprises. The shift towards funding credit though
securities markets also reflects fundamental changes in how individuals
accumulate assets, due to changes in technology, pension regulations, and
Long-term instruments such as home mortgages are arguably
better suited to securitization as a funding mode because of the maturity
mismatch inherent in depository institution funding. However, it is harder to make the same case
for the private securitization of some other types of loansfor example, credit
card receivables. Nevertheless, banks
play a prominent role in this type of securitization activity, so this may be a
way for banks to fund loans effectively by issuing secured debt while they
continue to be involved in all other aspects of the provision of credit
(including the relationship with the customer and the responsibility for
maintaining the quality of the pool of loans being funded). This alternative funding mode has allowed
banks to make more loans than they would have been able to if they had relied
on deposits alone as a funding source.
Thus, although commercial bankings on-balance-sheet activity
has declined as a piece of the credit-market pie, the industrys
off-balance-sheet activities are a growing source of income. Hence the ultimate finding of this study must
be that banking is evolving but does not appear to be declining. Even according to some fairly traditional
measures, the commercial banking industry remains remarkably important in
funding credit flows in the United Statesespecially credit flows to
What, then, can we say about the future of banking? Although the extent to which commercial banks
directly fund nonfinancial sectors in our economy has been stable since 1993,
such stability does not preclude future declines. Future increases in the economys debt capacity
are not likely to take the traditional form of intermediation. Thus, it will continue to be important for
researchers to study the evolving roles banks play in our financial sector, the
risks these roles pose for the industry, and the implications of these evolving
roles for broader financial stability.
For example, policy makers very much need evidence about the risks
inherent in the unbundling and repackaging of credit and about the implications
of these risks.
The secular shift by banks toward funding business lending
that is collateralized by real estate represents a shift to a type of lending
that has been associated with localized banking-sector problems. This association is likely to be most
problematic for community banks, which are more geographically focused in their
activities, than for larger banking companies operating a wide range of profit
centers over broader geographic areas.
In general, off-balance-sheet activities imply an ever more critical
role for large banking organizations.
The services that commercial banks provide in enhancing the
liquidity and credit quality of claims funded elsewhere undoubtedly reflect the
industrys unique status in our financial sector. The role of banks in making credit marketable
indicates that commercial banking remains a critical force in the modern flow
of funds that has contributed to the broader availability of credit in the U.S.
On the asset side of nonfinancial-sector balance sheets
there have also been fundamental changes in the way individuals hold financial
assets, particularly as changes in pension regulations and the availability of
mutual funds took hold during the 1980s.
In addition, changes in mechanisms used to conduct transactions and make
payments over the past several decades appear to have reduced the extent to
which individuals have to hold liquid assets as a share of their financial
portfolios for transactions purposes.52 Here we discuss the extent to which these
trends have made it easier for the growing volume of securities issued by
financial-sector firms to be absorbed.
The growth of the mutual-fund industry can be thought of as
a commoditization of investment opportunities in direct credit (and equity)
markets. By pooling many securities, a
mutual fund can reduce idiosyncratic risks and generate more-predictable risk,
return, and liquidity, compared with any given securities in the pool. Thus by choosing particular types of
securities, mutual funds can target particular characteristics for investors in
terms of risk, return, and even the social or ecological consciousness of the
underlying firms. Not only do personal
investors hold these funds, but institutional investorsparticularly life
insurance and pension fundsalso hold mutual-fund shares indirectly for their claimants.
Figure A.1 illustrates trends in the financial assets held
by the personal sector. Because of
inherently different risks and returns, we distinguish between holdings of debt
and holdings of equity mutual funds by the personal-sector portfolio. The resulting picture shows a dramatic
increase in individuals accumulation of financial assets, an increase
associated with the growth of pension wealth and increases in equities values
(until the past few years, at least).
The extent to which individuals direct holding of mutual funds has
facilitated the absorption of credit-market debt has been surprisingly
modest. And although money-market mutual
funds (MMMFs) have certainly displaced deposits somewhat, the overall level of
transactions accounts held by individuals (defined in the Federal Reserve Board
of Governors Survey of Consumer Finances to include bank accounts and nonbank
transactions accounts, such as MMMFs) as a share of GDP has remained fairly
stable over time. However deposits are
now a smaller share of the total portfolio of financial assets held by
individuals than thirty years ago. The
growth of insurance and pension reserves as a component of personal
financial-asset holdings has been the most prominent trend during the past few
In the early 1950s, roughly half of the personal sectors
financial portfolio consisted of claims on traditional intermediaries, that is,
life insurance companies, pension funds, and depository
institutions (figure A.2). And as noted, these intermediaries mainly held debt
issued by nonfinancial-sector borrowers; thus, intermediation tended to involve
a single layer: indirect liabilities held by individuals were used mainly to
fund nonfinancial borrowers directly.53 The other half of the personal-sector
portfolio was in the form of directly held securities (i.e., stocks and
bonds). Importantly, equities tended to
be held directly by individualsmost likely individuals with greater financial
The next few decades saw a shift in the personal-sector
portfolio toward the indirect liabilities issued by intermediaries, including
commercial banks and savings institutions.
By the mid-1970s, direct holdings of securities had fallen to around a
third of the personal sectors portfolio (we include custodial bank personal
trusts, first reported in the FFA in 1969, in this share). Mutual funds still accounted for only 1
percent of the personal sectors portfolio, and money-market mutual funds did
not yet exist. Traditional
intermediation had increased its share of the personal-sector portfolio to
almost 60 percent (figure A.3), and the market share allocated to deposits (and
currency) peaked at this time at 35 percent.
However, an interesting trend was under way as pension and insurance
sectors were increasing their holdings of equities (not depicted in the
figures). Thus, the decline in direct
equities held by individuals was offset by increases in equity investments by
In recent decades, trends evident in the 1970s have taken
off. As the size of the personal
sectors financial portfolio grew, so did the issue of securities by financial
intermediaries. Thus even though the
share of bonds and equities directly held by individuals remain close to 30
percent, it is now much more likely that holdings of securities are funding
financial intermediation. Importantly,
this is also true of mutual fund holdings (including MMMFs) and claims on
pension and insurance sectors, which now account for half of the
personal-sector portfolio. (See figure
The bottom line is that households have shifted from holding
securities directly to investing in intermediaries that invest in securities
(and in mutual-fund shares) that fund financial intermediaries as well as
nonfinancial borrowers. And whereas it
used to be mainly wealthier households that held securities, mutual funds and
pension plans have broadened the access of the average households access to
direct credit and equities markets. Thus
personal portfolio trends have facilitated the absorption of the greater amount
of debt being issued by direct credit marketsincluding debt issued by
financial sectorsincluding ABS-issuers and federally-related mortgage pools.
Altig, David, Susan Byrne, and Katherine Samolyk. 1992.
Is Household Debt Inhibiting the Recovery? Federal Reserve Bank of Cleveland Economic
Amel, Dean, and Michael Jacowski. 1989.
Trends in Banking Structure since the Mid-1970s. Federal Reserve Bulletin 75, no.
Antoniewicz, Rochelle L.
1996. A Comparison of the
Household Sector from the Flow of Funds Accounts and the Survey of Consumer
Finances. Finance and Economics
Discussion Series 96/26. Board of Governors
of Federal Reserve System.
Ascorbi, Anna, and Arthur Kennickell. 2003.
Changes in US Family Finances: Results from the 2001 Survey of Consumer
Finances. Federal Reserve Bulletin
89, no. 1:132.
Avery, Robert, and Allen Berger. 1991a.
Loan Commitments and Bank Risk Exposure.
Journal of Banking and Finance 15, no. 1:17392.
. 1991b. An Analysis of Risk-Based Capital and Its
Relation to Deposit Insurance Reform. Journal
of Banking and Finance 15, no. 4/5:84774.
Avery, Robert B., and Arthur B. Kennickell. 1991.
Household Savings in the U.S. Review
of Income and Wealth 37, no. 4:40932.
. 1993. U.S. Household Wealth: Changes from 1983 to
1986. In Research in Economic
Inequality: Festschrift to Nancy Ruggles 4, edited by Edward Wolff, 2768. JAI Press.
Benvenuiste, Lawrence M., and Allen N. Berger. 1986. An Empirical Analysis of Standby Letters of
Credit. Proceedings of a Conference on Bank Structure and Competition, Federal
Reserve Bank of Chicago: 387-412.
Berger, Allen N., Anil K. Kashyap, and Joseph Scalise. 1995.
The Transformation of the U.S. Banking Industry: What a Long, Strange
Trip Its Been. Brookings Papers on
Economic Activity 2:55218.
Berger, Allen N., and Gregory F. Udell. 1998.
The Economics of Small Business Finance: The Roles of Private Equity and
Debt Markets in the Financial Growth Cycle.
Journal of Banking and Finance 22, no.68:613673.
Bitler, Marianne P., Alicia M. Robb, and John D.
Wolken. 2001. Financial Services Used by Small Businesses:
Evidence from the 1998 National Survey of Small Business Finances. Federal Reserve Bulletin 87, no.
Boyd, John, and Mark Gertler. 1994a. The Role of Large Banks in the Recent
U.S. Banking Crisis. Federal Reserve Bank of Minneapolis Quarterly Review
18. No. 1.
. 1994b. Are Banks Dead? Or Are the Reports Greatly
Exaggerated? Federal Reserve Bank of
Minneapolis Quarterly Review 18, no. 3:223.
Boyd, John H., and Stanley L. Graham. 1991.
Investigating the Banking Consolidation Trend. Federal Reserve Bank of Minneapolis Quarterly
Review 15, no. 2:315.
Burger, Albert E.
1969. A Historical Analysis of
the Credit Crunch of 1966. Federal
Reserve Bank of St. Louis Review 51, no. 9:1330.
Canner, Glynn B., Thomas A. Durkin, and Charles A.
Luckett. 1998. Recent Developments in Home Equity
Lending. Federal Reserve Bulletin
84, no. 4:24151.
Carey, Mark, Mitch Post, and Stephen A. Sharpe. 1996.
Does Lending by Banks and Finance Companies Differ? Evidence on Specialization and Private Debt
Contracting. Finance and Economics
Discussion Series 96/25. Board of
Governors of the Federal Reserve System.
Carlson, John, and Katherine A. Samolyk. 1992.
The M2 Slowdown and Depository Intermediation. Federal Reserve Bank of Cleveland Economic
Carlson Keith M., 1993. On the Macroeconomics of Private
Debt. Federal Reserve Bank of St Louis Review: 75 no. 1: 53-66.
Carlstrom, C., and Katherine A. Samolyk. 1993.
Securitization: More than Just a Regulatory Artifact. Federal Reserve Bank of Cleveland Economic
Chase, Samuel. 1994.
Sixty Years of Banking in One Dozen Pages. Report of the Market Share
Task Force. American Bankers Association. Washington DC. Appendix D.
Cole, Rebel, and John Wolken. 1995.
National Survey of Small Business Finances. Federal Reserve Bulletin 18, no.
Cummings, Christine. 1987. The Economics of
Securitization. Federal Reserve Bank of
New York Quarterly Review, Autumn: 11-23.
DArista, Jane W., and Tom Schlesinger. 1994.
The Parallel Banking System. In Proceedings
of the Conference on Bank Structure and Competition, 288312. Federal Reserve Bank of Chicago.
Diamond, Douglas, W., 1984. Financial Intermediation and
Delegated Monitoring. Review of
Economic Studies. 51: 393-414.
Diamond, Douglas, W., and Philip Dybvig. 1983. Bank Runs,
Deposit Insurance, and Liquidity. Journal of Political Economy,
Duca, John V.
1992. The Case of the Missing
M2. Federal Reserve Bank of Dallas Economic
Review Q, II: 124.
Fama, Eugene. 1985.
Whats Different About Banks? Journal
of Monetary Economics 15: 29-40.
Fama, Eugene. 1980. Banking in a Theory of Finance. Journal of Monetary Economics 6:
Faust, Jon. 1990. Will Higher Corporate Debt Worsen Future
Recessions? Federal Reserve Bank of
Kansas City Economic Review. March/April: 19-34.
Federal Deposit Insurance Corporation (FDIC). 1997. History
of the Eighties, Lessons for the Future.
Vol. 1, An Examination of the Banking Crises of the 1980s and Early
Federal Reserve Bank of Boston, Monthly Mutual Funds Report
Federal Reserve Bank of Kansas City. 1986. Debt,
Financial Stability, and Public Policy: A Symposium Sponsored by the
Federal Reserve Bank of Kansas City.
Federal Reserve Board of Governors. 2002.
Report to the Congress on the Availability of Credit to Small
Federal Reserve Board of Governors, Flow of Funds Accounts
through Q2. 2002.
1997. The Mutual Fund
Industry. Federal Reserve Bank of Boston
New England Economic Review, 4572.
Freidman, Benjamin M.
1978. Crowding Out or Crowding
In? Economic Consequences of Financing
Government Deficits. Brookings Papers
on Economic Activity, 593654.
. 1980. Postwar Changes in American Financial Markets. In The American Economy in Transition,
edited by Martin Feldstein, 177.
University of Chicago Press.
. 1981. The Relative Stability of Money and Credit
Velocities in the United States: Evidence and Some Speculation. Working paper, no. 645. National Bureau of Economic Research.
. 1993. Ongoing Change in the U.S. Financial Markets:
Implications for the Conduct of Monetary Policy. Unpublished manuscript.
Greenspan, Alan. 1994.
Optimal Bank Supervision in a Changing World. Proceedings from the Conference on Bank
Structure and Competition. Federal
Reserve Bank of Chicago: 1-8.
Goldberg, Michael, and Peter Lloyd-Davies. 1985.
Standby Letters of Credit: Are Banks Overextending Themselves? Journal of Bank Research 16, no. 1:2839.
Gorton, Gary, and Richard Rosen. 1995.
Corporate Control, Portfolio Choice, and the Decline of Banking. Journal of Finance 50, no.
Gurley, John G., and Edward S. Shaw. 1956.
Financial Intermediaries and the Savings Investment Process. Journal of Finance. May: 25776 .
. 1957. The
Growth of Debt and Money in the United States, 18001950; A Suggested
Herring, Richard J. and Anthony M. Santomero. 2000.
What is Optimal Financial Regulation?
University of Pennsylvania.
Warton Financial Institutions Center working paper 00-34.
1992. Policy Implications of
Recent M2 Behavior. Federal Reserve Bank
of Kansas City Economic Review 77, no.3:2136.
Holland, David, Don Inscoe, William Kuta, and Ross
Waldrop. 1996. Interstate Banking: The Past, Present, and
Future. FDIC Banking Review 9,
Ip, Greg. 2002. Alternative Lenders Buy the Economy But Also
Pose Risk. Wall Street Journal,
1979. Competition between the
Commercial Paper Market and Commercial Banks.
Economic Review. Federal Reserve Bank of San Francisco. Winter
Kaufmann, George G. 1993.
The Diminishing Role of Commercial Banking in the U.S. Economy. in: The Crisis in American Banking. ed. Lawrence H. White. New York University
Kennickell, Arthur B., and Martha Starr-McCluer. 1996.
Household Saving and Portfolio Change: Evidence from the 19831989 SCF
Panel. Finance and Economics Discussion
Series 96/18. Board of Governors of the
Federal Reserve System.
Kennickell, Arthur B., Martha Starr-McCluer, and Annika E.
Sunden. 1997. Family Finances in the U.S.: Recent Evidence
from the Survey of Consumer Finances. Federal
Reserve Bulletin 83, no. 1:124.
Koppenhaver, G. D. 1999.
Circle Unbroken: Bank-Affiliated Money Market Mutual Funds. Proceedings from the Conference on Bank
Structure and Competition. Federal Reserve Bank of Chicago: 430-447.
1979. Survey of Standby Letters
of Credit. Federal Reserve Bulletin
65, no. 9:71619.
1993. Recent Trends in the Mutual
Fund Industry. Federal Reserve
Bulletin 79, no. 11:100112.
2003. The Economics of Money,
Banking and Financial Markets. 6th ed. update. Addison-Wesley.
Orphanides, Athanasios, Brian Reid, and David Small. 1994.
The Empirical Properties of a Monetary Aggregate that Adds Bond and Stock Funds
to M2; in Proceedings: A Symposium on
Mutual Funds and the Monetary Aggregates. St. Louis, Mo.: Federal Reserve
Bank of St. Louis.
Passmore, Wayne, and Elizabeth Laderman. 1998.
Is Mortgage Lending by Savings Associations Special? Federal Reserve Bank of San Francisco Review
0, no. 2:3045.
1986. Securitization. Federal Reserve Bank of Chicago Economic
Perspectives, vol. 10, no. 4:1631.
Pearce, Douglas K. 1985.
Rising Household Debt in Perspective. Federal Reserve Bank of Kansas
City Economic Review, July/August: 3-17.
Rhoades, Stephen A. 1985.
Mergers and Acquisitions by Commercial Banks, 1960-83. Staff Study no.
142. Board of Governors of the Federal
Rhoades, Stephen A.
1996. Bank Mergers and Industry
Wide Structure, 198094. Staff Study no.
169. Board of Governors of the Federal
1994a. U.S. Banking Sector
Trends: Assessing Disparities in
Industry Performance. Federal Reserve
Bank of Cleveland Economic Review 30, no. 2:217.
. 1994b. Banking and the Flow of Funds: Are Banks
Losing Market Share? Federal Reserve
Bank of Cleveland Economic Commentary.
. 1997. Small Business Credit Markets: Why Do We Know
So Little about Them? FDIC Banking
Review 10, no. 2:1432.
Savage, Donald T.
1982. Developments in Banking
Structure, 197081. Federal Reserve
Bulletin 68, no. 2:7785.
. 1993. Interstate Banking: A Status Report. Federal Reserve Bulletin 79, no.
1988. Developments in the U.S.
Financial System since the Mid-1970s. Federal
Reserve Bulletin 74, no. 1: 113.
Teplin, Allen M.
2001. The US Flow of Funds
Accounts and Their Uses. Federal
Reserve Bulletin 87, no. 7:43141.
1963. Commercial Banks as
Creators of Money. In Banking and
Monetary Studies, edited by Deane Carson, 40819. Richard D. Irwin, Inc.
2002. New Reporting Offers
Insight into Bank Activities. FDIC FYI,
1993. Is the Banking Industry in
Decline? Recent Trends and Future
Prospects from a Historical Perspective.
Federal Reserve Bank of St. Louis Review 75, no. 5:322.
1980. The Central Role of Credit
Crunches in Recent Financial History. Brookings
Papers on Economic Activity 2:277337.
Wolkowitz, B., P. Lloyd-Davies, B. Gendreayu, G. Hanweck,
and M. Goldberg. 1979. The Use of Contingencies and Commitments by
Commercial Banks, Below the Bottom Line.
Staff study. Board of Governors
of the Federal Reserve System.