View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

September/October 1993
Volume 78, Number 5

Federal Reserve
Bank of Atlanta

In This Issue:
D e r e g u l a t i o n a n d t h e Opportunities f o r
C o m m e r c i a l Bank Diversification
Private Insurance o f Public Debt:
A n o t h e r Look at t h e Costs a n d Benefits o f
Municipal Insurance
FYI—Commercial



Bank Profits







/«pnpmic
jgfeFiew
September/October 1993, Volume 78, N u m b e r 5




Federal Reserve
Bank of Atlanta

President
R o b e r t P. Forrestal
S e n i o r Vice President a n d
D i r e c t o r of R e s e a r c h
Sheila L. T s c h i n k e l
V i c e President a n d
A s s o c i a t e Director of R e s e a r c h
B. F r a n k K i n g

Research Department
William Curt Hunter, Vice President, Basic Research
Mary Susan Rosenbaum, Vice President, Macropolicy
Thomas J. Cunningham, Research Officer, Regional
William Roberds, Research Officer, Macropolicy
Larry D. Wall, Research Officer, Financial

Public Affairs
Bobbie H. McCrackin, Vice President
Joycelyn Trigg Woolfolk, Editor
Lynn H. Foley, Managing Editor
Carole L. Starkey, Graphics
Ellen Arth, Circulation

The Economic Review of the Federal Reserve Bank of Atlanta presents analysis of economic
and financial topics relevant to Federal Reserve policy. In a format accessible to the nonspecialist, the publication reflects the work of the Research Department. It is edited, designed, produced, and distributed through the Public Affairs Department.
Views expressed in the Economic Review are not necessarily those of this Bank or of the Federal Reserve System. '
Material may be reprinted or abstracted if the Review and author are credited. Please provide the
Bank's Public Affairs Department with a copy of any publication containing reprinted material.
Free subscriptions and limited additional copies are available from the Public Affairs Department, Federal Reserve Bank of Atlanta, 104 Marietta Street, N.W., Atlanta, Georgia 30303-2713
(404/521-8020). Change-of-address notices and subscription cancellations should be sent directly to the Public Affairs Department. Please include the current mailing label as well as any new
information. ISSN 0732-1813




(Contents
Federal Reserve Bank of Atlanta Economic Review
September/October 1993, Volume 78, Number 5

Deregulation and the
Opportunities f o r
Commercial Bank
Diversification
Larry D. Wall, Alan K. Reichert,
and Sunil Mohanty




Congress has resisted passing legislation deregulating the activities of commercial banks largely because of fears that banks might
take excessive risks and exploit the federal deposit insurance system,
fears fueled both by the collapse of the thrift industry and by the large
number of bank failures in recent years. Meanwhile, competition
from nonbank firms offering financial services has been strengthening in stride with developments in communications and informationprocessing technology. Banks must face this competition with hands
still tied by legislation originating in the 1930s.
An important issue in the debate over expanded bank holding
company powers is whether diversification would increase or decrease risk for these institutions. To address this question, the authors
of this article examine the portfolio effects of combining bank activities with nonbank activities currently impermissible or limited for
banks. The study considers the behavior of portfolios before and after
passage of the Depository Institutions Deregulation and Monetary
Control Act, constructing common pairwise portfolio combinations
and calculating efficient portfolios consisting of a bank and several
nonbank industries.
The pairwise studies suggest that many combinations would, at
least temporarily, increase the riskiness of bank holding companies;
banks would form the dominant part of most low-risk combinations.
On the other hand, merging with several nonbank firms might provide a bank holding company a substantially higher return and lower
risk than one restricted to traditional banking activities. Because financial conglomerates are built one merger at a time, these results
and those of other studies point to a difficult problem for banks and
policymakers: both must deal with the knowledge that a bank holding
company will not be able to form an efficient portfolio of activities
without being allowed to engage in individual mergers that may pose
at least a temporary increase in risk.

Private Insurance of Public
Debt: Another Look at the
Costs and Benefits of
Municipal Insurance
Stephen D. Smith and
Richard B. Harper

PYI— Commercial Bank
Profits in 1992
B. Frank King




The municipal bond market has changed dramatically in the past
two decades. In the mostly institutional market of twenty years ago
municipal bond insurance had just appeared but received little attention from investors experiencing default rates of less than 0.005.
Today the municipal bond market has been transformed into one in
which insured municipal issues are primarily held by retail investors.
What have been the sources of this changing investor demand? Has
the shift in the market significantly altered the benefits of insurance
to issuers? This article considers these questions.
The authors' results suggest that tax changes in the 1980s caused
the shift in the composition of investor demand. Using data from issues sold by the State of Florida and its subdivisions from 1990
through 1992, the study provides evidence suggesting that on average, state and local governments still realize a net financing benefit
from selling insured bonds. These benefits seem to have come from
lower prices associated with the cost of insurance through increased
competition among underwriters.
While this information is largely good news for issuers, to some
extent insurers have tried to offset the declining margins by increasing their leverage ratios. This fact and its implications for risk,
along with the increasing fiscal demands placed on state and local
governments, create a situation calling for greater regulatory awareness of the changing volume and pricing structure for municipal
bond insurance.

Thanks in part to falling interest rates, shrinking loans losses, and
econom
'mProv^nS
y> U.S. banks' 1992 earnings were at the highest levels in many years. This article examines some of the reasons
f o r this increased profitability in banks of all sizes both nationwide
and in the Southeast in 1992. Extensive tables provide details about
bank profitability from 1988 through 1992.

an




iE)eregulation and the
Opportunities for
Commercial Bank
Diversification
Larry D. Wall, Alan K. Reichert, and Sunil Mohanty

anking legislation and regulation originating in the 1930s prevents
commercial banks from being affiliated with nonbank providers of
financial services. 1 As a result, competition between financial services providers has been limited. For banks, regulations have during most of the period seemed a double-edged sword, restricting
their ability to leverage their financial talents in nonbanking areas but also
enhancing their value by reducing competition in the markets for traditional
banking services. However, continuing developments in communications and
information-processing technology as well as financial theory have severely
eroded the barriers that once protected commercial banks. Commercial banks
must face increasing competition from nonbank firms with their hands still
tied by restrictive legislation.

Wall is the research officer in
charge of the financial section
of the Atlanta Fed's research
department, Reichert is a professor of finance at Cleveland
State University, and Mohanty
is a doctoral candidate in the
finance department at Cleveland State University.

Federal Reserve Bank of Atlanta




Many banking organizations, especially larger ones, have pointed to their
declining fortunes and asked for such restrictions to be relaxed. Generally
sympathetic, bank regulatory agencies have been as flexible as possible in allowing bank entry into nonbank financial fields. However, the same legislation that limits banks also constrains the regulatory agencies. Banks have
extended their efforts to Congress, where they have petitioned with increasing urgency for m o r e than a decade for deregulation permitting them to
provide nonbanking services. The issues involved in requests for legislative
c h a n g e p r i m a r i l y c o n c e r n the w a y s in w h i c h d e r e g u l a t i o n w o u l d a f f e c t
(1) the cost and availability of financial services to domestic consumers,
(2) the competitiveness of U.S. financial firms in international markets, (3) systemic risk and the government safety net as nonbank firms continue expansion into the markets for traditional banking services, and (4) bank safety as
banks participate in nonbank financial activities.

Economic Review

1

Congress has resisted passing legislation deregulating banks largely because of fears that banks might
take excessive risks and exploit the federal deposit insurance system, fears fueled in part by the disaster that
followed deregulation of the thrift industry and in part
by the large number of bank failures in recent years. 2
The validity of the fears about banks' behavior hinges
on several questions. For example, would a portfolio of
b a n k and n o n b a n k activities carry m o r e or less risk
than traditional banking activities? Would bank managers adjust operations of nonbank (and bank) activities in ways that would increase or reduce their organizations' overall riskiness? Can financial problems in
a nonbank subsidiary be contained so as to minimize
the danger of systemic risk and costs to the insurance
f u n d ? A s s u m i n g that deregulation would also allow
nonbank firms to enter banking, how would traditionally nonbank firms manage their bank subsidiaries?
This study f o c u s e s first of all on the question of
whether deregulating bank activities would increase or
decrease banks' riskiness or would be likely to have no
impact. More specifically, it examines the portfolio effects of combining bank activities with nonbank activities in w h i c h b a n k p a r t i c i p a t i o n is e i t h e r c u r r e n t l y
impermissible or severely constrained. The approach is
to analyze the historic accounting rates of return on
various bank and nonbank activities reported to the Internal Revenue Service (IRS). The issues concerning
risk as they relate to bank management decisions and
to the containment of problems in nonbank subsidiaries
are addressed in the boxes on pages 4 and 8, respectively.
Several earlier studies have examined portfolio diversification as it is affected by deregulation of bank
activities. Providing a more recent analysis of the IRS
data, this study also presents a consistent analysis of
the behavior of portfolios that included bank and nonbank activities before the period of deregulation ushered in by the Depository Institutions Deregulation and
Monetary Control Act ( D I D M C A ) in 1980 and the period after its passage. Lastly, in addition to constructing
common pairwise portfolio combinations (that is, combining two industries at a time), this analysis also extends and updates Robert E. Litan's (1985) calculations
of efficient portfolios consisting of a bank and several
nonbank industries. That is, the portfolios that have the
lowest level of risk for any given return (or, equivalently, the highest return for any given level of risk) are calculated.
Results indicate that bank risk increased after the
passage of D I D M C A , but so did the riskiness of most
other financial industries. T h e analysis of p a i r w i s e

2

Economic Review-




combinations found more opportunities for risk reduction t h a n o t h e r r e c e n t s t u d i e s u s i n g d i f f e r e n t d a t a
sources have found. For example, this study found that
f o r m i n g a c o m b i n a t i o n of b a n k s and securities brokers/dealers m a y reduce risk if the securities operations
constitute no more than 25 percent of the consolidated
firm. The analysis of efficient portfolios for the 198189 subperiod suggested minimal opportunities for diversification gains in areas closely related to traditional
banking such as the thrift industry and personal and
business credit. However, the results suggest that portfolios including certain industries in which banks have
been seeking to remove barriers to growth, such as insurance and mutual funds, offer significant opportunities f o r i n c r e a s i n g return while l o w e r i n g risk. T h e
analysis of optimal bank holding company portfolios
and nontraditional activities also uncovered t w o surprises: (1) although the securities brokerage industry
can be risk-reducing in pairwise portfolios, it never entered into efficient portfolios, and (2) the real estate
subdivider and developer industry enters the efficient
portfolio even though it experienced a negative average
return on assets during the 1981-89 subperiod. The second finding, which reflects that risk-reduction gains
from diversifying into real estate subdivision and development were more important than the negative returns reported by this industry, helps to illustrate the
importance of examining portfolios of activities.

Ziterature Review
A variety of studies have e x a m i n e d the e f f e c t of
combining bank and nonbank activities into a single
portfolio. This section begins with a chronological review of a number of those studies. The review then discusses the wide variety of methodologies used in prior
studies and the principal differences and trade-offs involved in using various methodologies. It also suggests
that most studies find limited potential for reducing
portfolio risk by allowing banks to expand their investments in nontraditional activities.
Prior Studies. Arnold A. Heggestad (1975) used
IRS industry earnings data to examine the riskiness of
various industries between 1953 and 1967. 3 He measured riskiness by the coefficient of variation of returnon-equity for thirteen different industries. In addition,
Heggestad correlated industry earnings with returns to
banking. He discovered that commercial banking was
one of the least risky activities but also found that industries such as leasing, insurance, or real estate offer

September/October 1993

attractive risk-reducing diversification potential given
their negative correlation with banking.
Roger D. Stover (1982) examined the effect of diversification on the market value of bank holding companies using both industry data and firm-specific data
obtained from Moody's Industrial Manual, Moody s
Banking and Finance Manual, and Compustat for the
1959-68 period. He began by determining the debt capacity of a portfolio of bank and nonbank assets given
a fixed probability of failure. He then assumed that a
firm's value increases as its debt capacity increases.
Stover's results have direct implications for portfolio
risk. In his model, an increase in debt capacity would
imply a decrease in the risk of firm failure, with leverage held constant. He c o n c l u d e d that bank holding
company diversification to include activities outside of
banking increased the organization's value. His analysis suggests that banks may benefit from diversifying
into the following industries: fire and casualty insurance, investment banking, land development, and savings and loan companies.
Litan (1985) used IRS data to examine the portfolio
consequences of combining banks with sixteen other
financial services industries. He analyzed industry
return-on-assets figures for two subperiods, 1962 to
1972 and 1973 to 1981. The first part of the analysis
examined the coefficients of variation and correlations
of returns and found that the correlations between bank
and various nonbank activities are either small positive
or small negative values. 4 However, the risk-reduction
opportunities suggested by the correlations are at least
partially offset by higher coefficients of variation. Litan
then estimated efficient portfolios of the various activities for given return on assets values, a procedure that
incorporates differences in mean returns, standard deviations of returns, and correlations of returns. These efficient portfolios always place a substantial percentage
of their assets in banks but also place a significant fraction in nonbank activities. The fraction invested in nonbanking for the subperiod from 1973 to 1981 contained
from 46 percent in the riskiest portfolio to 12 percent in
the least risky portfolio. Litan also considered the impact of individual nonbank firms acquiring a "typical"
bank. He found that such an acquisition could increase
the risk exposure of some depositories even as it lowers
the risk for some nonbank parents.
Elijah Brewer III, Diane Fortier, and Christine Pavel
(1988) analyzed the relationship between bank risk and
nonbank activities using daily stock market returns for
a sample of 325 nonbank firms and 170 banking organizations during three very different e c o n o m i c periods—1980, 1982, and 1986. The nonbank firms were

Federal Reserve Bank of Atlanta




categorized into thirteen broad industry groups. Brewer, Fortier, and Pavel found that most of the currently
authorized n o n b a n k activities—especially m o r t g a g e
banking and consumer finance companies—are considerably riskier than commercial banking. On the other
hand, the variance of daily returns for impermissible
activities such as acting as insurance agents and brokers and property/casualty insurance underwriters were
relatively low. They conclude that a small investment
in a few such comparatively low-risk nonbank activities would actually reduce risk. Furthermore, a limited
investment in high-risk security-related activities would
not materially increase bank holding company risk.
John H. Boyd and Stanley L. Graham (1988) examined accounting data and stock market returns during
the 1971-84 period to determine the impact of nonbank
activities on the markets' assessment of bank holding
c o m p a n y risk. U s i n g both types of data, B o y d and
G r a h a m conducted hypothetical m e r g e r simulations
in which a risk-of-failure measure or statistic called a
z-score is used as the primary measure of risk. The results, while generally consistent with other studies,
yielded some significant differences. For example, with
stock market returns data the results indicate that mergers with property/casualty insurance companies would
r e d u c e b a n k risk, but the a c c o u n t i n g d a t a a n a l y s i s
yielded the opposite results. At the same time, both data sets indicated that diversification into life insurance
would reduce overall bank risk.
Richard J. Rosen and others (1989) focused on the
portfolio effect of bank diversification into real estate
investment over the period from 1980 to 1985. They
found that real estate investment trusts (REITs) had a
higher mean return on assets, reported returns negatively correlated with bank returns, and experienced far
more variable returns than bank returns. They found
that despite the higher returns and negative correlation,
investments greater than 6 percent of consolidated assets would yield a higher standard deviation of return
and coefficient of variation in return. Analyzing portfolios of banks and savings and loan real estate service
corporations, they failed to find any benefits to bank diversification into the service corporations.
Peter S. Rose (1989) examined both accounting and
stock m a r k e t r e t u r n s of a v a r i e t y of f i n a n c i a l and
nonfinancial firms during the 1966-85 period. His risk
measures included the mean, standard deviation, and
coefficient of variation of the various return measures
for firms in the different industries. Banks were found
to have a lower mean return but also a lower standard
deviation and a smaller coefficient of variation of returns than other industries. The correlation of the various

Economic Review

3

Management Influence
Although studies of the portfolio effect of banks entering currently impermissible nonbanking activities may
provide valuable insights into the relative riskiness of
various activities, these studies have only limited ability
to predict the actual consequences of product deregulation. The problem is that in deriving their statistical results portfolio studies assign (at least implicitly) a passive
role to bank managers and owners. In reality, the people
controlling banks would determine both the extent of
their organizations' entry into nonbanking activities and
the postmerger operational policies of their nonbank affiliates. Indeed, some operational changes in the nonbank
(and possibly bank) affiliates are almost inevitable.
Banks' passive portfolio investment in nonbanking activities is unlikely to generate significant synergies that will
make an acquisition valuable from the perspective of an
acquiring firm's shareholders.
Portfolio studies fail to consider managerial issues not
because researchers are unaware of their implication but
rather because such decisions cannot be accurately projected and there is little basis on which to impose related
assumptions on models. 1 The only way to know for certain how banks would use expanded powers is to deregulate bank activity. However, some insight may be gained
by reviewing the theoretical and empirical evidence on
banks' current risk-taking policies.
The theoretical case that banks may use expanded
powers to increase their riskiness is made by Boyd and
Graham (1986). 2 They noted that the fixed-rate deposit
insurance system that existed in 1986 created a moral
hazard problem. That is, the shareholders of banks that
invest in high-risk/high-return assets keep the gains if returns are high but if returns are significantly low are allowed to share losses with the FDIC. Brewer and Thomas
H. Mondschean (1991) provided evidence from insured
thrifts to support this hypothesis. They found that thrifts
increased their risk exposure by investing in junk bonds
and that the market rewarded the increase in risk through
higher stock returns.
A l t h o u g h d e p o s i t i n s u r a n c e m a y g i v e b a n k s an
incentive to take more risks, this incentive may be at least
partially offset by regulatory pressure and other considerations. For example, bank owners may not always gain
from an investment in high-risk assets. Banks make longterm investments in firm-specific capital (such as specialized software) that would be lost if the bank were to fail.
Furthermore, bank customers play an important role, and
many are willing to enter into long-term relationships
(such as interest rate swaps) with a bank only if they are
confident that the bank is not engaging in highly risky investments that could cause it to fail. Finally, risk also
threatens bank managers' loss of their firm-specific hu-

4
Economic Review


man capital along with their reputations as good managers if their bank fails.
Two studies suggest that shareholders and managers
do not generally prefer riskier policies. Robert A. Eisenbeis, Robert S. Harris, and Josef Lakonishok (1984)
studied the formation of one-bank holding companies
(OBHCs). O B H C activities, except investment banking,
were not regulated prior to the passage of the 1970
A m e n d m e n t s to the Bank Holding C o m p a n y Act. If
OBHCs provided a way to exploit deposit insurance to
gain value, then banks' announcements that they were
forming OBHCs should have uniformly increased shareholder wealth. However, Eisenbeis, Harris, and Lakonishok found that OBHCs generated unexpectedly large
returns only when bank holding company formation created an opportunity for geographic diversification. Thus,
contrary to the hypothesis that shareholders prefer increased risk exposure, in this case they appear to have favored reduced risk exposure. Most recently George J.
Benston, William C. Hunter, and Larry D. Wall (1993)
analyzed 302 U.S. bank mergers during the mid-1980s.
Their findings indicate that acquiring banks paid larger
acquisition prices for target banks that have a lower variance of return on assets.
Additional evidence concerning the ways in which
banks will actually use expanded powers is based on
observing the effect of nonbank affiliates on an organization's overall risk. This approach is less than ideal because nonbank affiliates are generally limited to performing activities that banks themselves could perform.
However, banks could increase risk by having these nonbank affiliates specialize in the riskier aspects of banking,
or they could reduce risk by exploiting affiliates as a
means of increasing the organization's geographic diversification.
Using accounting rates of return, Wall (1987) found
that the portfolio of nonbank subsidiaries at bank holding
companies are on average riskier than bank subsidiaries
but that the correlation of the two groups' return on equity is almost zero. When the bank and nonbank subsidiaries are combined into a single portfolio, on average
the level of risk is less than that of the banking subsidiaries by themselves. However, Wall also found an
economically significant minority of banks for which the
nonbank subsidiaries are risk-increasing. He then conducted tests to determine whether the nonbank affiliates
tended to accentuate risk (increase it at the riskier bank
holding companies) or moderate risk (decrease it at these
institutions). This test suggested that nonbank affiliates
were risk-moderating. Finally, in examining the riskiest
bank holding companies, Wall found that nonbank affiliates also tended to reduce somewhat these organizations'

September/October 1993

riskiness. Thus, he concluded that the evidence provided
no indication that bank holding companies would use
nonbank activities to take on additional risk and offered
weak support for the argument that greater diversification
would reduce risk.
Brewer (1989) examined the relationship between
bank holding company risk and their holdings of nonbank
affiliates using risk measures derived from stock returns.
He found some evidence that the proportion of bank holding company assets held in nonbank affiliates is inversely
related to bank holding company's riskiness. He then split
his sample into high- and low-risk bank holding companies. The nonbank activities were inversely related to the
riskiness of the high-risk bank holding companies, but
there was no significant relationship between nonbank activity and the riskiness of the less risky bank holding
companies. Brewer notes that his results tend to support
the case for expanded bank powers. However, he urges
caution in extending his results to new activities because
these activities may be riskier than banks' existing nonbank powers.
Brewer (1990) extended his previous study to examine
the effect of the composition of nonbank assets on bank
holding company risk as measured by stock returns. He
found that an increase in the proportion of bank holding
company assets invested in six of the seven nonbanking
activity categories that he defines are associated with reduced volatility of bank holding company returns, the one
exception being securities brokerage. He also found some
weak indications that most of the nonbanking activities
are associated with a reduction in the risk of failure. His

return m e a s u r e s with bank returns w a s generally positive but close to zero. R o s e also regressed cash flows
on various m a c r o e c o n o m i c a n d financial variables. H i s
f i n d i n g that t h e c a s h flow of a n u m b e r of i n d u s t r i e s
s h o w e d a w e a k e r c o r r e l a t i o n w i t h various m a c r o e c o n o m i c a n d f i n a n c i a l v a r i a b l e s t h a n w a s t r u e of c o m m e r c i a l b a n k i n g s u g g e s t s that diversification into
nontraditional p r o d u c t s c o u l d r e d u c e risk.
B o y d , G r a h a m , a n d R. S h a w n H e w i t t (1993) extend
the work of Boyd and G r a h a m by a d d i n g portfolio
w e i g h t s so that the n o n b a n k activities are a l l o w e d to
r a n g e f r o m 0 to a l m o s t 100 percent of the consolidated
entities' total assets. T h e y find that p a i r w i s e b a n k holding c o m p a n y m e r g e r s with life a n d p r o p e r t y / c a s u a l t y
i n s u r a n c e firms are m o s t likely to g e n e r a t e diversification gains. M e r g e r s with securities and real estate firms
achieved m a x i m u m risk reductions at e c o n o m i c a l l y insignificant levels of n o n b a n k activities (less than 5 percent of the p o s t m e r g e r f i r m ' s p r o f o r m a assets). 5

Federal Reserve Bank of Atlanta



results suggest that the riskiness of nonbank activities
should not be evaluated independent of the bank holding
companies' entire portfolio of assets.
Thus, an important element in evaluating proposals for
deregulation is assessing the goals of bank owners and
managers if they were to be permitted investment in riskier activities. Banks may use diversification as a method
for increasing or decreasing risk. Historically, deposit insurance has created an incentive for banks to take on
risk. 3 However, these incentives may be offset at many
banks by regulatory pressure, managerial risk aversion, or
the profitability of being able to make long-lived investments. The existing empirical literature does not support
the hypothesis that b a n k s have used their expansion
power to increase their overall risk. Hence, the evidence
suggests that bank holding company shareholders and
managers do not always prefer riskier policies.
Notes
1. The alternative of allowing researchers to choose their
own assumptions about owners and managers' behavior
would compromise the objectivity of portfolio studies. Researchers could find evidence supporting the position of
their choice if they were free to use whatever assumption
they wanted about banks' behavior in a deregulated environment.
2. See Kane (1985) for a broader discussion of the incentives
created by the deposit insurance system.
3. However, recent legislative changes may have reduced the
incentives to take risk that were created by deposit insurance. These changes are discussed in the box on page 8.

Evaluation of Prior Studies
T h e studies discussed above and others have app r o a c h e d a n a l y z i n g the e f f e c t s o n portfolios of diversification into n o n b a n k activities in a variety of w a y s . T h e
m e t h o d o l o g i e s used e a c h have significant strengths and
w e a k n e s s e s that it is i m p o r t a n t to u n d e r s t a n d in evaluating the current state of the literature.
M e a s u r e m e n t of Portfolio P e r f o r m a n c e . M o s t
studies of the e f f e c t of p o r t f o l i o d i v e r s i f i c a t i o n f o c u s
on o n e of t w o risk m e a s u r e s : the c o e f f i c i e n t of variation of s o m e return m e a s u r e or the risk of failure calculated using a c c o u n t i n g or m a r k e t data. T h e coefficient
of variation is, as n o t e d above, m e r e l y the variability of
r e t u r n s ( s t a n d a r d deviation of r e t u r n s ) d i v i d e d by the
expected return. T h e risk of failure incorporates a firm's
equity capital, its e x p e c t e d returns, a n d standard deviation of returns to p r o v i d e a m e a s u r e of the likelihood

Economic

Review

5

that a firm will experience losses that exceed its capital.
Risk of failure is a more direct measure of the primary
regulatory concern: Would increased participation in
nonbank activities make banks more or less likely to
fail? However, implicit in the risk-of-failure measure is
the assumption that the combined organization's capital
structure will be the sum of their individual premerger
capital structures, an assumption that may not be appropriate if regulators require higher postmerger capital
levels. Further, calculation of the risk of failure requires
data on premerger capital structures that may not be
available from some data sources.
A further consideration in evaluating portfolio performance is the perspective of bank owners and managers. Most studies focus on risk issues because that is
the regulator's concern. Bank owners and managers,
however, actually undertake mergers on the basis of the
effect of diversification on both the return and risk of
the c o m b i n e d organization. B a n k s m a y e n g a g e in a
risk-reducing merger if the reduction in their expected
return is not too large, but they may also be willing to
undertake higher risk if the increase in expected return
is sufficiently large. 6 Thus, a full analysis of the effect
of diversification on returns must consider both the
banks' and the regulators' perspectives. 7
F o r m a t i o n of Portfolios. T h e various studies of
bank mergers take three different approaches to forming
the portfolios for analysis. Some studies limit their analysis to three or four statistics: industry-average mean returns, industry-average standard deviations of returns,
industry-average coefficients of variation of returns on
assets, and the correlation of industry returns with banking returns. Looking at industry statistics alone does not
allow an easy determination of the change in risk that
results from combining different industries into a single
firm. For example, an industry might have a higher
standard deviation of returns than banking, but the returns may be negatively correlated with banks' returns.
Thus, it is not always clear whether the higher standard
deviation of a particular firm from this industry combined with a banking firm will increase the risk to the
postmerger organization or its negative correlation with
banking will generate less risk.
An alternative to using overall industry statistics is
to combine industries in pairs—banks and one nonbank
industry at a time. This approach provides for simultaneously considering the effects of expected return, the
standard deviation of returns, and the correlation between returns (as well as the capital positions of the
two firms, when appropriate). Perhaps most importantly, this approach has the advantage of corresponding
with actual bank behavior. Because firms typically en-

6
Economic Review


gage in one merger at a time, the concern to banks and
their regulators at any given point in time is the desirability of a particular pairwise combination.
T h e third alternative in examining portfolios is to
analyze efficient portfolios of banks and several nonbank industries. As discussed above, the term efficient
portfolio refers to one whose combinations produce the
most return for any given level of return variability (or,
equivalently, the least return variability for any given
return). These portfolios may contain firms operating
in only two industries (or in some cases a single industry). However, as Litan found, some efficient portfolios
are likely to contain multiple industries. E x a m i n i n g
portfolios of unique service products is advantageous
because it is the approach that banks should take from
a portfolio risk and return perspective. Thus, basing
public policy solely on the risk effects of pairwise mergers may impose significant social costs if it results in
policies that prevent the formation of efficient portfolios of bank and nonbank firms.
T i m i n g of A g g r e g a t i o n to I n d u s t r y Level. T h e
various studies take two approaches to the aggregation
of firm data into industry statistics. Some studies combine individual firms into a single industry before conducting any analysis, and others calculate the mean and
variability of returns for individual firms (and across
pairs of firms) and then aggregate the figures across all
firms in the industry (or in the pair of industries). The
major disadvantage of the first approach is that individual firms enter into mergers with specific firms, not
with broad industries. On the other hand, industry aggregate figures may be a better proxy for the expected
future distribution of returns to the extent that two conditions hold—that is, if within-industry differences primarily arise from regional economic conditions and if
firms within the industry are combining across regions.
Another advantage of using industry aggregates is that
spurious results in the formation of portfolios may be
less likely. An efficient portfolio is formed by looking
at an individual entity's "assets" to determine the combinations that produce an efficient set of portfolios.
These assets m a y be defined as entire industries or as
individual firms within industries. Obviously, the number of separate assets for inclusion in an efficient portfolio will increase dramatically if individual firm returns
are used rather than industry returns. In general, an increase in the number of assets is likely to increase the
c h a n c e s of i d e n t i f y i n g lower-risk p o r t f o l i o s . T h u s ,
Boyd, Graham, and Hewitt argue that random chance
favors the possibility that a risk-reducing portfolio will
be found using individual firm data even if there is not
a real opportunity for diversification to reduce risk.

September/October 1993

Use of Market or Accounting Data. Banks' and
n o n b a n k f i r m s ' returns m a y be m e a s u r e d using accounting or financial market data. The drawback to usi n g a c c o u n t i n g d a t a is that they are not p e r f e c t l y
correlated with e c o n o m i c returns. Firms often try to
smooth accounting data through time, producing reported returns that are deliberately low in the good
years and high in the bad years. If firms across different industries have unequal ability to s m o o t h their
accounting earnings, then accounting-based risk measures may not provide accurate interindustry comparisons of risk. Using accounting data has some appeal,
however. First, market data is typically available only
for the largest firms in an industry, so it clearly is more
limited than accounting data. In addition, regulators rely heavily on accounting figures in their evaluation of a
bank's financial condition.
Studies that rely on accounting data use two sources
of information: accounting data from the individual
firm's public financial statements prepared according
to generally accepted accounting principles ( G A A P )
and accounting data published by the Internal Revenue
Service for all firms in an industry prepared according
to IRS accounting rules. Each data set has its advantages. G A A P rules are intended to fairly present a firm's
performance over time, whereas IRS rules also reflect a
number of public policy decisions. For example, to encourage banks to hold state and local government obligations, IRS rules allowed banks to understate their
income by excluding the interest f r o m holding these
obligations. 8 Another advantage of using G A A P data is
that they are available at the individual firm level, and
IRS data are available only for an entire industry. On
the other hand, IRS data reflect a broad cross-section of
firms in an industry while public financial statements
are only available for the largest firms.
Overall Evaluation of Prior Studies. As the above
discussion suggests, there appears to be no single "correct" methodology. Each has advantages and disadvantages. Ideally, the different approaches would produce
consistent results confirming that individual findings
were not the result of a unique methodology. The majority of the studies do in fact seem to reach a consensus: p a i r w i s e c o m b i n a t i o n s of b a n k s and n o n b a n k s
seldom significantly reduce banking organizations' risk
exposure. The best opportunity for diversification gains
a p p e a r s to b e b a n k m e r g e r s with f i r m s e n g a g e d in
some aspect of the insurance industry.
The strongest challenge to the emerging consensus
in the literature comes from Litan's analysis of efficient
portfolios of multiple industries, in which he finds significant gains from combining banks with a number of

Federal Reserve Bank of Atlanta




other industries. Moreover, Litan's findings that there
may be more diversification benefits is consistent with
portfolio theory, which suggests that combinations of
multiple assets may be more efficient than any single
asset or unique pair of t w o assets. T h u s , his results
should not be easily dismissed.

iVew Research
T h e study reported on here provides a bridge between Litan's findings and those of other studies in two
important respects. First, this analysis is the first to analyze both pairwise combinations and efficient portfolio formation using the same data set and time period.
Doing so makes it possible to evaluate the implications
of the two methodologies. Second, this discussion updates Litan's IRS data set. His data set ends in 1981
while that of the present study extends to 1989. This
timeframe allows comparison of the portfolio gains of
bank entry into nonbanking activities in the period after
deregulation and with gains prior to deregulation. In
addition, the use of IRS data p e r m i t s a c o m p a r i s o n
with Litan. However, there are at least three limitations
to the work: First, there are the limitations inherent in
using IRS accounting data rather than market data or
G A A P accounting figures, described above. Further,
the IRS data are aggregated to the industry level before
the analysis. Finally, the coefficient of variation of return is analyzed but not the risk of failure b e c a u s e
the IRS data do not provide information on equity capital.

Data Source a n d Methodology
The data used in this study were taken f r o m the Internal Revenue Service publication Corporate
Income
Tax Returns for the major sector entitled "Finance, Insurance, and Real Estate" (Major Group 60) for selected years f r o m 1971 to 1989. Although the publication
contains a n u m b e r of data series for different industries, only t w o variables were consistently available
over the entire data period: total assets and net inc o m e . The publication reports data in t w o columns,
one labeled "Net I n c o m e " and the other "Deficit." The
column for net income shows total net income, as previously d e f i n e d , f o r c o r p o r a t i o n s reporting positive
earnings for the year. T h e deficit c o l u m n lists total
losses incurred by corporations reporting losses for the

Economic Review

7

Containing Risks
Suppose that banks are given authority to enter currently
impermissible activities and one of these new nonbank affiliates experienced severe financial problems. Could the problems in a nonbank affiliate be contained so that they do not
generate substantial social costs? If the answer is positive,
then advocates of deregulation would argue that risk considerations should not limit bank holding company expansion
into nonbank activities; bank holding companies should be
allowed to expand regardless of whether the new activities
will increase or reduce risk.
The primary social costs of a banking organization's failure are its potential for creating systemic risk and the cost to
the FDIC and the Fed (through its discount window) of
protecting the banks' creditors. Systemic risk arises if the
failure of one or more banks impairs the operation of the financial system to such a degree that the real (nonfinancial)
economy is adversely affected to a significant degree. The
failure of individual institutions may propagate through the
banking system is a variety of ways, such as contagious
bank runs, credit losses on interbank liabilities that cause
other banks to become insolvent, and the failure of major
payments networks to settle.
The solution to systemic risk situations since the 1930s
has been for the government to provide a safety net for commercial banks: the Federal Reserve provides liquidity to
problem institutions through the discount window while the
FDIC absorbs losses as a result of having provided deposit
insurance. The combined actions of the Fed and the FDIC
apparently have prevented bank failures from generating
systemic risk. However, this solution has in effect socialized
bank losses, which in turn creates two social costs. First, in
this framework the government has a direct financial stake
in preventing banks from becoming highly insolvent and
then failing. The economic and political costs of a bailout
can be high. Second, FDIC coverage of losses means that
private owners keep all the gains if their gamble is successful, but the government shares in the losses if a bank fails.
As noted in the box on page 4, most banks have good reasons for not taking excessive risk to exploit the government
safety net. However, the thrift debacle has shown that it only takes a few institutions implementing high-risk policies to
create enormous deposit insurance losses.
This box begins with a review of the issues associated
with containing the riskiness of nonbank activities in the
period preceding the FDIC Improvement Act of 1991 (FDICIA). An understanding of the issues during that time is
important in part because most discussions of this topic
were written prior to FDICIA. Moreover, virtually all empirical studies of risk associated with deregulation rely on
pre-FDICIA data, and having a clear picture of the preFDICIA rules is valuable in interpreting these studies' results. The box continues with a review of the changes put in
place by FDICIA. These new rules may significantly reduce

8
Economic Review


the potential social costs associated with the failure of a
nonbank affiliate. 1

Pre-FDICIA Rules
The failure of a nonbank subsidiary could pose both a direct and an indirect threat to its bank holding company bank
affiliates. 2 The most direct threat to the bank affiliates
would occur if the creditors of the nonbank affiliate could
"pierce the corporate veil" and usurp bank resources to satisfy their claims on the nonbank affiliate. 3 However, most
analysts have suggested that this threat is remote provided
that the bank holding company and its affiliates followed
correct legal operating procedures.
Another direct threat to the bank affiliates was that the
bank holding company's management would divert bank resources to prevent the failure of a nonbank affiliate, bank
holding company management might seek to prevent the
failure of a nonbank affiliate to preserve its organization's
credibility in the financial markets. If a bank holding company were to abandon a failing subsidiary, its remaining affiliates would experience a higher cost of funds and reduced
market clout.
Regulators could use a number of tools to prevent bank
holding companies from withdrawing excessive funds from
their bank affiliates. The dividend payments of most banks
were limited, with a common restriction being that a bank
might not pay more than its current year's earnings plus the
sum of the previous two years' additions to retained earnings without prior regulatory approval. Section 23A of the
Federal Reserve Act limited the diversion of funds through
the mispricing of interaffiliate transactions by requiring that
all interaffiliate transactions occur at "arms' length" prices.
The regulators could also use capital adequacy standards to
prevent banks from reducing their capital to unsafe levels.
Finally, the regulators had the general power to order banks
to cease any banking practice deemed unsafe or unsound.
The methods by which nonbank affiliates might directly
threaten the bank holding company were therefore controlled. However, nonbank problems could pose indirect
threats harder to block. For example, the bank holding
company may order its bank affiliates to invest in highreturn/high-risk assets in an attempt to boost earnings and
thereby increase dividends. If the bank regulators perceived
such an action by the bank holding company a number of
supervisory tools were available to prevent the bank affiliates from becoming riskier.
Another indirect way in which the banking affiliates
could be threatened by a nonbank affiliate's problems lies in
the fact that bank depositors who confuse the identity of the
bank and nonbank affiliates may demand withdrawal of their
funds upon hearing of the problem. Further, even more sophisticated depositors may withdraw funds if they interpret
the problems in the nonbank affiliate as a measure of the

September/October 1993

quality of the banking affiliates' management. The Federal
Reserve could prevent a run on a solvent bank from creating
a systemic risk situation by lending to all affected banks
through the discount window.
Thus, in pre-FDICIA days banks were substantially but
not totally isolated from their nonbank affiliates. The regulators could slow the withdrawal of bank capital and prevent
obvious attempts to exploit the bank affiliates in order to
save nonbank operations. However, they could not guarantee that a determined and ingenious bank holding company
management would not exploit the bank nor could they prevent nonbank affiliate problems from causing a ran at their
bank affiliates.

A possible concern regarding least costly resolution is
that the policy may actually encourage deposit runs at banks
whose nonbank affiliates are having financial problems. The
threat of these runs may be beneficial before a bank makes
it investment decision in that it could discourage the bank
holding company from following high-risk policies with its
nonbank affiliates. However, once a nonbank affiliate has
encountered problems, regulators may need to take action to
prevent a bank run from becoming a systemic crisis. Although FDICIA imposes some limitations on the Fed's ability to lend to banks, the Fed still has the power to lend to
viable banks and can do so as needed to prevent problems a1
a nonbank affiliate from turning into a systemic crisis.

Post-FDICIA

Conclusion

FDICIA contains a number of provisions to limit the
government's deposit insurance exposure. 4 Probably the two
most important provisions are those for prompt corrective
action (PCA) and for least costly resolution. The PCA provision instructs regulators to manage problem-bank situations in a manner that minimizes the cost to the FDIC. The
least costly resolution provisions generally require that failed
banks be resolved in the least costly manner to the FDIC. As
a practical matter, the latter is intended to tightly limit the
number of cases in which the FDIC has to protect uninsured
depositors and other creditors from losses.
The PCA provision specifies five levels of capital adequacy, with regulatory restrictions increasing as a bank's
capital level drops. The intent of these guidelines is to discourage banks from taking excessive risks and to address
problems before they threaten a bank's viability. However,
if a bank's viability is seriously jeopardized, PCA provides
for early bank closure before the bank's accounting net
worth is depleted.
In the study that laid the basis for PCA, Benston and
George G. Kaufman (1988) presented a proposal designed
to eliminate totally the risk of loss to the FDIC. Under their
plan banks that would be classified under PCA as well capitalized would be allowed to engage in currently impermissible activities without exposing the FDIC to significant risk.
Although Congress adopted most of the Benston and Kaufman plan in FDICIA, legislators did not adopt their recommendation that banks be provided increased benefits or incentives that would encourage them to seek "well-capitalized"
status.
The restrictions providing for least costly resolution are
important because they encourage market discipline on the
part of bank creditors. Least costly resolution reduces the
regulator's discretion in providing de facto deposit insurance
coverage, thereby encouraging large depositors to evaluate
their banks more carefully before making a deposit and providing strong incentives to withdraw funds before their bank
fails. The policy of least costly resolution may reinforce the
early-closure provisions of PCA and help force the closure of
failing banks before they can create substantial FDIC losses.

Before FDICIA a number of limitations were placed on
banking organizations to protect bank affiliates from nonbanking problems. These limitations were unlikely to provide total protection but could be expected to slow the
spread of the problem. By enhancing the regulators' ability
to aggressively address problems in the banking subsidiaries and by increasing market discipline, FDICIA may
substantially reduce the implicit value of the subsidy provided by deposit insurance to high-risk banking organizations. The reduction in the deposit insurance subsidy should
discourage banking organizations from taking excessive
risks through nonbank affiliates and hence reduce the probability that the problems of the nonbank affiliates will generate significant losses to the FDIC. The combination of the
pre-FDICIA rules plus FDICIA itself should significantly
reduce the risks to banks and the FDIC from nonbank activities. It is important to keep in mind, however, that regulation that is strict enough to eliminate risk totally may
significantly limit the very synergies financial firms seek in
combining bank and nonbank activities.

Federal Reserve Bank of Atlanta



Notes
1.The analysis in this section presumes that systemic risk
problems would arise only if a nonbank subsidiary created
problems for its bank affiliates. If the failure of the nonbank
subsidiary could generate systemic risk, then that problem
would deserve separate analysis. However, if the failure of a
nonbank subsidiary could cause a systemic problem, then the
failure of another large but independent firm in the same industry may also give rise to systemic risk, and an analysis of
the entire industry would be justified.
2. See Wall (1984) for a more detailed discussion of the preFDICIA environment.
3. The creditors of one corporation ordinarily lack the power to
take the assets of another corporation to satisfy the creditors'
claims, even if the two corporations are affiliated. However,
under certain circumstances the courts will let creditors
"pierce the veil" and take assets from affiliated corporations.
4. See Camell (1992), Pike and Thomson (1992), Todd (1992),
and Wall (1993).

Economic

Review

9

year. To provide a comprehensive picture of industry
performance, this study calculated aggregate industry
profits f o r the year by subtracting the deficit figure
from the net income figure. The appendix identifies the
types of firms that make up the various industries. A
portfolio optimization software package entitled "Asset
Allocation P a c k a g e " (Release 1.1), written by Steve
Gladin and Robert Radcliffe, was used to estimate efficient bank and nonbank portfolios. Although the program can calculate efficient frontiers for as many as
fifty securities or asset classes, the presence of highly
correlated returns in the IRS data limited the number of
asset classes to approximately twelve.

Returns, Volatility, and Correlations
with Bank Holding Companies
The IRS divides the broad category of finance, insurance, and real estate into several industry groups,
which are further broken down into narrower industries. Each of the broad industry groups is presented in
Table 1 with their respective mean returns on assets,
coefficient of variation of return on assets (volatility),
and the correlation of return on assets between each industry and the bank holding company industry category.
The results are further broken down into the prederegulation period, from 1974 to 1980, and the deregulation
period, from 1981 to 1989.

and commodity brokers and services, and (2) holding
and other investment companies. 1 0
Analyzing the stability of the various risk and return
measures over the two subperiods suggests that banking became riskier in the second subperiod, but so did
most of the other industries. Although the relative rankings in the different categories are generally stable over
time, some significant changes did occur. Perhaps the
most notable are in the con-elations with bank holding companies, where some correlations changed signs across the
two subperiods (but never from being significant with
one sign to being significant with the opposite sign).
The mean return on assets, coefficient of variation
of return on assets, and correlation of returns on assets
with bank holding companies for the narrow industry
definitions are presented in Table 2 for the subperiod
from 1974 to 1980. The narrow bank holding company
category, like the broad banking category, has a low return on assets and a low variability of returns. Assuming a 5 p e r c e n t e q u i t y - t o - a s s e t ratio, b a n k h o l d i n g
companies earned a return on equity of less than 5.5
percent. To the extent that IRS income accurately measures economic returns, this rate of return to bank holding company owners is clearly inadequate.
Seven industries seem to have presented good diversification opportunities with bank holding companies
between 1974 and 1980 in that they have both higher
returns on assets and lower coefficients of variation:
other banks, business credit institutions, life insurance,
insurance agents and brokers, lessors of mining and oil
properties, regulated investment companies, and other
h o l d i n g and investment c o m p a n i e s . F u r t h e r m o r e , a
number of these industries reported earnings that were
negatively correlated with banking.

T h e b r o a d c a t e g o r y of b a n k i n g ( w h i c h i n c l u d e s
bank holding companies, mutual savings banks, and
other banks) has one of the lowest returns on assets of
the seven industry groups in both subperiods (0.29 and
0.24, respectively). If it is assumed that banks typically
maintained a 5 percent equity capital-to-assets ratio
during this period, then banking would be yielding only a 5 percent to 6 percent return on equity to its shareholders over the two subperiods. 9 On the other hand,
the industry has one of the lowest coefficients of variation of return on assets despite its low return on assets
indicating minimal variability in returns. However, the
insurance agents, brokers, and service industry and the
h o l d i n g and other investment c o m p a n i e s industries
both showed higher returns on assets and lower coefficients of variation during the 1981-89 subperiod.

Table 3 presents similar statistics for the 1981-89
subperiod. Bank holding companies' return on assets
dropped and their coefficients of variation increased,
but in both cases their ranking improved. Only regulated investment companies and other holding and investment companies had both a better average return on
assets and lower coefficient of variation during this
subperiod. However, eight industries had negative correlations (not all of which are statistically significant)
with bank holding companies, raising the possibility
that bank holding c o m p a n y diversification into these
industries could result in a lower variability of returns.

Given that b a n k h o l d i n g c o m p a n i e s are a m a j o r
component of the broad banking industry category, it is
not surprising that the banking industry is the industry
category most highly correlated with these institutions.
Two industries have significant negative correlations
with banking during the 1981-89 period: (1) securities,

Only three of the industries improved their returns
on assets during the 1981-89 subperiod: condominium
management and co-ops, regulated investment companies, and real estate investment trusts. Similarly, only
three industries reduced their coefficients of variation
of return on assets in the latter subperiod: commodity

10

Economic Review-




September/October 1993

brokers/dealers, regulated investment companies, and
real estate investment trusts. The rankings of mean return on assets and return on assets variability in the later subperiod are positively correlated with those in the
first subperiod, but some significant shifts occurred.
The signs of the bank holding company industry correlations with other industries tend to be consistent over
time but changed in a few industries.

Pairwise Portfolio Formation
The analysis presented above of individual industry
returns and their correlation with bank holding companies suggests potentially beneficial c o m b i n a t i o n s .
However, the more definitive way to analyze potential
pairwise combinations is to examine actual hypothetical combinations, as in Table 4. The approach follows
that of Boyd, Graham, and Hewitt in examining combinations of bank holding c o m p a n i e s and o n e other
industry using assets f r o m the two industries in combination in various percentages.

There are only two clear patterns across both subper i o d s : In the b r o a d c a t e g o r y of other h o l d i n g and
investment companies, performance of the various industries improved relative to all other industries, and
returns to industries in that category became less correlated with b a n k i n g . O n e m a y s p e c u l a t e that these
c h a n g e s resulted less directly f r o m deregulation in
these industries than from banks' continuing inability
to make a substantial entry into other industries. While
the financial sector was moving from deposit-based to
securitized financing, banks were unable to follow their
customers adequately.

Evaluating the results in Table 4 requires a standard
for judging the extent to which bank holding companies should be allowed to diversify into other activities.
One of several standards that seem reasonable is that
bank holding companies should be allowed to diversify
to the extent that doing so minimizes their risk as measured by the coefficient of variation in return on assets.

Table 1
Industry Volatility and C o r r e l a t i o n Analysis' 1
Mean R O A
Industry

Period

Value ( % )

Volatility

Rank1'

Value ( % )

Corr. w / B H C s
Rank 0

Value

Rank d

Banking

(a) 1974-80

.29

(Broad category)

7

25.2

2

(b) 1981-89

.24

.87*

5

6

65.3

3

.96*

7

Credit Agencies

(a) 1974-80

(Other than banks)

.35

6

89.2

7

(b) 1981-89

.27

-.40

7

2

163.7

6

.68*

6

Insurance

(a) 1974-80

1.48

4

(Broad category)

36.6

(b) 1981-89

3

.61

3

.85*

78.6

5

5

.66

5

Insurance Agents,

(a) 1974-80

7.15

1

Brokers, Service

21.8

(b) 1981-89

3.74

1

2

.76*

4

54.5

2

.63

4

Real Estate

(a) 1974-80

1.61

3

(Broad category)

63.3

6

(b) 1981-89

.20

6

237.5

6

-.43

Security, Commodity

(a) 1974-80

1.29

5

Brokers and Services

52.1

5

(b) 1981-89

.59

4

-.23

69.3

1

4

-.70*

2

Holding and Other

(a) 1974-80

4.66

2

Investment Companies

37.6

4

(b) 1981-89

6.40

1

.60

19.2

1

3

-.90*

1

a
b
c
d

Higher-ranking activities indicate a more desirable merger partner for bank holding
Industries with higher ROAs receive a higher ranking.
Industries with lower earnings volatility are ranked higher.

.94*

7
3

companies.

Industries with large negative correlations with bank holding companies are ranked higher.
Indicates that the correlation coefficient is statistically significant at the 5 percent level or better.

Federal Reserve Bank of Atlanta



Economic Review

11

Table 2
Banking Industry Volatility and Correlation Analysis for 1 9 7 4 - 8 0
Mean R O A
Industry

Value ( % )

Volatility
Rank

Value ( % )

Corr. w / B H C s
Rank

Value

Rank

Bank Holding Companies

.27

19

30.6

8

Mutual Savings Banks

.20

21

131.4

19

.16

7

Other Banks (Not M S B s or B H C s )

.34

17

22.3

3

.62

14

Savings and Loans

.26

20

116.0

17

.23

8

Personal Credit Institution

1.34

12

58.9

13

.23

9

Business Credit Institution

1.47

11

29.3

7

.69

15

.29

18

100.6

15

.47

11

Life Insurance

1.56

10

12.5

1

.52

13

Mutual Insurance (except life)

1.29

13

144.0

20

.76"

17

Other Insurance

1.21

14

126.0

18

.83*

20

Insurance Agents, Brokers, Service

7.51

2

21.8

2

.76*

18

Real Estate Operators

1.85

6

34.2

10

-.94"

1

10.93

1

26.4

5

-.20

4

1.67

8

31.7

9

-.16

5

-1.67

23

48.5

11

-.45

3

.90

16

220.2

22

.96*

22

Other Real Estate

1.79

7

109.0

16

.71

16

Security Brokers, Dealers

1.20

15

58.9

12

-.08

6

Commodity Brokers/Dealers

2.27

5

92.4

14

-.54

2

Regulated Investment Companies

5.90

3

28.1

6

.51

12

Real Estate Investment Trusts

-.28

22

799.0

23

.43

10

Small Business Investment

1.61

9

170.3

21

.84*

21

Other Holding and Investment
Companies

3.71

4

25.4

4

.81*

19

Other Credit Agencies

and Lessors of Buildings
Lessors of Mining, O i l Properties
Lessors of Railroad Properties
C o n d o m i n i u m Management
and Co-ops
S u b d i v i d e s and Developers

'Indicates

that the correlation

coefficient

12
Economic Review


is statistically

significant

at the 5 percent

1.00*

23

level or better.

September/October 1993

Table 3
Banking Industry Volatility and Correlation Analysis for 1 9 8 1 - 8 9

Value ( % )

Industry

Corr. w / B H C s

Volatility

Mean R O A
Rank

Value ( % )

Rank

Value

Rank

Bank Holding Companies

.26

15

34.4

3

1.00*

23

Mutual Savings Banks

.05

17

1,884.1

23

.79*

19

Other Banks (Not M S B s or B H C s )

.27

14

79.2

10

.95"

22

-.60

22

128.8

15

.56

15

Personal Credit Institution

.84

9

104.5

14

.38

11

Business Credit Institution

-.04

18

1,248.9

21

.48

13

Other Credit Agencies

.22

16

200.5

17

.91*

21

Life Insurance

.60

11

83.0

11

.31

10

Mutual Insurance (except life)

.39

13

340.1

18

.60

16

Other Insurance

.62

10

155.2

16

.66

18

Insurance Agents, Brokers, Service

3.73

3

54.5

8

.63

17

Real Estate Operators

1.10

8

64.0

9

-.86*

2

Lessors of Mining, O i l Properties

8.32

1

48.5

5

-.65

4

Lessors of Railroad Properties

1.53

7

53.8

7

-.30

6

-1.22

23

98.7

13

.55

14

S u b d i v i d e s and Developers

-.28

21

406.3

19

.84*

20

Other Real Estate

-.09

19

903.5

20

.41

12

.52

12

84.2

12

-.71*

3

Commodity Brokers/Dealers

1.78

5

51.6

6

.12

9

Regulated Investment Companies

7.63

2

22.6

1

-.93*

1

Real Estate Investment Trusts

3.34

4

41.2

4

-.22

7

Small Business Investment

-.10

20

1,687.4

22

-.33

5

Other Holding and Investment
Companies

1.63

6

31.9

2

-.04

8

Savings and Loans

and Lessors of Buildings

Condominium Management
and Co-ops

Security Brokers, Dealers

"Indicates

that the correlation

coefficient

Federal Reserve Bank of Atlanta



is statistically

significant

at the 5 percent

level or better.

Economic

Review

13

Table 4
I m p a c t of D i v e r s i f i c a t i o n o n Bank R O A
P a i r w i s e Portfolios, 1 9 8 1 - 8 9
(Percent of
Industry
Combination

5

10 Percent

Percent

Diversification)

25 Percent

50 Percent

Average

Coefficient

ROA

of Variation

Mutual Savings Banks

.25

51.5

.24

71.7

.20

147.7

.16

Independent Banks

.26

36.4

.26

38.6

.26

45.1

.26

(BHCs combined with)

Average

Coefficient

Average

Coefficient

Average

Coefficient

ROA

of Variation

ROA

of Variation

ROA

of Variation

75 Percent
Average

Coefficient

ROA

of Variation

344.6

.11

56.4

.26

245.3

90 Percent
Average

Coefficient

ROA

of Variation

736.1

.07

1,233.1

67.8

.27

74.7

-.38

154.2

-.51

145.2

.70

96.0

.79

101.5

.04

965.7

-.007

5,699.2

.23

152.6

.22

180.8

Banking

Credit Agencies
Savings and Loans

.22

51.0

.17

79.2

.04

503.3

.17

Personal Credit
Institutions

.29

37.9

.32

44.1

.41

62.5

.55

Business Credit
Institutions

.24

40.1

.23

48.0

.19

85.9

.11

Other Credit Agencies

.26

40.0

.26

47.6

.25

Insurance

.24

83.3
230.1
108.9

Life Insurance

.28

34.5

.30

36.3

.34

46.0

Mutual Insurance
(except life)

.27

50.7

.43

61.9

.52

.27

74.0

.56

79.7

69.9

.29

127.5

.32

211.6

.35

281.4

.37

317.8

Other Insurance

.28

44.0

.30

54.5

.35

82.8

Insurance Agents,
Brokers, Service

.43

38.9

116.5

.53

139.1

.58

43.1

149.3

.61

1.13

49.0

2.00

52.4

2.87

53.7

3.39

54.2

Real Estate Operators
and Lessors of Buildings

.30

19.1

.34

11.8

.47

26.0

.67

46.1

57.1

1.01

61.6

Lessors of Mining,
Oil Properties

.66

24.1

1.06

33.4

2.27

42.5

4.30

46.3

6.30

47.8

7.52

48.2

Lessors of Railroad
Properties

.32

25.5

.39

24.9

.58

34.0

.90

44.8

1.21

50.4

1.40

52.7

Condominium
Management and Co-ops

.18

69.1

.11

159.5

-.11

309.7

-.48

130.6

-.85

107.7

-1.08

101.6

Subdividers and
Developers

.23

58.0

.21

89.5

.13

267.9

-.01

7,812.1

-.14

606.3

-.22

457.3

Other Real Estate

.24

44.0

.23

58.9

.17

132.6

470.1

.00

9,999.0

-.05

1,367.7

Real Estate




lo CT;
C Ö
C

This standard would be appropriate if the only advantage of bank holding company involvement in nonbank
activities is the potential for portfolio risk reduction.
For example, bank holding companies could invest as
much as 10 percent in securities brokers/dealers because such a portfolio has the lowest c o e f f i c i e n t of
variation of any portfolio of bank holding companies
and securities brokers/dealers." Limited to those industries for which nonbank activities reduce risk, the
analysis finds four nonbank holding c o m p a n y industries that are risk-minimizing at 5 percent and four that
are risk-minimizing at 10 percent. 1 2
Another reasonable standard is that bank holding
companies should be allowed to diversify as long as the
nonbank activity does not increase the coefficient of
variation above that of b a n k h o l d i n g c o m p a n i e s by
themselves. This standard is based on the idea that
bank holding companies should be allowed to diversify
as long as the diversification does not increase their
portfolio risk. For example, a bank holding company
would be able to invest as much as 25 percent of its
portfolio in lessors of railroad properties because the
coefficient of variation at that level, 34.0 percent, is
less than that of a portfolio of 100 percent bank holding
company assets—34.4 percent. 1 3 In this case the maxim u m permitted diversification a c c o r d i n g to Table 4
would involve two industries at the 10 percent level,
four industries at the 25 percent level, and two industries at the 90 percent level. 14

o-i co
O <-'

«

-a

o
E
E
o
U
3

<
J

OJ

Jo
n
s
0)
Q

>

,
tC l
c

"E O
)
a

y
41
E
Jü
-C a. in
to
E aj
a; c o
D C
o o >
O
£ c i
-a U
-a o ¿r
e c
e ¿5
C
U
PS QJ "O C Sic
a a 3 U
3
>O ¡s> ©c E S ' E tS E> ff C
E
LU ¿3
E fc
:5
O
l
ß
& "rä <
O u o <
o > odE 15 >
L
>
Ol o 0J
m
c DiU
Jr <S) U Ü
X
ctE

Federal Reserve Bank of Atlanta




.D
C

C

"O ro
sz a.
re E
fcS

J iß

5 1

A third standard would permit diversification to the
extent that it does not increase bank holding companies' risk above that of currently permitted activities.
T h i s w o u l d be a r e a s o n a b l e standard if it is agreed
that other specific industries in the c a t e g o r y of finance, real estate, and insurance should be subject to
n o stricter s t a n d a r d s than are a p p l i e d to activities
closely related to banking. A fair comparison may be
with " o t h e r " banks, a group consisting largely of ind e p e n d e n t b a n k s not a f f i l i a t e d w i t h b a n k h o l d i n g
c o m p a n i e s . T h e s e o r g a n i z a t i o n s are insured by the
F D I C whereas in this study's sample only the banking
affiliates of the bank holding companies are explicitly
covered by deposit insurance. This industry had a coefficient of variation of return on assets of 79.2 for the
1981-89 subperiod. According to this standard, bank
holding companies should be allowed to enter every
industry and to have as much as 90 percent of their
assets in the n o n b a n k holding c o m p a n y industry in
ten industries. 1 5
The analysis of pairwise combinations generates results that favor deregulating bank activities more than
do those of Boyd, Graham, and Hewitt. The findings

Economic Review

15

likely stem f r o m three m e t h o d o l o g i c a l d i f f e r e n c e s :
(1) the use of IRS accounting data aggregated to the industry data rather than the f i r m - s p e c i f i c G A A P accounting and market value data that Boyd, G r a h a m ,
and Hewitt use, (2) use of the coefficient of variation of
return on assets rather than the risk-of-failure measure,
and (3) differences in the sample periods. However, the
results focus attention on a question that Boyd, Graham, and Hewitt do not fully address: What standard
s h o u l d be a p p l i e d to j u d g e the v a l u e of p o t e n t i a l
diversification? The question is an important one because each standard has different implications. The riskminimizing standard is far more restrictive, for example, than one that would allow banks into any activity
that is not riskier than activities currently permitted.

Efficient Portfolios of Activities
While studying pairwise combinations offers some
interesting insights into the potential portfolio effects
of bank holding company diversification, this analysis
is not able to determine which portfolios of multiple
activities are most efficient—that is, which portfolios
offer the highest return for the lowest standard deviation of return on assets. For example, the pairwise analysis m a y suggest that a bank holding c o m p a n y ' s investment in certain industries would reduce returns and
increase risk. However, w h e n multiple industries are
combined into efficient portfolios, those same industries may enter into at least some of the portfolios.
Given the limitations imposed by multicollinearity
in the data set, the efficient portfolio analysis focuses
on two sets of firms. The first is the set of conventional
credit-granting industries that bank holding companies
can currently enter without limitation: mutual savings
banks, savings and loans, personal credit institutions,
and business credit institutions. 16 The second includes
industries that at least some commercial banks have
frequently taken a strong interest in entering: securities
brokers and dealers; commodity brokers and dealers;
life insurance; insurance agents, brokers, and servicers;
regulated investment companies; real estate operators
and lessors of buildings; and real estate subdividers and
developers. 17
Table 5 presents the results of forming portfolios of
these traditional credit-granting intermediaries over
the 1974-80 subperiod. T h e first efficient portfolio is
formed at the mean bank holding company return on
assets for the period, .27. Thereafter, the portfolios begin with a return on assets of .3 and increase incremen-

16



Economic Review-

tally by .1. Bank holding companies dominate the portfolios with the lowest standard deviations but decline in
importance as the means and standard deviations of return on assets increase. Mutual savings banks enter into
only the lowest risk portfolio while savings and loans
do not enter any of the efficient portfolios. Business
credit dominates at higher return levels, with the share
of personal credit institutions also increasing. Although
the standard deviations of return increase u n i f o r m l y
with the mean (as expected in a set of efficient portfolios), the coefficients of variation of return are remarkably stable at around 25 percent. Thus, the coefficients
of variation for these portfolios is consistently below
that of bank holding companies by themselves. This result could be interpreted as suggesting that there should
have been no limits on these activities during the 197480 subperiod.
Table 5 also presents the efficient portfolios of traditional credit-granting industries for the 1981-89 subperiod. In contrast to the earlier subperiod, only bank
holding companies and personal credit institutions enter the efficient portfolio in this later period. Furthermore, the coefficient of variation in return on assets
consistently increases as the share devoted to nonbank
holding company industries increases. The implication
is that any significant diversification out of the bank
holding company industry will increase the portfolio's
riskiness. In a n s w e r to those w h o believe that bank
holding companies should not be allowed to expand
into any activity that increases risk, these results argue
that b a n k holding c o m p a n i e s should not have been
allowed to invest in any of these traditional creditgranting activities during the 1981-89 period.
T h e portfolios of activities that bank holding companies would like to enter is presented in Table 6. For
the first subperiod, 1974-80, more than 80 percent of
the assets are devoted to bank holding companies for
the two portfolios that have a return on assets standard
deviation of 0.06, which is lower than the standard deviation for bank holding companies by themselves. The
efficient portfolio with the same standard deviation of
return as bank holding companies, 0.08, places only 46
percent of its assets in bank holding companies. Furthermore, bank holding companies' share of the assets
in the efficient portfolio drops to zero when the portfolio standard deviation reaches 0.14. Note also that all of
the efficient portfolios have a substantially lower coefficient of variation than the bank holding companies by
themselves. Life insurance becomes a dominant part
of the portfolio at intermediate return values. Insurance
agents dominate the portfolio at higher levels of return. 18 Only two industries never appear in the first sub-

September/October 1993

period's efficient portfolio: real estate operators, and
real estate subdividers and developers.
The bank holding company's efficient portfolios and
the activities they would like to enter for the 1981-89
subperiod are also presented in Table 6. Once again,
bank holding companies are the dominant portion of
portfolios that have a lower standard deviation than
bank holding c o m p a n i e s by themselves (that is, less
than 0.09 percent). However, as the portfolio return and
standard deviation increase, banks' share of the portfolio falls dramatically. The efficient portfolio for a standard deviation of 0.09, which equals the bank holding
companies standard deviation for this period, places only 54 percent of its assets in bank holding companies.
The three main nonbank holding company activities in
the portfolio are life insurance, regulated investment
companies, and real estate subdividers and developers.
The only industry that never enters the efficient portfolio is securities brokers and dealers. All efficient portfolios
are less risky than banking as measured by the coefficient of variation of return on assets.
Because the real estate subdividers and developers
industry is the only one to report a negative average return on assets during this period, it is interesting that the
industry is included in the efficient portfolios during the
second subperiod. The negative return on assets suggests that the industry could only be included in the efficient portfolio if it produced offsetting reductions in risk
via a negative correlation with other elements of the
portfolio. This conjecture is supported by an examination of the (unreported) correlation matrix, which shows
a - 0 . 8 8 correlation between real estate subdividers and
developers and regulated investment companies. 1 9 The
inclusion of real estate subdividers and developers in
the efficient portfolio indicates the benefit of considering a wide variety of assets in forming efficient asset
portfolios. 211
One important point that emerges from comparing
the 1974-80 tables with the 1981-89 tables is that the
results depend in part on the time period and industry
changes that m a y occur. For example, life insurance
was a dominant part of the low-risk portfolio in the
1974-80 period, but it became much less important in
the low-risk portfolio for 1981-89 subperiod. While the
historical data used in this study provides some insight
into the relative returns and risks of different types of
portfolios, one cannot definitely say which portfolios
will perform the best during the 1990s because no one
knows exactly how the various financial services industries will perform.
Overall, analyzing efficient portfolios of bank holding companies plus selected nonbanking activities sug-

Federal Reserve Bank of Atlanta



gests that in the two periods studied, nontraditional activities would have provided better diversification than
traditional credit-granting industries. T h e traditional
activities apparently added very little to bank performance during the later subperiod, whereas the restricted
a c t i v i t i e s — t h o s e that b a n k s want to e x p a n d i n t o —
would have generated substantially improved portfolio
performance.

.Efficient Portfolios Using Bank Holding
Company Regulatory Returns
A potential criticism of the above results is that bank
holding companies' average return on assets using IRS
data is unreasonably low, with an average return on assets between 1981 and 1989 of only 0.26. The low return on assets m a y reflect parts of the tax code that
allow taxable e a r n i n g s to be significantly less than
G A A P earnings. For example, excluding the interest on
state and local government obligations reduced the return on assets of bank holding companies reporting to
the Fed by somewhere between 0.20 and 0.36 percentage points between 1981 and 1989.
As a check on the findings based on IRS data, the
efficient portfolios for the 1981 -89 subperiod were recalculated, replacing the IRS's reported average return
on assets with the average pretax return on assets for
all consolidated bank holding c o m p a n i e s filing with
the Federal Reserve System in the relevant year. Previous casual inspection of Federal Reserve and G A A P
figures suggests that the two sources generally report
similar after-tax net income and total assets. Unfortunately, no similarly comprehensive measure was available for the other industries in our sample. Hence, it
was necessary to continue using IRS data for these industries although G A A P accounting returns for these
industries are likely to differ somewhat from IRS accounting returns. If other industries also receive special
tax breaks their IRS reported net income may also be
less than their accounting net income, and the use of
IRS data for every industry except bank holding companies may bias these results in favor of including a
larger proportion of bank holding companies in the efficient portfolios.
The use of figures reported by the Federal Reserve
produced a higher average return on assets of 0.72 for
the 1981-89 subperiod. However, the standard deviation of returns also increased, leaving the coefficient of
variation of return on assets essentially unchanged at
36 percent. One surprise is that the correlation between

Economic Review

17

Table 5
Efficient Portfolios
Traditional Bank-like Activities
1974-80
Efficient Risk a n d Return C o m b i n a t i o n s
Mean ROA (%)

0.27
29.63

0.30
0.08
26.66

95.44

Standard Deviation ROA ( % )
Coefficient of Variation R O A (%)

0.40

0.50

0.10
25.00

0.13

0.15

26.00

97.34

88.79

0.00

0.08

0.00

0.60

0.70

0.80

0.20

25.00

0.17
24.29

25.00

80.24

71.69

63.14

0.00

0.00

0.00

54.59
0.00

0.00

0.00

9.46

11.46

0.90
0.22
24.44

25.00

24.55

25.00

1.30
0.32
24.62

46.04

37.49

28.94

20.39

11.84

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

15.46

0.00
17.46

0.00

13.46

19.46

21.45

1.00

1.10

1.20

0.25

0.27

0.30

1.40
0.35
25.00

Associated Portfolio Allocations
Bank Holding Companies
Mutual Savings Banks

4.56

Savings and Loans

0.00

Personal Credit Institutions

0.00

0.00

0.00

0.00

Business Credit Institutions

1.46

5.46

0.00
7.46

0.00

1.20

3.46

14.30

20.85

7.75

27.40

33.95

40.50

47.05

53.60

60.15

66.70

3.29
0.00
0.00
23.45
73.25

1981-89
Efficient Risk a n d Return C o m b i n a t i o n s
M e a n ROA (%)

0.26

Standard Deviation ROA (%)

0.09

Coefficient of Variation ROA (%)

0.30

0.40

0.12

0.50

0.25

0.60

0.39

0.70

0.54

0.80
0.83

0.68

34.62

40.00

62.50

78.00

90.00

97.14

100.00

93.10

75.86

58.62

41.38

24.14

6.90

0.00

0.00

0.00

0.00

0.00

0.00

103.75

Associated Portfolio Allocations
Bank Holding Companies
Mutual Savings Banks

0.00

Savings and Loans

7 7
llon^^75^rcei^fn

0.00

0.00

0.00

0.00

24.14

0.00

41.38

0.00

0.00

6.90

0.00

Business Credit Institutions
To T/nntZ
Z d Z i ^ ' Z r

0.00

Personal Credit Institutions

58.62

75.86

93.10

0.00

0.00

0.00

0.00

0.00

0.00

1
T

7

T

8
bu^ess

C(inZ

«

cr^it

°f

€
S,

rf

X

T

ymg

V T t , 0
assnciat^n'1^

ileVGlS

a deS'rediate
n

0 f

°fret
r m -

a. return

^

"

S k (mean

C°mpUter

- For example'

w e m thCn CalCUlated
J°

U m 3nd

r6tUmS
Pr°gram

and
tHen

deviations)-

ln

combinations

in the 1974 80 p e r i o d the third Column

-

5' 6' and
7 each
™lumn represents a different efficient portfoof assets that minimized the risk for that level
ofreturnTheXsTn-

Tables

P™ents the results

^en
the program was asked to calculate he
invests 88.79 percent of its assets in bank holding companies, 3.46 percent of itsportfolios in personal Credit institu-

m T U a ' S a V m 8 S^
Z r c e i f ^ ^ ^ ^ r ^ ^ T ^
^
?
' Z
percent. I he coefficient of variation given these return and standard deviation values is 25 percent.




Standard
id6ntified

°

r SaVin8S

^

hanS-

The

Slandard

d6Viation

aSSOciated

with

this

risk portfolio is 0.10

Table 6
Efficient Portfolios
N o n t r a d i t i o n a l A c t i v i t i e s IRS B a n k H o l d i n g C o m p a n y

Returns

1974-80
Efficient Risk and Return C o m b i n a t i o n s
Mean R O A ( % )

0.47

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

4.50

5.00

5.50

6 00

6 50

7 00

Standard Deviation R O A ( % )

0.06

0.06

0.08

0.09

0.14

0.25

0.36

0.48

0.59

0.71

0 83

0 95

1 07

1 23

1 45

Coefficient of Variation R O A (%) 12.77

12.00

8.00

6.00

7.00

10.00

12.00

13.71

14.75

15.78

16.60

17.27

17.83

18*92

2071

84.67

82.66

2.48

2.73

46.23

13.59

0.00

0.00

0.00

0.00

0.00

0.00

12.34

43.21

31.54

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

19.44

0.00

0.00

17.59

0.00

0.00

15.75

53.14

0.00

13.90

45.94

0.00

12.06

38.74

8.68

8.37

24.35

18.56

6.53

17.15

28.45

2.90

9.95

38.34

0.93

2.76

48.23

0.00

0.00

58.12

0.00

0.00

68.01

14.23

0.00

77.90

12.30

0.00

70.44

10.36

0.00

10.62

8.43

0.00

0.00

6.50

0.66

0.00

4.57

3.91

0.00

2.88

5.00

0.00

2.27

6.08

0.00

2.23

7.17

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

A s s o c i a t e d Portfolio A l l o c a t i o n s
Bank Holding Companies
Security Brokers/Dealers
Commodity Broker/Dealers
Life Insurance
Ins. Agents, Brokers, Serv.
Regulated Investment Co.
Real Estate Operators
Subdivides and Developers

6.75

9.68
3.39

8.25

10.21

0.00

0.00

2.83

1.74
16.16

18.09

18.14

7.52

0.00
0.00

0.00

0.00

60.69

0.00

69.85

88.00

22.63

12.00

21.17

0.00

0.00
0.00

1981-89
Efficient Risk and Return C o m b i n a t i o n s
Mean R O A ( % )

0.54

1.00

1.50

2.00

2.50

3.00

3.50

4.00

4.50

5.00

5.50

6 00

6 50

7 00

Standard Deviation R O A ( % )

0.03

0.05

0.09

0.13

0.18

0.23

0.31

0.40

0.50

0.61

0.73

0 88

1 04

1 30

7 50
1 63

Coefficient of Variation ROA ( % )

5.56

5.00

6.00

6.50

7.20

7.67

8.86

10.00

11.11

12.20

13.27

14.67

16D0

18.59

21.73

93.34

76.66

54.26

0.00

0.00

31.86

0.00

9.46

0.00

0.00

20.88

28.13

26.58

41.58

45.48

32.62

53.34

29.05

0.00

38.06

30.59

0.00

34.54

28.55

0.00

31.02

24.27

0.00

26.56

19.92

0.00

20.81

15.57

0.00

0.00

11.22

2.67

0.00

7.79

10.64

0.00

5.63

18.61

25.97

0.00

0.00

23.86

0.00

0.00

18.16

0.00

0.00

12.45

0.00

0.00

6.74

0.00

0.00

3.14

0.00

0.00

2.42

0.00

0.00

1.71

0.00

0.00

0.99

0.00

0.00

0.00

0.00

61.21

A s s o c i a t e d Portfolio A l l o c a t i o n s
Bank Holding Companies
Security Brokers
Commodity Broker/Dealers
Life Insurance
Ins. Agents, Brokers, Serv.
Regulated Investment Co.
Real Estate Operators
Subdivides and Developers




16.07

0.00

6.38

0.00

13.63

1.31

3.47

3.45

9.31

3.21

15.06

0.00

0.00

0.00

5.35

0.00

11.87

0.00

18.39

0.00

24.92

0.00

24.44

0.00

18.83

0.00

13.23

0.00

7.63

0.00

0.00

0.00

6.17

0.00

70.95

16.20

3.34

0.00

83.80

0.00

96.66

0.00

0.00

0.00

0.00

0.00

0.00

Table 7
Efficient Portfolios,

1981-89

Federal Reserve Bank Holding C o m p a n y

Returns

Traditional Bank-like Activities
Efficient Risk a n d Return C o m b i n a t i o n s
Mean R O A ( % )

0.55

0.60

0.65

0.17

0.17

0.70

0.17

0.75

30.09

30.09

0.18

0.21

28.33

26.15

0.56

25.71

28.00

70.00

67.97

70.55

74.60

0.00

0.00

79.12

0.00

0.00

75.00

2.95

0.57

0.00

33.33

2.99

0.00

0.00

10.22

12.16

0.00

0.00

10.19

13.98

0.00

18.85

15.76

0.00

18.89

Coefficient of Variation R O A ( % )

0.55
0.17

67.92

Standard Deviation R O A (%)

16.72

11.42

25.00

66.67

5.12

0.00

0.00

0.80

A s s o c i a t e d Portfolio A l l o c a t i o n s
Bank Holding Companies
Mutual Savings Banks
Savings and Loans
Personal Credit Institutions
Business Credit Institutions

Nontraditional Activities
Efficient Risk a n d Return C o m b i n a t i o n s
Mean ROA ( % )

1.77

Standard Deviation R O A (%)

0.14

Coefficient of Variation ROA ( % ) 7.91

2.00

2.50

0.15

3.00

3.50

0.18

4.00

0.31

4.50

5.00

10.00

11.11

0.73

0.88

7.00

8.86

0.61

6.50

7.67

0.50

6.00

7.20

0.40

5.50

7.50

0.23

12.20

1.04

13.27

14.67

1.30

1.63

16.00

18.57

21.73

28.66

9.36

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

10.64

2.67

6.17

0.00

18.61

16.07

0.00

26.58

25.97

0.00

26.73

23.86

0.00

20.85

18.16

0.00

0.00

12.45

0.00

0.00

6.74

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

7.50

A s s o c i a t e d Portfolio A l l o c a t i o n s
Bank Holding Companies

37.54

Security Brokers/Dealers

0.00

Commodity Brokers/Dealers

0.00

Life Insurance
Ins. Agents/Brokers/Serv.
Regulated Investment
Real Estate Operators
Subdividers and Developers




18.15
4.17
17.01
0.00
23.14

5.10

7.12

11.22

20.26

15.57

27.33

31.02

19.92

24.27

34.54

28.55

38.06

30.59

32.62

41.58

45.48

53.34

29.05

61.21

16.20

3.34

70.95

83.80

96.66

0.00

0.00

25.13

0.00

0.00

29.46

24.44

0.00

0.00

18.83

0.00

13.23

7.63

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

IRS returns on assets and Federal Reserve returns on
assets is statistically insignificant. This increase in variance and the insignificant correlation suggests that differences in return on assets are caused by more than
just the tax treatment of municipal securities. Unfortunately, the IRS data lack sufficient detail for determining other contributing factors.
Table 7 gives the results of recalculating efficient
portfolios using IRS data. The results are quite similar
to those in Table 6, with perhaps the biggest difference
being that the lowest reported standard deviation of returns in Table 7 exceeds those reported in Table 6. The
reason for this change is that the standard deviation of
bank holding companies' return on assets increased as
calculated for Table 7 so that very low-risk portfolios
were no longer feasible.
Bank holding companies continue to dominate the
portfolios of traditional banking activity in Table 7 for
the 1974-80 period. Personal credit and business credit
still enter into at least some efficient portfolios, but
bank holding companies account for more than twothirds of most of the efficient portfolios. T h e d o m i nance of bank holding companies fades only when the
mean return on assets reaches 0.8. Moreover, the coefficient of variation of return on assets increases substantially when the mean return goes to 0.8.
In terms of nontraditional activities, Table 7 indicates
that bank holding companies enter into the lowest-risk
portfolios, but their proportion of the portfolio quickly
drops to zero at higher levels of return. C o m m o d i t y
brokers, the life insurance industry, insurance agents,
regulated investment trusts, and real estate subdividers
and developers all enter into at least some of the efficient portfolios. Security brokers and real estate operators are not included in any efficient portfolio.

Conclusion
A n i m p o r t a n t issue in the d e b a t e over e x p a n d e d
bank holding company powers is whether diversification would increase or decrease risk for these institutions. The emerging consensus in the literature using
pairwise combinations of individual firm data was that
deregulation would at best provide insignificant portfolio diversification gains in return on assets and would at
worst increase bank holding companies' risk. The primary challenge to this consensus has been Litan's analysis of efficient portfolios using IRS industry data. The
research reported here found greater benefits for larger
combinations of industries than Litan found. A direct

Federal Reserve Bank of Atlanta




comparison is not readily accomplished, however, because Litan did not calculate pairwise combinations.
This study sought to reconcile the analysis of pairwise combinations with efficient portfolio analysis and
to update Litan's analysis. It was found that using the
IRS industry data produces m o r e risk-reducing pairwise combinations than seem to occur in studies using
other data. T h e s e d i f f e r e n c e s m a y d e v e l o p directly
from the use of IRS data or the use of the coefficient of
variation in return on assets. Further work seems desirable for pinning d o w n w h y the results of this study
vary from those of others using pairwise industry combinations.
Another finding of this analysis is that some efficient multiple-activity portfolios generate substantially
higher returns and with lower risk (as measured by the
coefficient of variation in return on assets) than does
investing all assets in bank holding c o m p a n i e s or in
pairwise combinations of bank holding companies and
a nonbank industry. For example, the least risky pairwise combination between 1981 and 1989 consisted of
95 percent bank holding company and 5 percent regulated investment companies. This portfolio had a coefficient of variation of 5.1 percent and a return on assets
of 0.63. T h e lowest-risk (lowest coefficient of variation) efficient portfolio contained bank holding comp a n i e s ; c o m m o d i t y b r o k e r s / d e a l e r s ; life i n s u r a n c e ;
insurance agents, brokers, and servicers; regulated investment companies; and subdividers and developers.
This efficient portfolio had a marginally lower coefficient of variation of 5.0 percent, and a substantially
higher return on assets of 1.00 percent. Thus, the finding of diversification gains is not solely a function of
the use of IRS data. Furthermore, the results are confirmed by the attempt to correct for the biases in IRS
m e a s u r e m e n t of b a n k h o l d i n g c o m p a n y i n c o m e by
substituting Federal Reserve pretax income figures for
bank holding companies.
As noted in the introduction, it is important that policymakers consider more than the portfolio risk effect
of allowing bank holding companies to enter into and
to e x p a n d their operations in traditionally n o n b a n k
activities. These results in combination with those of
other studies appear to indicate a problem for policymakers who emphasize the benefits of portfolio diversification. Financial conglomerates are built one merger
at a time, and the pairwise studies suggest that many
combinations would, at least temporarily, increase the
riskiness of bank holding companies. If regulators permit certain pairwise mergers, then a bank holding company m a y b e riskier, especially if it c h o o s e s not to
m a k e further acquisitions or if conditions prevent it

Economic Review

21

f r o m m e r g i n g with additional industries. O n the other

inherent in k n o w i n g that a bank h o l d i n g c o m p a n y will

h a n d , a b a n k h o l d i n g c o m p a n y that h a s m e r g e d with

never be able to f o r m this efficient portfolio of activi-

several n o n b a n k f i r m s m a y h a v e a substantially higher

ties without b e i n g allowed to e n g a g e in p a i r w i s e m e r g -

return a n d lower risk than one restricted to traditional

e r s that m a y i n d i v i d u a l l y c a u s e at least a t e m p o r a r y

b a n k i n g activities. R e g u l a t o r s f a c e the difficult p r o b l e m

increase in risk.

Appendix
Finance, Insurance, and Real Estate Industry B r e a k d o w n
The 1RS corporate returns publication breaks the general finance, insurance, and real estate sector down into
twenty-three minor industry groups as follows (a brief description of some industries is included as an aid to readers):
Banking
• Mutual savings banks
• Bank holding companies—including both one-bank and
multibank holding companies
• Independent banks—excluding mutual savings banks
and bank holding companies
Credit Agencies
• Savings and loan associations
• Personal credit institutions—establishments primarily
engaged in providing loans to individuals and establishments engaged in financing retail sales by installment
plan and in automobile financing
• Business credit institutions—establishments engaged in
making loans to business and agricultural enterprises,
such as short-term business credit institutions (commercial finance companies), accounts receivable and commercial paper factoring, direct financing of working
capital, captive automobile finance companies (for example, GMAC), mercantile financing, and so forth
Insurance
• Life insurance companies
• Mutual insurance companies—excluding life or marine
and certain fire or flood insurance companies
Insurance Agents and Brokers
• Agents and brokers dealing in insurance, and organizations offering services to insurance companies and policyholders, such as insurance claim adjusters
Real Estate
• Real estate operators and lessors of buildings—including firms that operate and lease but do not develop real

 22


Economic Review-

•
•
•
•

property, such as operators of commercial and office
buildings, retail establishments and shopping centers,
and so forth
Lessors of mining, oil, and similar properties
Lessors of railroad property—including firms such as
airport leasing offices, landholding offices, and others
Condominium management and cooperative housing associations
Subdividers and developers—including firms engaged
in subdividing real property into lots and in developing
them for resale on their own account

Security, C o m m o d i t y Brokers and Services
• Security brokers, dealers, and flotation companies—
establishments engaged in the purchase, sale, and brokerage of securities and those, generally known as
investment bankers, that originate, underwrite, and distribute securities issues
• Commodity contract brokers and dealers, security and
commodity exchanges, and allied services—firms that
buy and sell commodity contracts on either the spot or
future basis for their own account or the account of others or that provide investment advice regarding securities to companies and individuals on a contractual or fee
basis, and so forth
Holding and Other Investment Companies
(excludes bank holding companies)
• Regulated investment trusts—a wide range of firms
such as open and closed-end investment funds, money
market mutual funds, unit investment trusts, and so forth
• Real estate investment trusts—firms engaged in closedend real estate investments or related mortgage assets
that meet the requirements of the amended Real Estate
Investment Act of 1960, such as mortgage investment
trusts, mortgage and realty trusts, and real estate investment trusts
• Small business investment trusts

September/October 1993

Notes
1. See Benston (1990) for an analysis of the evidence supporting the restrictions on bank involvement in securities activities. These restrictions were reinforced by limits incorporated in various pieces of legislation, with perhaps the
two most important pieces being the Bank Holding Company Act of 1956 and the Act's 1970 amendments.
2. Although the problems with thrift deregulation are frequently cited as an argument against any additional deregulation
of insured depositories, the argument contains a number of
flaws. First, thrifts had suffered massive losses in their traditional business of home mortgage lending in the early 1980s.
Second, one of the biggest deregulatory measures, the elimination of interest rate restrictions on consumer deposits, was
forced by the high interest rate environment of the late 1970s
and early 1980s. The alternative to deregulation was a drastic shrinkage in the size of both the thrift and commercial
banking industries as funds flowed out to unregulated money
market mutual funds. Third, the credit losses that occurred in
nontraditional activities were due in no small part to the fact
that the regulators were loosening capital requirements,
thereby encouraging rapid expansion at financially weak
thrifts. The commercial bank regulators in contrast progressively tightened their capital regulations during the 1980s.
3. Other studies from the 1970s include Johnson and Meinster
(1974) and Eisemann (1976). Wall and Eisenbeis (1984)
also use IRS data to examine the portfolio effect of banks'
diversification into selected nonbanking activities. Unfortunately, the study is limited to firms with positive income and
hence not as comprehensive as these results.
4. The coefficient of variation is the standard deviation of return on assets divided by the mean return on assets. Industries with higher coefficients of variation are riskier. The
correlation of returns captures the relationship between the
returns on assets of different industries and bank holding
companies. If the con-elation is positive, the nonbank industry would lend to experience high returns when bank holding
companies have high returns and low returns when banks
have low returns. If the correlation is negative, the nonbank
industry would tend to experience high returns when banks
have low returns, and low returns when banks have high returns. If the correlation is zero, there is no relationship between the returns of the nonbank industry and bank holding
companies. Bank holding companies are more likely to gain
diversification benefits from combining with industries that
have a low positive or a negative correlation of returns with
bank holding companies.
5. That is, Boyd, Graham, and Hewitt created a new firm with
95 percent of the assets coming from the bank and 5 percent
from the nonbank firm.
6. See Boyd and Graham (1986) for a discussion of the issue of
managerial incentives.
7. Boyd, Graham, and Hewitt point out that an important issue
in evaluating the effect of a merger is the purchase price paid
for the target by the acquiring organization. Virtually all studies of historical data implicitly assume that no premium will

Federal Reserve Bank of Atlanta



be paid to the target. Researchers use this assumption not because it is realistic but because they have no good basis for
determining the likely magnitude of the takeover premium.
8. Banks' ability to deduct the interest from municipal securities was severely constrained by the Tax Reform Act of
1986. For a summary of the changes see "A Farewell to Municipal and Other Tax Effects," ABA Ranking
Journal
(February 1987): 35-36.
9. The banking regulators are currently requiring most banking
organizations to maintain a minimum equity-capital-toassets ratio of 4 percent to 5 percent.
10. The holding and other investment company category includes regulated investment trusts (which includes mutual
funds), real estate investment trusts, and small business investment trusts.
11. The set of possible portfolios with each industry includes a
portfolio of 100 percent bank holding company assets. Thus,
if a portfolio containing a 5 percent investment in the nonbank holding company industry has a coefficient of variation
exceeding that of bank holding companies alone—34.4 percent, for example—then the maximum investment allowed
in that industry would be less than 5 percent. Note also that
it is not possible to specify the exact percentage of assets that
a bank holding company should be able to hold in a nonbanking industry because the analysis looks at diversification only at selected points. Thus, the finding that the
risk-minimizing portfolio contains a 10 percent investment
in an industry is consistent with the actual risk-minimizing
point lying between 5 percent and 25 percent diversification.
12. The four that are risk-minimizing at the 5 percent level
are lessors of mining and oil properties, commodity brokers/
dealers, regulated investment companies, and real estate investment trusts. At the 10 percent level the four that are riskminimizing are real estate operators and lessors of buildings,
lessors of railroad properties, security brokers/dealers, and
other holding and investment companies.
13. Bank holding companies would generally be allowed to invest somewhat more than these stated maximums because
these figures are based on consideration of only a limited
number of portfolio alternatives. For example, the maximum
investment in security brokers/dealers would be the point at
which the risk of the pairwise portfolio equaled that of bank
holding companies alone—somewhere between 25 percent
and 50 percent.
14. The two that are no riskier than bank holding companies at
the 10 percent level are lessors of mining and oil properties
and commodity brokers/dealers; at the 25 percent level, real
estate operators and lessors of buildings, lessors of railroad
properties, security brokers/dealers, and real estate investment trusts. The two that are less risky at the 90 percent level
are regulated investment companies and other holding and
investment companies.
15. The ten industries are independent banks; insurance agents,
brokers, and service; real estate operators and lessors of buildings; lessors of mining and oil properties; lessors of railroad

Economic

Review

23

properties; security brokers/dealers; commodity brokers/dealers; regulated investment companies; real estate investment trusts; and other holding and investment companies.
16. Mutual savings banks and savings and loans would generally
have to convert to stock institutions before they could be acquired by commercial banks.
17. Looking at historical accounting returns, the real estate subdivider and developer industry was an unprofitable activity
according to 1RS data. However, the industry' was included
anyway because some banks showed a strong interest in it
given their substantial volume of lending to subdividers and
developers.
18. Note that if efficient portfolios were formed the total assets
of the combined firm would likely have to shrink as the proportion devoted to insurance agency increased.

19. A positive correlation exists between the profitability of bank
holding companies and real estate subdividers and developers. Thus, it is not surprising that this industry fails to enter
the lowest risk portfolio that contains predominately bank
holding companies and includes only a minimal investment
in regulated investment companies.
20. One note of caution should be considered in interpreting
Table 6. Some of the industries in the table—for example,
insurance agents and regulated investment companies—use
proportionately far fewer of their own assets than bank holding companies do to generate a similar amount of revenue.
Thus, increasing the proportion of the portfolio invested in
these industries may also imply shrinking the total size of the
firm.

References
Benston, George J. The Separation of Commercial and Investment Banking: The Glass-Steagall Act Revisited and Reconsidered. New York: Oxford University Press, 1990.
Benston, George J., William C. Hunter, and Larry D. Wall.
"Motivations for Bank Mergers and Acquisitions: Enhancing
the Deposit Insurance Put Option versus Increasing Operating Net Cash Flow." Federal Reserve Bank of Atlanta unpublished working paper, July 1993.
Benston, George J., and George G. Kaufman. "Risk and Insolvency Regulation of Depository Institutions: Past Policies
and Current Options." Federal Reserve Bank of Chicago,
Staff Memorandum, 1988.
Boyd, John H., and Stanley L. Graham. "Risk, Regulation, and
Bank Holding Company Expansion into Nonbanking." Federal Reserve Bank of Minneapolis Quarterly Review 10
(Spring 1986): 2-17.
. "The Profitability and Risk Effects of Allowing Bank
Holding Companies to Merge with Other Financial Firms: A
Simulation Study." Federal Reserve Bank of Minneapolis
Quarterly Review 12 (Spring 1988): 3-20.
Boyd, John H., Stanley L. Graham, and R. Shawn Hewitt.
"Bank Holding Company Mergers with Nonbank Financial
Firms: Effects on the Risk of Failure." Journal of Banking
and Finance 17 (1993): 43-63.
Brewer, Elijah, III. "Relationship between Bank Holding Company Risk and Nonbank Activity." Journal of Economics
and Business 41 (1989): 337-53.
. "The Risk of Banks Expanding Their Permissible Nonbanking Activities." The Financial Review 25 (November
1990): 517-37.
Brewer, Elijah, III, Diana Fortier, and Christine Pavel. "Bank Risk
from Nonbank Activities." Federal Reserve Bank of Chicago Economic Perspectives 12 (July/August 1988): 14-26.
Brewer, Elijah, III, and Thomas H. Mondschean. "An Empirical
Test of the Incentive Effects of Deposit Insurance: The Case

24



Economic Review-

of Junk Bonds at Savings and Loan Associations." Federal
Reserve Bank of Chicago Working Paper 91-18, September
1991.
Carnell, Richard S. "A Partial Antidote to Perverse Incentives:
The FDIC Improvement Act of 1991." In Rebuilding Public
Confidence through Financial Reform, edited by Peter Dickson. Columbus, Ohio: The Ohio State University, 1992.
Eisemann, Peter C. "Diversification and the Congeneric Bank
Holding Company." Journal of Bank Research (Spring
1976): 68-77.
Eisenbeis, Robert A., Robert S. Harris, and Josef Lakonishok.
"Benefits of Bank Diversification: The Evidence from Shareholder Returns." Journal of Finance 39 (1984): 881-92.
Heggestad, Arnold A. "Riskiness of Investments in Nonbank
Activities by Bank Holding Companies." Journal of Economics and Business 27 (Spring 1975): 219-23.
Johnson, Rodney D„ and David R. Meinster. "Bank Holding
Companies' Diversification Opportunities in Nonbank Activities." Eastern Economic Journal (October 1974): 145365.
Kane, Edward J. The Gathering Crisis in Federal Deposit Insurance. Cambridge, Mass.: MIT Press, 1985.
Litan, Robert E. "Evaluating and Controlling the Risks of Financial Product Deregulation." Yale Journal on Regulation 3
(Fall 1985): 1-52.
Pike, Christopher J., and James B. Thomson. "FDICIA's Prompt
Corrective Action Provisions." Federal Reserve Bank of
Cleveland Economic Commentary (September 1, 1992).
Rose, Peter S. "Diversification of the Banking Firm." The Financial Review 24 (May 1989): 251-80.
Rosen, Richard J., Peter R. Lloyd-Davies, Myron L. Kwast, and
David B. Humphrey. "New Banking Powers: A Portfolio
Analysis of Bank Investment in Real Estate." Journal of
Banking and Finance 13 (July 1989): 355-66.

September/October 1993

Stover, Roger D. "A Re-examination of Bank Holding Companies' Acquisitions." Journal of Bank Research 13 (Summer
1982): 101-8.
Todd, Walker F. "FDICIA's Discount Window Provisions."
Federal Reserve Bank of Cleveland Economic. Commentary
(December 15, 1992).
Wall, Larry D. "Insulating Banks from Nonbank Affiliates."
Federal Reserve Bank of Atlanta Economic Review 69 (September 1984): 18-28.

Federal Reserve Bank of Atlanta



. "Has Bank Holding Companies' Diversification Affected Their Risk of Failure?" Journal of Economics and Business 39(1987): 313-26.
. "Too-Big-to-Fail after FDICIA." Federal Reserve Bank
of Atlanta Economic Review 78 (January/February 1993): 114.
Wall, Larry D., and Robert A. Eisenbeis. "Risk Considerations
in Deregulating Bank Activities." Federal Reserve Bank of
Atlanta Economic Review 69 (May 1984): 6-19.

Economic

Review

25




ifrivate Insurance
of Public Debt:
Another Look at the
Costs and Benefits of
Municipal Insurance

r

Stephen D . Smith and Richard B. Harper

he market for municipal securities has changed dramatically in the
past two decades. Twenty years ago the dollar value of outstanding state and local government bonds was less than $200 billion,
with the majority of these securities held by institutional investors.
Commercial banks alone accounted for roughly half the total demand, leaving a relatively small segment to retail investors. Municipal insurance had just appeared but received little attention in a market experiencing
default rates of less than 0.005.

Smith holds the H. Talmage
Dobbs, Jr., Chair of Finance
at Georgia State University
and is a visiting scholar at the
Federal Reserve Bank of
Atlanta. Harper, formerly an
intern at the Federal Reserve
Bank of Atlanta, is a student
in the MBA Program at
Indiana University.

Federal Reserve Bank of Atlanta




Today's municipal bond market profile is quite different. The total dollar
value outstanding has increased, on a nominal basis, by roughly 600 percent
(to 1.2 trillion) since 1973. Retail investors now account for 75 percent of total holdings (Board of Governors of the Federal Reserve System). At the same
time, private bond insurance has become so prevalent that the insured Aaa
bond stands as the most common type of rated security in the tax-exempt
market (Thomas R Addison, Jr. 1990).
Such growth in the volume of municipal insurance would probably not
have occurred if using these third-party guarantees had not generally benefited securities issuers. Research based on data f r o m the 1970s and early
1980s suggests that the average benefit to issuers using insurance was positive but no larger than 25 basis points. However, these studies used data
f r o m a period during which institutional investors dominated the market.
There has been little examination of m o r e recent data to determine the costs
and benefits of third-party insurance.

Economic Review

27

A number of factors suggest that the net benefits of
insurance probably have changed over the last decade.
On the one hand, various tax reform acts in the 1980s
contributed to a decline in institutional holdings in general, and bank holdings in particular, in the tax-exempt
segment of the market for state and local bonds. Most
of that decline has been replaced by retail investment.
For a number of reasons, retail investors are likely to be
m o r e concerned about default risk than their institutional counterparts and therefore willing to pay more
for greater insurance benefits. 1 Increased competition
among underwriters may also have generated additional
net benefits to issuers. Growth in the insurance market
has come, however, at the expense of increasing leverage ratios on the part of insurers. The implication is that
guarantees may have increasingly covered greater risk
over this period. Any increase in risk for insurers, other
things held constant, would tend to reduce the net benefits of insurance guarantees attached to securities.
The purpose of this article is first to document that
the investor profile in the municipal bond market has
in fact changed dramatically. The discussion then analyzes the results of tests conducted to assess whether
insurance continues to provide benefits in the current
market. Using data f r o m issues sold by the State of
Florida and its subdivisions from 1990 through 1992,
the study provides evidence suggesting that, on average, state and local governments still realize a net financing benefit from selling insured bonds.
The results reported here have implications for municipal bond issuers, regulators, and investors alike.
State and local governments will find some comfort
knowing that, despite radical changes in the composition of investor demand, employing a third-party enhancer m a y still lower their interest costs. It should be
of value to regulators, concerned about the increasing
riskiness of third-party enhancers, that the Florida data
provide no evidence that investors are penalizing insured issuers on the assumption that insurance signals
dilution in insurer capital. Finally, potential investors,
including banks, can use the findings to help analyze
the yield characteristics of alternative investment strategies, particularly the potential for yield loss accompanying insured bonds as compared with those that come
to the market without insurance.

i/nderstanding Bond Insurance
Municipal bond insurance, in its purest form, is an irrevocable guarantee from a third-party credit enhancer

28

Economic Review-




that it will pay the debt service of a particular bond issue to investors should the bond issuer be unable to
make the payment. As compensation for attaching this
guarantee, the credit enhancer is paid a fee by the municipality or its representative. In turn, the agencies that
rate the default risk of bonds have, since 1984, used the
presence of a guarantee f r o m a Aaa-rated insurance
company to justify assigning the issue a Aaa rating.
As in other markets, an issue's credit rating is not
the only characteristic by which it is judged in the market. On the whole, however, having a municipal bond
insurance policy attached decreases an issue's overall
default risk and increases its marketability, generating a
lower overall borrowing cost in the primary market and
increasing the issue's liquidity in the secondary market.
The dominant theme of earlier empirical studies on
municipal bond insurance and debt insurance in general has been that of whether issuers save on financing
costs when insurance is used. Although the consensus
of these studies is that the attachment of insurance does
reduce an issuer's borrowing costs in the primary market, results on the net benefits of the insurance have
been mixed. Ronald W. Forbes and Michael H. Hopewell (1976) and Ronald C. Braswell, E. J o e Nosari,
and Mark A. Browning (1982) have argued that their
samples showed minimal net benefits of insurance. On
the other hand, Charles W. Cole and Dennis T. Officer
(1981) and David S. Kidwell, Eric H. Sorensen, and
John M. Wachowicz, Jr. (1987), have shown that the reduction in interest cost owing to insurance is highly
significant and that the largest benefits accrue to the
least c r e d i t w o r t h y i s s u e r s . M o r e r e c e n t l y , L. Paul
Hsueh and R R . C h a n d y (1989) suggest that insured
bonds of underlying A credit quality gain no benefits
f r o m the use of insurance. Given the time f r a m e of
their data set, practitioners would seem to agree with
this assessment (Sylvan G. Feldstein 1983).
In addition to the changing nature of demand and
supply, several points of methodology and measurement make the accuracy of these earlier results questionable in today's market. First, several of the authors
cited employed the net interest cost (NIC) measure instead of the true interest cost (TIC), a choice with potential significance. While the TIC measure accounts
for the time value of money, the NIC does not, and its
use could produce misleading results. The TIC also incorporates the cost of the insurance premium and is
widely accepted as the current industry b e n c h m a r k .
(See the box on page 37 for a more detailed comparison of the TIC and NIC). Moreover, given the rapidly
evolving market of the late 1980s and early 1990s, a
look at more recent data seems worthwhile. For exam-

September/October 1993

million payment by the A M B A C Indemnity Corporation was one of the only payments investors received
from the defaulted $2.2 billion bond issue.

pie, several of the earlier studies included only issues
sold by competitive bid, or they restricted their focus
to g e n e r a l obligation b o n d s ( b o n d s s e c u r e d by the
pledge of the issuer's full faith and credit). The current
industry structure, however, is quite d i f f e r e n t f r o m
these samples. First, 75 percent of the new-issue municipal volume in 1991 was issued through negotiated
underwriting. In addition, bonds whose debt service is
to be paid f r o m revenues of the specific project, socalled revenue bonds, make up 70 percent of the new
issues for each of the past ten years (Bond Buyer 1992,
"Quarterly Statistics Supplement," 3A).

71ie Evolving Municipal Bond Market
Before turning to the data, it is worthwhile to consider the n a t u r e of participants in this m a r k e t . T h e
transition in the municipal bond market f r o m an institutional investor base to a p r e d o m i n a n t l y retail investor base seems to have started in the early 1980s
(see Charts 1 and 2). Chart 1 documents the increased
retail holdings of tax-exempt municipal bonds. Chart 2
s h o w s the c o m p o s i t i o n of r e t a i l i n v e s t o r d e m a n d
across tax-exempt investment vehicles. As Chart 1 indicates, commercial banks, traditionally the dominant
holders of m u n i c i p a l b o n d s , a c c o u n t e d f o r only 25
percent of the growth in municipal bonds during the
1975-86 period (Board of Governors). While state and
local g o v e r n m e n t b o n d s o f f e r e d p r e m i u m after-tax
yields to other investors, banks saw the value of purchasing them diminished by tax reform legislation in
1982 and 1984. Prior to the legislation, b a n k s h a d

A final point is that these studies' use of pre-1984
data is also a limitation because Moody's, one of the
two m a j o r rating agencies, made a significant policy
change in 1984. Although Standard and Poor's (S&P),
the other m a j o r rating agency, bestowed its highest
rating on municipal issues guaranteed by credit enhancers before July 1984, M o o d y ' s refused to recognize the presence of bond insurance and rated an issue
solely on the basis of its underlying f u n d a m e n t a l s .
This contrast in policies produced split ratings for insured issues in the marketplace. M o o d y ' s changed its
policy in 1984 in the wake of a default by the Washington Public P o w e r Supply System, w h e n the $ 1 3

Chart 1
T a x - e x e m p t M u n i c i p a l H o l d i n g s by S e c t o r of I n v e s t o r D e m a n d
(Volume)
Millions

1,200,000
1,000,000
800,000
600,000

400,000
200,000

0
1975

1977

1979

1981

Total

1983

1985

1987

1989

Retail
Bank

1991
Institutional

P & C Insurers

S o u r c e : C a l c u l a t e d by the Federal Reserve Bank of Atlanta using data from the quarterly F l o w of Funds report from the B o a r d of G o v e r n o r s
of the Federal Reserve System.

Federal Reserve Bank of Atlanta



Economic Review

29

been allowed to deduct all interest expense on deposits
used to purchase or carry tax-exempt obligations. The
tax reform acts trimmed the deduction of such interest
expense to 80 percent (Public Securities Association
1990, 107). This loss of tax-free income had a significant impact on the attractiveness of municipal bonds
versus other investments.
With passage of the Tax R e f o r m Act of 1986 the
transition f r o m an institutional to a retail investor base
accelerated. Banks were again the investors most adversely affected. Some have argued that the provisions
of the act "essentially lock banks out of the tax-free municipal bond m a r k e t " ("A F a r e w e l l to M u n i c i p a l s "
1987). Specifically, the legislation essentially eliminated the deduction of interest expense on deposits used to
p u r c h a s e municipal bonds. (The only exemption involves small local government issues of less than $10
million dollars.) Moreover, the corporate tax rate was
lowered from 46 percent to 34 percent, dramatically reducing these bonds' after-tax yield potential. Only in
rare circumstances would the addition of a municipal
bond, bought after July 7, 1986, contribute a higher
after-tax yield than those obtainable from most taxable
fixed-income instruments. The 1986 act also included a
new alternative m i n i m u m tax rate, which further re-

d u c e d bank d e m a n d for state and local g o v e r n m e n t
bonds. The net result was a 58 percent decline in municipal debt held by commercial banks over the 198692 period (Board of Governors).
W h i l e municipal b o n d s b e c a m e less attractive to
banks in the 1980s, they became more appealing to retail investors. Despite the fact that tax reform legislation lowered the highest marginal personal tax rate, the
elimination of most alternative tax shelters for individuals apparently shifted retail demand toward municipal b o n d s . T h e i n c r e a s e d p r e s e n c e of m u t u a l and
money market funds, which focus investments in taxexempt securities, has also made state and local bonds
more accessible to a greater array of retail investors.
Retail investors often save on transactions costs and
obtain less risk through diversifying with these investment vehicles.

(Growth in Insurance
Whatever reasons are behind the growth in supply
of insurance for state and local bonds, most of the expanded volume has been purchased by the retail sec-

Chart 2
M u n i c i p a l H o l d i n g s by C o m p o s i t i o n of Retail Investor D e m a n d
(Volume)
Millions

Total

Retail
Mutual Funds

Individual
M o n e y Market Funds

S o u r c e : C a l c u l a t e d by the Federal Reserve B a n k of Atlanta using data from the quarterly F l o w of Funds report from the B o a r d of G o v e r n o r s
of the Federal Reserve System.


Economic Review30


September/October 1993

tor. For the period since 1983, Chart 3 shows both the
value of new municipal issues and the percentage of
new issues insured.
Because virtually all insured bonds carry a Aaa rating, the supply shift toward insured issues has led to a
relative increase in the supply of Aaa securities. Had
the same securities come to market without insurance,
their so-called stand-alone ratings would have been
mainly in the " B a a " to "A" range. Researchers at Donaldson, Lufkin, and Jenrette have documented that this
relative increase in the supply of Aaa securities has
been accompanied by a reduction in the yield spread
between low- and high-grade bonds (Addison 1990).
Chart 4 represents the relationship between the percent
of all bonds carrying insurance and the yield spread.
T h e r e l a t i v e i n c r e a s e in the s u p p l y of h i g h - g r a d e
bonds is the primary factor behind the change in yield
spreads. The result is lower prices (and higher yields)
for these securities in comparison with lower-rated,
stand-alone securities. Reinforcing this effect is the fact
that capturing the extra yield on uninsured bonds is attractive for institutional investors. As insured volume
increases, the volume of bonds rated from Baa to A1
declines. With the declining volume of uninsured issues, competition for these bonds becomes m o r e in-

tense and their prices are bid up (yields are bid down)
relative to those in the insured market.
Besides the fact that spreads tend to be lower when
the level of rates declines, another potentially important
factor may be at work in these data on yield spreads.
A s Chart 5 shows, there was a dramatic increase in
leverage ratios during the late 1980s. The chart depicts
both total exposure and leverage ratios in this insurance
industry f r o m 1987 through 1992. A l t h o u g h the increase in leverage ratios has slowed in the past f e w
years, investors, concerned about the high risk of insurers, may still discount some issues that are given the
Aaa rating only because of their insurance guarantees.
In other words, it is generally accepted that investors
do not view an insured Aaa bond as having the same
value as one with a stand-alone Aaa rating, given the
difference in default risk of the issues and the risk of
insurers. Because the more recent Aaa samples contain
a large percentage of insured bonds, the average Aaa
bond may carry a higher yield than it would if only
stand-alone Aaa's were in the sample.
This last point also has a direct bearing on whether
issuers can still expect to benefit f r o m the use of private insurance guarantees. On the one hand, the growing
pool of retail investors and the higher value they place

Chart 3
N e w M u n i c i p a l Issues, Insured v e r s u s Total V o l u m e
(1983-92)

Millions

Percent Insured

40

140,000

20,000

1983

1984

1985

• Insured Volume

1990

1986
Total Volume

1991
Percent Insured

S o u r c e : M o n t h l y data for insured issues and total v o l u m e of n e w long-term m u n i c i p a l issues w e r e supplied by Securities Data C o m p a n y .
Quarterly figures are the sum of the monthly data for the preceding three months, a n d the percent insured is the quotient of the
quarterly figures for insured v o l u m e and total v o l u m e . C a l c u l a t e d by the Federal Reserve Bank of Atlanta.

Federal Reserve Bank of Atlanta



Economic Review

31

Chart
Yield Spread and

4

Insured V o l u m e as a P e r c e n t a g e of Total

N e w

Issue

Volume

(1983-92)
Baa-Aaa

Spread

(Percent)

Percent Insured

2.0

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

S o u r c e : T h e yield spread is the quarterly average of the spread b e t w e e n B a a a n d A a a m u n i c i p a l issues, as recorded monthly by M o o d y ' s
Investors Service. T h e percent insured is the quotient of the quarter's insured n e w issue v o l u m e divided by the quarter's total n e w
issue v o l u m e , as supplied by Securities Data C o m p a n y .

Chart

5

Total Exposure and Leverage Ratios of Municipal

Bond

B i l l i o n s of D o l l a r s

Insurers
Leverage Ratio

650

i

r

550

--

140

450

-

130

350

-

-

120

250

--

-

110

150

4-

100
1987

1

1989
Net Exposure

S o u r c e : Fitch Investors Service, Inc., Bond

32

Economic




150

Review-

Insurers'

New

Horizons,

1990

1991

1992

Leverage Ratio
Fitch Research Financial Institutions Special Report ( M a y 3, 1993): 13.

September/October

1993

on insurance would seem to suggest increased benefits. However, any such benefits may be offset by the
perceived riskiness of private insurers, with factors
such as higher leverage ratios tending to reduce the
value of the insurance. These potentially conflicting
factors complicate the question of whether issuers can,
on average, reduce their interest costs by purchasing
insurance in the current environment. The discussion
that follows investigates these pricing issues by examining a recent sample of state and local bond issues
from the state of Florida.

Costs and Benefits to Issuers in Florida
Interest cost savings from bond insurance may still
be possible when retail, rather than institutional, customers are the main investors. Indeed, as mentioned
above, earlier studies that use data from a time when
most municipal bond purchasers were institutions may
understate the potential benefits to issuers in today's
market, depending on whether these new investors perceive private insurers as having taken on too much risk.

lar issue to certain market and issue characteristics.
This topic has been the subject of several empirical
studies, and a fairly standard list of market factors and
issue characteristics has emerged as candidates for independent variables. Historically, the TIC of a municipal bond has depended largely on such things as the
issue's rating, size, maturity, revenue source, and the
prevailing level and term structure of interest rates at
the time of pricing. Other factors include the bidding
process, redemption provisions, regional segmentation
of the market, and the supply of tax-exempt bonds relative to taxable bonds at the time of pricing. With minor modifications, the research reported here also used
a subset of these variables. T h e model used for estimating the TIC of uninsured issues is given by:

TICn = a + blLNSIZEN + b2LNFMATN
+ b3TOPTIERN

+ b4BOTTIERN

+ b5GON + b(IRATEN

(2)

+ e,

where the specific definitions of these variables are as
given in Table 1.

where ^C" U N I N S U R E D is the estimated borrowing cost
for an issue given its underlying credit rating, and the
p a r a m e t e r s t a k e n f r o m the u n i n s u r e d s a m p l e a n d
are the actua
T I C INSURED
l borrowing cost for the insured issue. If NET BENEFITS
is, on average, positive, some credibility would be lent to the hypothesis
that interest cost benefits still accompany the use of insurance.

P r o x i e s f o r the s i z e a n d m a t u r i t y of t h e i s s u e
(LNSIZE and LNFMAT) are expected to be positively
correlated with the interest cost—the former because
larger issues can only be absorbed by offering the issue at a lower price, and the latter because it is typically believed that investors demand a yield premium in
order to hold longer-term bonds. TOPTIER and BOTTIER (Bottom Tier) are d u m m y variables for ratings
classifications. Issues are grouped into high investmentgrade and lower investment-grade classes. With nonrated r e v e n u e b o n d s as the c o m p a r i s o n g r o u p , the
coefficients for both TOPTIER and BOTTIER should
be negative. Consistency requires that the coefficient
for TOPTIER is more negative than that for BO'ITIER
(that is, TOPTIER should carry a lower yield). GO is a
d u m m y variable for general obligation bonds, a factor
that should reduce an issue's TIC in the primary market because bonds carrying the full faith and credit of
the issuer are g e n e r a l l y c o n s i d e r e d less risky than
bonds w h o s e repayment d e p e n d s on a single source
(revenue bonds). IRATE is a proxy for the prevailing
level of interest rates. Its sign should be positive since,
other things being equal, higher rates on alternative securities, such as Treasuries, force all other issuers—including state and local g o v e r n m e n t s — t o increase
interest rates offered on their securities. This move, in
turn, causes the measured 77Cs to rise.

The specification used here basically follows that in
Kidwell, Sorensen, and Wachowicz (1987). The first
model relates the overall borrowing cost for a particu-

Table 1 contains the coefficient estimates and other
relevant information f r o m the estimation of equation
(2) f o r a s a m p l e of 132 uninsured bonds issued by

This section reports on a test of the hypothesis that
insurance generates positive cost savings. The method
used starts by estimating a model of the factors that influence costs for noninsured Florida municipal bonds
issued during the last three years. The next step inv o l v e s u s i n g the m o d e l ' s p a r a m e t e r s to predict an
" u n i n s u r e d " borrowing cost for a sample of insured
Florida municipal bonds issued over roughly the same
period of time. This approach provides, in essence, a
cost for the insured issues as if they had come to the
market uninsured. The cost of insurance is already factored into the estimate of the borrowing cost (the TIC).
Therefore, the estimated net interest cost benefit of insurance accruing to the issuer is simply the following:
NET BENEFITS

= TiC„msmED

Federal Reserve Bank of Atlanta



- TlCImmED

(1)

Economic Review

33

Table 1
R e g r e s s i o n for True Interest C o s t ( T I C )
U n i n s u r e d Florida B o n d s in t h e Primary M a r k e t
(1990-92)

TIC = a + b} LNSIZE
+ b4 BOTTIER

+ b2 LNFMAT+

b,

TOPTIER

+ b- CO + bb IRA TE + e
Coefficients

T-Ratio

Constant

-3.5516

-2.0738 3

LNSIZE

-0.1559

-2.2874 a

LNFMAT

0.7105

6.2896' 1

TOPTIER

-1.0922

-3.8804"

BOTTIER

-0.9474

-3.8120 a

CO

-0.6125

-2.8131 a

IRATE

1.2672

6.361 3 a

Adjusted R2

0.5545

M e a n TIC (percent)

7.0694

Number of Observations
LNSIZE

132

= total dollar size of the issue (10 5 ) in natural loga-

rithms.
LNFMAT

= years to final maturity in natural logarithms.

TOPTIER

= d u m m y variable, w h i c h equals I if the bond has

a M o o d y ' s credit rating of Aa or higher; 0, otherwise.
BOTTIER

(Bottom Tier) = d u m m y variable, w h i c h equals 1 if

the issue's credit rating is investment grade but less than
Aa; 0, otherwise. Noninvestment grade (lower than Baa-)
and nonrated bonds are the excluded set.
CO = dummy variable, w h i c h equals 1 if the issue is a general obligation issue and 0 if it is a revenue bond issue.
IRATE = prevailing level of interest rates as measured by the
average rate on the twenty-year Treasury Bond in the week
corresponding to the sale of the bond issue.
a

Significant

at the 5 percent

significance

level.

S o u r c e : Regressions run o n data from actual debt financings of governmental district a n d agencies in the State of Florida, as c o l l e c t e d by the
State of Florida D i v i s i o n of B o n d Finance. N o n i s s u e specific data
w e r e provided by the Federal Reserve System.

state and local g o v e r n m e n t s in F l o r i d a f r o m 1990
through 1992. All of the independent variable^ have
the expected signs, and their coefficients are statistically significant at conventional levels. For example,
the c o e f f i c i e n t of BOTTIER is - . 9 5 . T h u s , the data

34



used here imply that an uninsured, lower-tier investm e n t - g r a d e b o n d will have, other t h i n g s b e i n g the
same, a TIC that is on average 95 basis points lower
than that of a nonrated revenue bond. The coefficient
on TOPTIER ( - 1 . 0 9 ) implies an additional 14 basis
point advantage to being a top-tier investment-grade
bond.

Economic Review-

/ n t e r e s t Cost Analysis
Issues from the State of Florida were chosen for this
study for several reasons. First, Florida has historically
been among the top four states in terms of the volume
of newly issued insured bonds. Florida also led all
states in newly insured volume during the first half of
1992. Interestingly, insured volume represents around
two-thirds of total new-issue volume in Florida. 2
Table 2 presents the results of the interest cost/benefit analysis for a sample of 219 insured Florida issues. This group consists solely of bonds with underlying stand-alone credit ratings that are a bottom-tier
investment grade. The procedure first involved estimating stand-alone TICs for the insured issues by multiplying the parameters of equation (2) by the relevant
independent variables. The difference between these
estimated TICs and the actual 77Cs for the issues are
estimates of the net interest cost benefits accruing to
the issuer (equation f l j ) through the attachment of
bond insurance. The mean predicted TIC was 6.90894
percent while the actual average TIC was 6 . 6 1 8 9 7 .
The mean net interest cost benefit, 28.997 basis points,
is statistically significant using a variety of tests.
As an example, the Wilcoxon Z-statistic was calculated to test the h y p o t h e s i s that the d i f f e r e n c e s between the estimated and actual TICs were less than or
equal to zero. T h e Wilcoxon is a so-called matched
pairs/signed rank test. 3 In essence, it attempts to answer the question of whether the observed n u m b e r of
positive values of TIC - TIC could have occurred by
chance. T h e Z-statistic is found significant at the 5
percent level, and the hypothesis that insurance provided no interest cost savings is thus rejected. In other
words, there is less than a 5 percent likelihood that the
extraordinary number of positive values of TIC - TIC
could have occurred by chance.
Several observations can be inferred concerning the
distribution of benefits across default risk classes. Previous data (Table 1) suggest that uninsured bonds with
credit ratings characteristic of the top tier (Aa ratings
or better) trade at yields about 14 basis points lower

September/October 1993

Table

2

D i s t r i b u t i o n o f N e t B e n e f i t s o f I n s u r a n c e in T e r m s
Interest Cost Savings for Insured Florida Municipal

of
Bonds

(1990-92)

All
(Number

Issues
of Issues:

219)
Percent

Number
11

5.02

9

N e t Benefits < - 4 0 b.p.a

4.11

- 4 0 b.p. < N e t Benefits < - 2 0 b.p.
- 2 0 b.p. < N e t Benefits < 0 b.p.

21

9.59

0 b.p. < N e t Benefits < 2 0 b.p.

32

14.61

20 b.p. < N e t Benefits < 4 0 b.p.

47

21.46

4 0 b.p. < N e t Benefits < 60 b.p.

50

22.83

60 b.p. < N e t Benefits < 80 b.p.

31

14.16

N e t Benefits > 8 0 b.p.

18

8.22

M e a n E s t i m a t e d TIC: 6 . 9 0 8 9 4 5

M e a n Net Benefits: 0 . 2 8 9 9 7 2

M e a n I n s u r e d TIC:

Wilcoxon

6.618973

Subsample with Insurance
(Number

of Issues:

Z-Statistic:-11.9134b

Prices

100)
Number

Percent

Net Benefits < - 4 0 b.p.

0

0.00

- 4 0 b.p. < N e t Benefits < - 2 0 b.p.

4

4.00

- 2 0 b.p. < N e t Benefits < 0 b.p.

12

12.00

0 b.p. < N e t Benefits < 2 0 b.p.

18

18.00

2 0 b.p. < N e t Benefits < 4 0 b.p.

20

20.00

4 0 b.p. < Net Benefits < 6 0 b.p.

28

28.00

15

15.00

3

3.00

60 b.p. < N e t Benefits < 80 b.p.
N e t Benefits > 80 b.p.
M e a n E s t i m a t e d TIC:
M e a n I n s u r e d TIC:

3

The notation

b

The calculated

Note:

"b.p."

The base data set was composed
tained

M e a n Cost of Insurance: 0 . 0 6 5 8 6 3

stands for basis points; one basis point equals .01 x 1 percent.

value is less than -2.33, the critical

were estimated

M e a n Net Benefits: 0 . 3 2 2 5 8 6

6.844326

6.521741

value for the one-tailed

of insured issues offered by Florida

for all issues. All issues had underlying

in Moody's

vided by insurance

Municipal
is simply

Finance

Manual

the estimated

Wilcoxon

Z-test statistic at the 5 percent

municipalities

from 1990 through

ratings of at least Baa but less than Aa, as determined

and conversations

TIC minus the actual

with officials

of the individual

significance

level.

1992, and "stand-alone"

municipalities.

from information
The net benefit

TICs
conpro-

TIC.

S o u r c e : P r e m i u m rates for individual insurance policies w e r e c o l l e c t e d by the State of Florida D i v i s i o n of B o n d F i n a n c e a n d by individual
Florida municipalities.


Federal Reserve Bank of


Atlanta

Economic

Review

35

than bottom-tier bonds. Thus, the estimated benefits of
insurance to top-tier issuers is u n d o u b t e d l y smaller
t h a n t h o s e in the s a m p l e of b o t t o m - t i e r u n i n s u r e d
bonds. However, the net benefits varied considerably
even across the bottom-tier bonds considered here. Indeed, while the mean and median estimates of net benefits are positive, almost 20 percent of distribution
falls in the negative range.
Some independent data on the gross cost of insurance to issuers were also examined. In particular, insurance premiums were calculated for 100 of the 219
insured issues. P r e m i u m s f o r this s a m p l e w e r e assumed to be paid in full at the time of issuance. As rep o r t e d b y the A s s o c i a t i o n of F i n a n c i a l G u a r a n t y
Insurers in its 1991 annual report, up-front premiums
constituted almost 98 percent of all direct premiums
written by the industry f r o m 1988 through 1991.
T h e cost of the insurance premium (as a percentage) is calculated using the following formula:
IC

=
PREMIUM

PREMIUM
(PARAMOUNT • AVERAGE

^
LIFE)'

net cost of insurance. As Table 2 shows, the subsample
for w h i c h costs w e r e collected had a m e a n of only
6.586 basis points, nearly 5.2 basis points less than the
c o r r e s p o n d i n g m e a n i n s u r a n c e cost in the K i d w e l l ,
Sorensen, and Wachowicz sample. This figure, interestingly, almost exactly equals the difference between
the estimated net benefits the two studies report. It appears, t h e r e f o r e , that c o m p e t i t i o n a m o n g credit enhancers has driven down the cost of insurance. Such
an increase in competition is consistent with the fact
that insurers have increased their leverage ratios; in
other words, one approach to offsetting lower profits
per dollar of insurance involves writing more guarantees per dollar of stockholders' invested capital. Other
factors, of course, also play a role. Nevertheless, the
evidence presented here is broadly consistent with the
hypotheses that issuers save on costs by using insurance and that these savings are, on average, inversely
related to stand-alone credit ratings.

(3)

Conclusion
where PREMIUM is the cost of the insurance policy,
prorated to the level of debt service; PAR AMOUNT is
the total face value of a bond issue; and AVERAGE
LIFE is the average maturity of a bond issue, after factoring in sinking f u n d provisions. Using equation (3),
the m e a n gross cost of insurance for this subsample
was .06586 percent or 6.586 basis points. Since the estimated net benefits already have costs f a c t o r e d in,
these premium estimates imply that the gross benefits
of insurance were, on average, in the range of 35 to 4 0
basis points.
Despite the differences in the dates covered and the
composition of the data sets, these results compare favorably with those of previous studies, especially Kidwell, Sorensen, and Wachowicz (1987). The mean net
benefit of the overall sample in the present study was
about 29 basis points (see Table 2); the m e a n net benefit for the entire Kidwell, Sorensen, and Wachowicz
sample was 22.4 basis points. Given the differences in
the samples, this degree of stationarity is somewhat
surprising. T h e s a m p l e of general obligation bonds
used by Kidwell, Sorensen, and Wachowicz benefits
less on average from insurance than the revenue bonds
used in this research, but that result is expected, given
that the spread between Aaa-rated general obligations
and similarly rated revenue bonds is positive.
T h e most striking d i f f e r e n c e between the results
here and those of earlier studies involves the estimated

36

Economic Review-




The municipal bond market has changed dramatically
in the past twenty years. What was once an uninsured
market involving mostly institutions has been transformed into one in which insured municipal issues are
primarily held by retail investors. This article has considered the potential sources of this changing investor demand and whether the shift has significantly altered the
benefits of insurance to issuers. Results suggest that tax
changes in the 1980s caused the shift in the composition
of investor demand. Interestingly, when compared with
earlier studies, the data used here suggest that net benefits to issuers have increased. Moreover, the additional
benefits (roughly 5 to 7 basis points) seem to have come
from lower prices associated with the cost of insurance.
Increased competition among underwriters has led to
lower average fees paid by issuers for the insurance
guarantee.
T h e s e facts are, by and large, g o o d n e w s for issuers. However, to some extent insurers have tried to
offset the declining margins by increasing their levera g e ratios. T h i s fact and its i m p l i c a t i o n s f o r risk,
along with the increasing fiscal d e m a n d s placed on
state and local governments, create a situation calling
for greater regulatory awareness of the changing volu m e and pricing structure for municipal bond insurance.

September/October 1993

Box
The All-in True Interest Cost {TIC) measure was used
in this study in order to provide the most accurate measure of the issuer's true borrowing cost. The main advantage of the TIC is that it incorporates, in contrast to the
historically more often used Net Interest Cost (NIC), the
time value of money into the cost calculation.
The historical acceptance of the NIC can be attributed in part to the complicated structure of municipal
bonds. Unlike corporate and Treasury securities, municipal financings are typically structured with bonds in
separate serial and term maturities, with the coupon
rates varying across maturities. This structure makes the
quick calculation of an issue's time-value, weighted- interest cost extremely difficult without the aid of a computer. Although it neglects the time value of money, the
NIC calculation enables underwriters to make a firstpass calculation of the overall cost. The NIC is found by
taking the sum of the total interest payable and the discount from par, or face, value (or the difference between
interest payable and the premium over par value) and dividing by the par value of the issue times the issue's average maturity. This approach is analogous to the yield
to maturity approximations found in many finance textbooks.
The TIC is the rate at which future debt service payments must be discounted in order to equal the net bond
proceeds to the issuer. The gross bond proceeds typically
include the par amount, price discount or premium, and
accrued interest. The underwriting spread, the cost of insurance (if applicable), and issuance costs are then deducted from the gross bond proceeds to produce a net
figure. The calculation of the TIC is similar to the calculation of a bond's yield-to-maturity or internal rate of return. Future cash flows—in this case semiannual debt
service payments—are individually discounted back to
the delivery date at the true interest cost, which is found

through trial and error. Mathematically, it can be depicted as
NBP=t

, -H)

:
where
NBP

=

DSi
r
frac

=
=
=

i

=

t

=

•

the net bond proceeds to the issuer at the
date of delivery;
debt service payment at period i;
annualized all-in true interest cost;
period between the date on which interest
first begins to accrue and the date of delivery, expressed in semiannual periods;
the number of semiannual compounding
periods between the date of the current
debt service payment and frac,
total number of semiannual compounding
periods.

The difference between the TIC and NIC measures
lies in the fact that the net interest cost measure is independent of the timing of the interest payments. The interest can be paid out in full the first year or the last
year, and the NIC would remain unchanged. In the TIC
calculation, however, a dollar of interest today would be
worth much more than a dollar of interest five years
from now.
In addition to the fact that the TIC has now become
the norm in competitive bidding, it also has an additional
advantage over the NIC for this particular study. Bond
insurers have been reluctant in the past to divulge premium rates. However, unlike the NIC, the TIC includes the
cost of insurance. Therefore, a separate calculation for
the cost of insurance is no longer necessary in order to
examine the potential net benefits provided by insurance.

Notes
1. There arc a number of reasons that individual investors may
be more concerned about default risk than their institutional
counterparts. Higher risk aversion, less diversification, and
limited information are all possibilities that could explain
this phenomenon.
2. Florida is cited as one of the four most prolific states in issuing municipal bonds and insured municipal bonds by the
Bond Buyer in its "Quarterly Statistics Supplement" on July 27, 1992, and by all four major credit enhancers in their
annual reports for 1991.

Federal Reserve Bank of Atlanta



3. The Wilcoxon Z-statistic is calculated using the following
equation:
z_

T - N(N + 1) / 4
~ «JN(N + l)(2N + 1 ) / 2 4 '

where T is the sum of the negatively signed differences and
N is the number of pairs.

Economic

Review

37

References
Addison, Thomas P., Jr. "Bond Insurance and Quality Yield
Spreads." Donaldson, Lufkin, and Jenrette Securities Corporation, Municipal Market Comment, October 1, 1990, 1-7.
AMBAC Indemnity Corporation. 1991 Annual Report. 1992.
Association of Financial Guaranty Insurers. Report of Combined Financial Results and New Business Written for the
Years Ended December 31, 1991, 1990, 1989, and 1988.
Bond Buyer. "Decade of Municipal Finance." Seminnual Review, July 1992.
Board of Governors of the Federal Reserve System. Flow of
Funds Accounts, Tax-Exempt Securities Holdings.
Braswell, Ronald C., E. Joe Nosari, and Mark A. Browning.
"The Effect of Private Bond Insurance on the Cost to the Issuer." Financial Review 17 (November 1982): 240-51.
Cole, Charles W., and Dennis T. Officer. "The Interest Cost Effect of Private Municipal Bond Insurance." Journal of Risk
and Insurance 48 (September 1981): 435-49.
"A Farewell to Municipals." ABA Banking Journal (February
1987): 35-36.
Feldstein, Sylvan G. "Municipal Bond Insurance and Pricing."
In Municipal Bond Handbook, edited by Frank J. Fabozzi
and Sylvan G. Feldstein, 783-97. Homewood, 111.: Dow
Jones-Irwin, 1983.

38
Economic Review


Financial Guaranty Insurance Company. 1991 Annual Review.
1992.
Financial Security Assurance. 1991 Annual Review and Company Profile. 1992.
Fitch Research. "Bond Insurers' New Horizons." May 3, 1993.
Forbes, Ronald W., and Michael H. Hopewell. "Private Municipal Bond Insurance: A Theoretical and Empirical Analysis." Presented at the annual meetings of the Western
Economic Association, San Francisco, 1976.
Hsueh, L. Paul, and P.R. Chandy. "An Examination of the
Yield Spread between Insured and Uninsured Debt." Journal of Financial Research 12 (Fall 1989): 235-43.
Kidwell, David S., Eric H. Sorensen, and John M. Wachowicz,
Jr. "Estimating the Signaling Benefits of Debt Insurance:
The Case of Municipal Bonds." Journal of Financial and
Quantitative Analysis 22 (September 1987): 299-313.
Municipal Bond Insurance Association. 1991 Annual Report.
1992.
Public Securities Association. Fundamentals of Municipal
Bonds. 4th ed. 1990.

September/October 1993

Commercial Bank
Profits in 1992
B. Frank King

n 1992 banks' profitability reached its highest levels in more than two
decades, boosted by widening interest margins c o m b i n e d with decreasing interest rates and easing loan problems. Banks in the Southeast shared this g o o d f o r t u n e . 1 T h e n a t i o n ' s (and r e g i o n ' s ) c o m mercial banks took advantage of a relatively steep yield curve and
mild loan demand to shorten the maturity of their liabilities, reduce deposit
growth, and increase interest margins. These positive steps, along with several other factors related to lower interest rates and an improving economy,
accounted for most of the substantial increase in these b a n k s ' returns on assets and equity.
Although successful in 1992 and through most of 1993, banks' current
strategy may hold germs of problems. If in the future interest rates go above
1993's falling long-term rates—particularly if the spread between short-term
and long-term rates continues to narrow from the extreme 1992 gap—earnings from these strategies could be jeopardized.
The author is vice president
and associate director of research at the Atlanta Fed. He
thanks Sheila Tschinkel and
Larry Wall for helpful comments and Robert Goudreau
and S herley Wilson for
research assistance.

Federal Reserve Bank of Atlanta



In the 1992 environment of modestly declining interest rates, banks of all
sizes were able to cut interest expenses (per dollar of assets) relative to 1991.
The expense reduction exceeded banks' 1992 decline in tax-equivalent interest revenue. T h e resulting increase in risk-adjusted net interest margins
played a m a j o r role in raising bank profitability across the board from 1991
to 1992. 2 (Tables 1-4 provide interest margin data on the nation's banks for
the years 1988 through 1992. A detailed discussion of the various profitability measures and their significance can be found in the appendix.)

Economic Review

39

Intermediation Profitability
Falling interest expenses in 1992 resulted from two
m a j o r sources. As Chart 1 demonstrates, market interest rates for debt of all maturities declined modestly in
1992. These declines affected the rates that banks had
to pay for f u n d s . W h i l e rates fell across the board,
those for short maturities moved further below those
for long maturities. For instance, the spread between
y i e l d s on f i v e - y e a r U.S. T r e a s u r i e s and equivalent
three-month yields averaged 268 basis points in 1992.
This spread compares with average spreads of 184 basis points in 1991 and 62 basis points in 1990.
Banks took advantage of the steeper yield curve to
reduce interest expenses further through pricing and
promotion decisions that drove off deposits originally
in longer-maturity certificate accounts and increased
the proportion of their liabilities in shorter maturities.
The share of banks' deposits in demand deposits, other
transactions accounts, money market deposit accounts,
and savings accounts increased to 61.6 percent f r o m
56.0 percent in 1991. (Table 5 presents deposit shares
for the nation's banks.) Allan D. Brunner and William
B. English (1993, 655) found that time deposit maturities actually increased during 1992 but not enough to
overcome reductions in the share of time deposits in
total bank liabilities. Similar changes also took place
in 1991 as the yield curve steepened and supported a
modest increase in interest margins.

Chart 1
Interest Rates o n U . S . Treasuries by Maturity
Percent

S o u r c e : Federal

Reserve

Bulletin,

T a b l e 1.35; three-month bill adjusted to

b o n d equivalent.

4
 0


Economic Review-

B a n k s were aided in increasing margins in 1992
by several factors. For one thing, they faced diminished competition f r o m savings and loans, particularly f r o m t r o u b l e d S & L s p a y i n g h i g h interest rates.
Regulatory pressure on weak banks and S&Ls, including pressure to raise capital/asset ratios, probably diluted aggressive competition f r o m these institutions.
In addition, lower interest rates contributed to higher
interest margins by reducing the cost of carrying other (that is, foreclosed) real estate and nonperforming
loans.
Interest rate and economic changes during the year
were e c o n o m y w i d e in scope, consequently affecting
banks of all sizes. Banks in each size class delineated
here took advantage of the economic changes. 3 Each
dropped interest expenses per dollar of interest earning
assets. The smallest decline, of 132 basis points, was
for banks in the smallest size category. The largest declines, of around 165 basis points, c a m e in the t w o
largest categories.
Although liability maturities shortened in 1992, evidence on b a n k s ' asset maturities suggests that the
changes were small (Brunner and English 1993, 665).
The proportions of debt security portfolios at the short
e n d — t h a t is, securities with less than three m o n t h s
and with three m o n t h s to one year until maturity—increased slightly from 1991 to 1992. Debt securities in
the one- to five-year maturity category also increased
while debt in the more-than-five-years maturity category declined. T h e maturity structure of loans and
leases remained relatively stable f r o m 1991 to 1992.
Reducing interest expense by shortening the maturity of their liabilities allowed b a n k s of all sizes to
avoid a c o m m e n s u r a t e decline in interest earnings.
Overall interest revenue per dollar of interest earning
assets fell 123 basis points, but interest expense fell
159 basis points. Further improvement in risk-adjusted
interest margins came from substantial reductions in
loan-loss expense, which is subtracted from interest
revenue here in order to adjust to the risk inherent in
each b a n k ' s loan portfolio. 4 An improving e c o n o m y
and several years of cleansing bad loans and of limited
commercial and apartment construction (and lending)
led to improved loan-loss records. Consequently, nonp e r f o r m i n g loans retreated and l o a n - l o s s e x p e n s e s
declined sharply in 1992. Overall the average riskadjusted margin rose 65 basis points to 3.76 percent.
The largest category of banks (assets greater than $ 1
billion) showed the largest gains, with an increase of
76 basis points. However, consistent with the historical
record, their adjusted interest margin remained well
below that of other banks.

September/October 1993

Other Sources of Profits
Several factors influenced the share of gains from
wider interest margins that showed up in returns on assets and equity, though none had a large effect. For
banks as a whole these factors counteracted each other
and made no net contribution to earnings. Gains from
securities sales, spurred by declines in interest rates
over the past three years, continued to add revenues as
did the l o n g - t e r m t r e n d of i n c r e a s i n g f e e i n c o m e .
However, these gains were almost exactly offset by increases in noninterest expenses. (See Tables 6-8 for
data on noninterest income and expenses.)
In 1992 as in 1991, declining interest rates induced
gains in the values of securities held in banks' portfolios and allowed m o d e s t additions to their earnings
through securities sales. Such earnings increased in
1992 over 1991 for banks of all sizes; however, the securities gains and their increases were small relative to
gains in net interest margins.
Banks of all sizes also recorded m o d e s t gains in
earnings f r o m noninterest income in 1992. Banks in
the largest and smallest size categories increased these
earnings more than banks of other sizes. 5 The rising
n o n i n t e r e s t s h a r e of b a n k s ' r e v e n u e s h a s b e e n a
decade-long trend, driven by banks' declining share of
the financial intermediation business, separation of
service charges from credit charges, and the entry into
new markets as well as the introduction of new products. Banks have come to depend increasingly on inc o m e f r o m this e a r n i n g s category; the greatest dependence is among the largest banks. 6 Their noninterest income as a percent of total assets has increased by
31 percent during the last five years and stood at 2.20
percent of total assets in 1992.
While fee-based income has been rising as a perc e n t a g e of a s s e t s and total e a r n i n g s o v e r the last
decade, noninterest e x p e n s e s have also increased in
p r o p o r t i o n to a s s e t s . Until the last t h r e e years the
growth rate was slow. Since 1989, however, these figures have j u m p e d , with large banks recording by far
the largest increase. In 1989 large banks' ratio of noninterest expenses to total assets was similar to that of
all other banks except the smallest, which had higher
expenses. Since that time the largest banks' noninterest expenses have risen well above those in other categories of banks exceeding $25 million in asset size
to a level almost equal that of the smallest banks.
For most banks the recent increase in noninterest
expenses seems to have resulted from a combination
of factors connected with asset quality deterioration in

Federal Reserve Bank of Atlanta




the late 1980s and early 1990s. Both wage and salary
expense and occupancy expenses as percentages of total assets have remained stable since 1988 for each
bank size group. T h e other expense category, which
includes losses on other real estate and sales of loans
as well as deposit insurance fees, has, however, increased by one-third over this period. 7 This j u m p in
noninterest expenses essentially wiped out increased
income f r o m securities gains and noninterest income
in 1992, leaving virtually no net effect on return on assets and equity.

T h e Bottom Lines
Dominated by interest margin gains, overall return
on assets (ROA) and equity (ROE) rose substantially
in 1992. (See Tables 9 and 10 for data on returns on
assets and equity.) Last year's ROA of 0.96 percent
was at the highest level in at least the past two decades
while the 1992 R O E of 13.33 percent almost equaled
1988's level of 13.50 percent, the highest R O E of the
1980s. 8 Banks of all sizes m a d e large gains in both
measures of profitability. A f t e r recording the lowest
returns in 1991, the two size classes of banks with assets greater than $500 million gained most in 1992 as
sharp declines in loan-loss expense reinforced rising
interest margins. The smallest banks completed their
recovery to earnings levels similar to other banks' after languishing well below all but the largest since
1984. Their return was aided most by their ability to
limit declines in interest revenue and boost their noninterest earnings more than other banks.

Distribution of Gains b y
Bank Condition
Banks of all sizes and conditions experienced 1992's
impressive profitability gains. The relatively equal distribution of improvement—among banks that had been
performing well and those that had run into trouble—
tends to reduce concern about the health of the industry.
T h i s analysis of the y e a r ' s profitability distribution
ranks each bank from lowest to highest return on assets
and records the return on assets of the bank at the top of
the lowest (first) quartile (the best of the poor performers), the median bank (representative of middle-level
performers), and the bank at the top of the third quartile
(weakest of the best p e r f o r m e r s ) . C o m p a r i n g t h e s e

Economic Review

41

banks' returns with those of previous years clarifies the
picture of last year's distribution of gains.
As one might expect, the representative banks in all
three of these profitability classes had improved earnings in 1992. (See Tables 11-13 for returns on assets
of banks in the low, medium, and high p e r f o r m a n c e
categories.) Although banks in the classification showing highest profits improved most, they did not far outgain the median or least profitable banks. Only in the
largest size class (which also has the second fewest
banks) did one profitability group substantially outgain the others. Here, the banks with the lowest returns on assets improved by more than 40 basis points,
nearly doubling the gain by the median profitability
bank and more than doubling that of the bank at the
beginning of the highest quartile of profitability.

2?anks in the Southeast
The forces that generally accounted for improved
bank performance across the nation also played out in
the S o u t h e a s t with similar results. E x c e p t in a f e w
special cases, southeastern banks generally followed
national patterns. 9 (See Tables 14-22 for data on profitability of banks in the Southeast.) The m a j o r deviations appeared at both ends of the size spectrum and
were of quite different natures.
The region's small banks made large gains from a
lagging position. (See Table 21 for data on the ROA of
small s o u t h e a s t e r n b a n k s . ) For several years these
b a n k s l a g g e d o t h e r small b a n k s (as well as larger
banks) in returns on assets and equity. In 1991, for instance, the smallest banks in the region earned on ave r a g e only 0.14 percent on assets c o m p a r e d with a
0.62 percent ROA earned by banks in the smallest size
category for the nation as a whole and a 0.66 percent
ROA for all southeastern banks. In 1992, the region's
smallest banks' ROA soared to an average of 0.77 percent. At the same time, however, they continued to und e r p e r f o r m b a n k s of similar size in the rest of the
nation and larger banks in the region. While the continuing lag reinforced doubts about the long-run viability of some of the region's small banks, it should be
noted that in 1992 ROA moved closer to the national
average of 0.95 percent for small banks and to the 1.06
percent average for all banks in the region.
Last year's profitability gain for the region's smallest
banks is largely attributable to vastly improved performance of newer banks in Florida and Georgia. Robert
G o u d r e a u and B. Frank K i n g (1991) and G o u d r e a u

42




Economic Review-

(1992) noted that the large number and sad plight of
banks established in Florida and Georgia since 1987
have been the m a j o r drags on overall small bank profits. A s they continued to mature in a more hospitable
economy, the performance of these banks in 1992 approached that of older small banks and of other southeastern banks; most of the new b a n k s b e c a m e profitable. Earnings for the small bank category as a whole
were boosted relative to banks of other sizes by reduced noninterest expenses and gains in noninterest
income. These gains would have raised returns even
more had not the smallest banks continued to record
higher levels of loan-loss provisions than all but the
largest of their larger competitors.
T h e region's largest banks lost a bit of ground to
their counterparts in the rest of the country in 1992,
nonetheless retaining the lead in ROA and R O E enjoyed for four years running. In 1991 their ROA lead
was 16 basis points (36 percent), but the margin was
closed to 12 basis points (13 percent) in 1992's strong
recovery.
Rising profitability, driven by higher adjusted net
interest margins, characterized banks in each of the region's states in 1992. (See Tables 23-28 for data on the
profitability of the region's banks by state in the 198892 period.) Banks in Alabama and Georgia fared best,
as they have for most of the past decade. In Georgia,
i m p r o v e m e n t s a m o n g the smallest institutions aided
p r o f i t a b i l i t y s u f f i c i e n t l y f o r the s t a t e ' s b a n k s as a
whole to surpass Alabama banks slightly in ROA.
B a n k s in L o u i s i a n a , however, achieved the m o s t
spectacular gains. After a decade of ROAs that hovered around zero, the state's banks achieved a level
somewhat above the regional average. This leap occurred mainly b e c a u s e Louisiana's banks were able
to reduce loan-loss expenses by 60 percent, a m u c h
l a r g e r c u t t h a n b a n k s in t h e r e g i o n ' s o t h e r s t a t e s
achieved. T h e improved loan-loss p e r f o r m a n c e was
not the result of a m o r e resilient state economy, however; in fact, the Louisiana e c o n o m y lagged those of
t h e r e g i o n ' s o t h e r s t a t e s in 1992 ( S t a c y K o t t m a n
1992). It is m o r e likely that L o u i s i a n a b a n k s ' bad
loans had been charged off or acquired by the F D I C in
previous years.

S o m e Implications of
1992's Performance
I m p r o v e d p e r f o r m a n c e by b a n k s of all sizes and
conditions in 1992 allowed most of the industry to in-

September/October 1993

crease its capital ratios; it also allowed banks' shareholders, management, and deposit insurers to breathe
a little easier. W h e t h e r the profitability improvement
has resulted f r o m falling interest rates and shrinking loan losses or f r o m the significant reductions of
e x c e s s c a p a c i t y in the f i n a n c i a l s e r v i c e s i n d u s t r y
o v e r the past several y e a r s — a n d both f o r c e s h a v e
probably contributed—some of the practices followed
to achieve these gains may hold threats for banks' future performance.
There is evidence that banks achieved wider interest margins at least in part by shortening the maturity
of their deposit liabilities (see above). U n l e s s they
were e f f e c t i v e l y h e d g e d , these a c t i o n s have m a d e
banks m o r e vulnerable to the flattening yield curve
that has occurred in 1993 and to interest rate increases
that m a y c o m e at s o m e time in the f u t u r e . Smaller
gaps between short- and longer-term interest rates limit banks' ability to use the yield curve to their advantage, and overall rate increases m a y threaten banks
with losses as shorter-lived liabilities m a t u r e m o r e
quickly than longer-term assets.
In addition, if loan d e m a n d picks up, banks m a y
have to struggle to regain C D customers induced by
their 1992 pricing policies to move their funds to other
investment vehicles. Growth in alternatives to bank de-

posits has been significant for the past two years. Money market mutual f u n d s and bond and stock mutual
funds have grown swiftly. Larger banks have worked
hard to promote various types of mutual funds as a way
of maintaining customer relationships. These institutions have little experience, however, in regaining the
d e p o s i t s directed into these f u n d s . M o r e generally,
banks have minimal experience in recovering from an
economic environment like that of 1991 and 1992, and
it is difficult to assess the seriousness of problems that
they may face in the future in drawing customers back.
Whatever is involved, there is little doubt that it will require significant marketing expenses, which will cut into future earnings.
At least some of 1992's impressive bank earnings
gains can be attributed to an improving economy and
several years of cleansing banks of bad practices and
bad paper. These positive forces should benefit banks'
future ability to produce earnings for their shareholders
and safety for their depositors and insurer. Nonetheless,
some of the gains have been bought with policies that
risk future earnings if a flatter yield curve and rising interest rates and loan demand return. If banks attempt to
reduce risks connected with rising rates and demand by
restructuring portfolios and regaining customers, future
earnings are also likely to be reduced.

Appendix
Profitability Measures
Three core measures have been used to provide information on bank performance: adjusted net-interest
margin, return on assets, and return on equity. Adjusted
net-interest margin gauges the d i f f e r e n c e between a
bank's interest income and expenses and is roughly similar to a b u s i n e s s ' s gross p r o f i t margin, except that
b a n k s ' sales of f e e - b a s e d services are not included.
Gross profit is the amount received from sales minus the
cost of goods or services sold; other expenses such as
sales, advertising, salaries, and rent have not been deducted. For banks, this indicator is calculated by subtracting interest e x p e n s e f r o m t a x - a d j u s t e d interest
revenue (net of loan-loss provisions) and dividing that
result by net interest-earning assets. For this calculation,
interest revenue from tax-exempt securities is adjusted
upward by the bank's marginal tax rate to avoid penalizing institutions that hold substantial state and local securities portfolios, which earn less interest but reduce tax
burdens.

Federal Reserve
 Bank of Atlanta


Loan-loss expenses are subtracted from interest revenue to place banks that make lower-risk loans at lower
interest rates on a more equal footing with commercial
banks that make higher-risk loans, which can generate
greater interest income. For example, interest rates on
credit cards have been substantially higher than rates on
prime commercial loans, but loan losses on credit cards
have also been larger.
Banks also bring in revenue in noninterest forms such
as loan origination fees; deposit service charges; and
charges for letters of credit, loan commitments, and other
off-balance-sheet services, to name a few. In recent years
gains from sales of securities have provided added income as well. Moreover, banks incur noninterest expenses such as expenditures on employee salaries, costs of
computer equipment, and maintenance. Therefore, Bank
X with a comparatively low adjusted-interest margin
may achieve a higher return on assets than Bank Y,
which attained a larger margin. That is, Bank X may
record a higher return on assets by realizing higher noninterest revenues or lower noninterest expenses.

Economic Review

43

Adjusted Net Interest Margin =

The return on assets ratio—the result of dividing a
bank's net income by its average assets—gauges how profitably a bank's management is using the firm's assets. The
return on equity figure tells a bank's shareholders how
much the institution is earning on the book value of their
investments. ROE is calculated by dividing a bank's net income by its total equity. The ratio of ROA to ROE falls as
the bank's capital-to-assets ratio rises. Smaller banks typically have higher capital-to-asset ratios.
Analysts who want to compare profitability while ignoring differences in equity capital ratios tend to focus on
ROA. Those wishing to focus on returns to shareholders
look at ROE. Highly capitalized banks that post the same
return on assets as less well capitalized competitors will
record a lower return on equity. Since return on equity is
computed by dividing a bank's net income by the book
value of its equity capital, a bank's return on equity will
decline as equity capital increases, if net income remains
fixed.

Adjusted Interest Revenues - Interest Expense
Average Interest-Earning Assets
Return on Assets =
Net Income
Average Consolidated Assets
Return on Equity =
Net Income
Average Equity Capital

Average interest-earning assets, consolidated assets,
and equity capital are derived by averaging beginning-,
middle-, and end-of-year balance-sheet figures. The expected interest-income component of net interest margin
incorporates two significant adjustments from ordinary
interest income. If profits before tax are greater than zero, the lesser of revenue from state and local securities
exempt from federal tax or the bank's profits before tax
is divided by 1 minus the bank's marginal federal tax
rate. Loan-loss expenses are subtracted from interest
revenue.

Profitability Data and Calculations
The bank data in this article are taken from reports of
condition and income filed with federal bank regulators
by insured commercial banks. The sample consists of all
banks that had the same identification number at the beginning and end of each year. The number of banks in the
1992 sample is 11,368.
The three profitability measures used in this study are
defined as follows:

Table 1
Adjusted N e t Interest Margin as a P e r c e n t a g e of Interest-Earning Assets
(Insured commercial

banks by consolidated

assets)

Year

All
Banks

$0-$25
million

$25-$50
million

$50-$ 100
million

$100-$ 500
million

$500 millionS i billion

$1 bill ion-t-

1988

3.65

4.04

4.15

4.25

4.22

3.71

3.43

1989

3.09

4.22

4.29

4.35

4.31

4.04

2.56

1990

3.03

4.27

4.26

4.25

4.10

3.87

2.55

1991

3.11

4.31

4.29

4.25

4.08

3.56

2.68

1992

3.76

4.66

4.72

4.67

4.46

4.24

3.44

S o u r c e : Figures in all tables h a v e been c o m p u t e d by the Federal Reserve Bank of Atlanta from data in " C o n s o l i d a t e d Reports of C o n d i t i o n
for Insured C o m m e r c i a l B a n k s " a n d " C o n s o l i d a t e d Reports of I n c o m e for Insured C o m m e r c i a l B a n k s , " 1988-92, filed w i t h e a c h
bank's respective regulator.

44



Economic Review-

September/October 1993

Table 2
Interest Expense as a Percentage of Interest-Earning Assets
(Insured commercial

banks by consolidated

assets)

Year

All
Banks

$0-$25
million

$25-$50
million

$50-$ 100
million

$1 00-$500
million

$500 millionS i billion

$1 billion+

1988

6.27

5.36

5.39

5.42

5.48

5.67

6.63

1989

7.39

5.91

6.01

6.04

6.19

6.42

7.93

1990

7.09

5.85

5.96

5.96

6.03

6.19

7.55

1991

5.72

5.23

5.30

5.29

5.28

5.18

5.92

1992

4.13

3.91

3.81

3.81

3.73

3.53

4.30

Table 3
Tax-Equivalent Interest R e v e n u e as a P e r c e n t a g e of Interest-Earning Assets
(Insured commercial
$25-$50
million

banks by consolidated
$50-$ 100
million

assets)

$100-$500
million

$500 millionS i billion

Year

All
Banks

$0-$25
million

1988

10.57

10.12

10.18

10.24

10.29

10.18

10.72

1989

11.57

10.71

10.86

10.89

11.08

11.16

11.82

1990

11.22

10.61

10.72

10.72

10.78

11.05

11.40

1991

10.00

9.97

10.06

10.04

10.01

9.84

10.00

1992

8.77

8.95

8.86

8.86

8.71

8.55

8.78

$1 billion+

Table 4
Loan-Loss Expense as a P e r c e n t a g e of Interest-Earning Assets
(Insured commercial

banks by consolidated

assets)

Year

All
Banks

$0-$25
million

$25-$50
million

$50-$100
million

$100-$500
million

$500 million$1 billion

$1 billion+

1988

0.65

0.72

0.63

0.57

0.59

0.79

0.66

1989

1.10

0.58

0.56

0.50

0.58

0.69

1.33

1990

1.10

0.49

0.50

0.51

0.65

0.98

1.30

1991

1.17

0.42

0.47

0.50

0.65

1.10

1.40

1992

0.88

0.38

0.33

0.39

0.53

0.77

1.03

F e d e r a l R e s e r v e B a n k of A t l a n t a




Economic

Review

45

Table 5
D e p o s i t Classes as a P e r c e n t a g e of Total D o m e s t i c
(Insured

commercial

Deposits

banks)

Year

Transactions
Accounts

MMDAs

Other
Savings

Time Deposits
less than $ 100,000

Time Deposits
more than $ 100,000

1988

32.1

17.7

9.3

24.8

16.1

1989

30.6

16.2

8.8

27.6

16.7

1990

29.5

16.3

8.6

29.7

15.8

1991

29.4

17.1

9.5

30.7

13.3

1992

31.6

18.7

11.3

28.3

10.2

Table 6
Securities G a i n s (Losses) b e f o r e Taxes as a P e r c e n t a g e of Total Assets*
(Insured

commercial

banks

by consolidated

assets)

Year

All
Banks

$0-$25
million

$25-$50
million

$50-$ 100
million

$100-$500
million

$500 million$1 billion

$1 billion+

1988

0.01

0.00

0.00

0.01

0.01

0.00

0.01

1989

0.02

0.00

0.01

0.01

0.01

0.00

0.03

1990

0.01

0.00

0.00

0.00

0.00

0.01

0.02

1991

0.09

0.05

0.05

0.06

0.06

0.07

0.10

1992

0.12

0.11

0.08

0.08

0.09

0.08

0.13

" 0.00 indicates

securities

gains (losses) that are less than 0.01 percent

of total assets.

Table 7
Noninterest I n c o m e as a P e r c e n t a g e of Total Assets
(Insured

commercial

banks

by consolidated

assets)

Year

All
Banks

$0-$25
million

$25-$50
million

$50-$ 100
million

$100-$500
million

$500 million$1 billion

$1 billion+

1988

1.44

0.91

0.75

0.82

0.89

1.12

1.68

1989

1.52

1.08

0.78

0.86

0.97

1.11

1.76

1990

1.63

1.07

0.82

0.83

0.93

1.26

1.91

1991

1.73

1.05

0.84

0.88

1.05

1.24

2.02

1992

1.88

1.24

0.90

0.90

1.14

1.32

2.20

4
Economic
 6


Review-

S e p t e m b e r / O c t o b e r 1993

Table 8
Total N o n i n t e r e s t Expense as a P e r c e n t a g e of Total Assets
(Insured commercial

banks by consolidated

assets)

Year

All
Banks

$0-$25
million

$25-$50
million

$50-$ 100
million

$100-$500
million

$500 millionS i billion

$1 billion+

1988

3.37

3.84

3.39

3.31

3.36

3.49

3.35

1989

3.39

3.86

3.41

3.31

3.40

3.34

3.39

1990

3.49

3.92

3.45

3.32

3.33

3.54

3.53

1991

3.73

3.96

3.56

3.40

3.49

3.59

3.82

1992

3.90

4.06

3.60

3.44

3.60

3.75

4.02

Table 9
Percentage Return on Assets
(Insured commercial

banks by consolidated

assets)

Year

All
Banks

$0-$25
million

$25-$ 50
million

$50-$ 100
million

$100-$ 5 00
million

$500 millionS i billion

$1 billion+

1988

0.84

0.36

0.61

0.77

0.81

0.58

0.89

1989

0.50

0.60

0.73

0.88

0.92

0.88

0.35

1990

0.50

0.60

0.71

0.81

0.80

0.77

0.39

1991

0.54

0.62

0.72

0.83

0.83

0.53

0.44

1992

0.96

0.95

1.05

1.10

1.07

0.95

0.93

$100-$500
million

$500 millionS i billion

$1 billion+

Table 10
P e r c e n t a g e Return o n Equity
(Insured commercial

banks by consolidated

- $50-$ 100
million

assets)

Year

All
Banks

$0-$25
million

$25-$50
million

1988

13.50

3.79

6.96

9.15

10.67

8.67

16.40

1989

7.93

6.15

8.15

10.12

11.93

12.72

6.21

1990

7.85

6.02

7.81

9.29

10.21

10.37

6.86

1991

8.04

6.26

7.85

9.40

10.50

7.43

7.35

1992

13.33

9.49

11.12

12.13

12.79

12.59

13.90

F e d e r a l R e s e r v e B a n k of A t l a n t a




Economic

Review

47

Table 11
Percentage Return on Assets
2 5 t h Percentile A c c o r d i n g to Profitability
(Insured commercial

banks by consolidated

assets)

Year

All
Banks

$0-$25
million

$25-$50
million

$50-$ 100
million

$100-$500
million

$500 millionS i billion

$1 billion+

1988

0.51

0.20

0.53

0.64

0.71

0.56

0.71

1989

0.58

0.37

0.58

0.70

0.77

0.64

0.50

1990

0.52

0.36

0.53

0.63

0.65

0.41

0.10

1991

0.56

0.45

0.55

0.67

0.64

0.51

0.21

1992

0.79

0.68

0.81

0.86

0.85

0.74

0.62

Table 1 2
Percentage Return on Assets
5 0 t h Percentile A c c o r d i n g to Profitability
(Insured commercial

banks by consolidated

assets)

Year

All
Banks

$0-$25
million

$25-$50
million

$50-$ 100
million

$100-$ 500
million

$500 millionS i billion

$1 bill ion-f-

1988

0.93

0.78

0.93

0.98

1.04

1.00

1.01

1989

0.98

0.84

0.98

1.04

1.07

1.06

0.96

1990

0.93

0.82

0.93

0.98

1.01

0.98

0.74

1991

0.95

0.86

0.94

1.00

1.01

0.94

0.81

1992

1.13

1.03

1.14

1.18

1.18

1.10

1.02

Table 13
Percentage Return on Assets
7 5 t h Percentile A c c o r d i n g to Profitability
(Insured commercial

banks by consolidated

assets)

Year

All
Banks

$0-$25
million

$25-$50
million

$50-$ 100
million

$100-$ 500
million

$500 million$1 billion

$1 billion+

1988

1.24

1.14

1.24

1.28

1.33

1.29

1.21

1989

1.29

1.20

1.28

1.34

1.36

1.30

1.20

1990

1.23

1.16

1.23

1.26

1.28

1.30

1.12

1991

1.24

1.18

1.24

1.27

1.28

1.25

1.16

1992

1.44

1.34

1.44

1.48

1.46

1.37

1.33

 4 8


Economic

Review-

September/October

1993

Table 14
Adjusted N e t Interest Margin as a Percentage of Interest-Earning Assets
(Insured commercial

banks in the Southeast by consolidated

assets)

Year

All SE
Banks

$0-$25
million

$25-$50
million

$50-$ 100
million

$100-$500
million

$500 millionS i billion

$1 billion+

1988

4.19

4.30

4.27

4.35

4.37

4.04

4.12

1989

3.80

4.17

4.35

4.29

4.26

3.50

3.56

1990

3.51

4.15

4.33-

4.18

4.11

4.00

3.08

1991

3.71

4.04

4.18

4.18

4.14

3.81

3.44

1992

4.38

4.63

4.75

4.71

4.51

4.43

4.25

Table 15
Interest Expense as a Percentage of Interest-Earning Assets
(Insured commercial banks in the Southeast by consolidated assets)

Year

All SE
Banks

$0-$25
million

$25-$50
million

$50-$ 100
million

$100-$500
million

$500 millionS i billion

$1 billion+

1988

5.66

5.53

5.59

5.60

5.45

5.76

5.73

1989

6.48

6.23

6.34

6.32

6.19

6.53

6.62

1990

6.28

6.08

6.20

6.17

6.07

6.34

6.36

1991

5.23

5.45

5.52

5.42

5.27

5.22

5.16

1992

3.53

3.96

3.89

3.81

3.62

3.41

3.43

Table 16
Tax-Equivalent Interest R e v e n u e as a P e r c e n t a g e of Interest-Earning Assets
(Insured commercial

banks in the Southeast by consolidated

assets)

Year

All SE
Banks

$0-$25
million

1988

10.49

10.54

10.55

10.52

10.43

10.37

10.51

1989

11.07

11.24

11.31

11.14

11.05

10.99

11.05

1990

10.85

11.00

11.09

10.98

10.82

11.39

10.75

1991

9.84

10.16

10.33

10.25

10.04

9.78

9.67

1992

8.50

9.22

9.09

8.99

8.63

8.39

8.34

F e d e r a l R e s e r v e B a n k of A t l a n t a




$25-$50
million

$50-$ 100
million

$100-$500
million

$500 millionS i billion

Economic

Review

$1 billion+

49

Table 17
Loan-Loss Expense a s a P e r c e n t a g e of Interest-Earning Assets
(Insured

commercial

banks

in the Southeast

by consolidated

assets)

Year

All SE
Banks

$0-$25
million

$25-$50
million

$50-$ 100
million

$100-$ 5 00
million

$500 millionS i billion

$1 billion+

1988

0.64

0.71

0.69

0.58

0.61

0.56

0.66

1989

0.79

0.84

0.62

0.53

0.60

0.95

0.88

1990

1.06

0.76

0.56

0.62

0.64

1.05

1.30

1991

0.90

0.67

0.63

0.65

0.63

0.76

1.07

1992

0.59

0.63

0.44

0.47

0.50

0.55

0.65

Table 18
Securities G a i n s (Losses) b e f o r e T a x e s as a P e r c e n t a g e of Total Assets'
(Insured

commercial

banks

in the Southeast

by consolidated

assets)

Year

All SE
Banks

$0-$25
million

$25-$50
million

$50-$100
million

$100-$500
million

$500 million$1 billion

$1 billion+

1988

0.02

(0.01)

0.00

0.01

0.01

0.02

0.02

1989

0.02

0.00

0.01

0.01

0.00

0.00

0.04

1990

0.02

0.00

0.00

(0.01)

(0.01)

0.01

0.03

1991

0.11

0.08

0.07

0.05

0.06

0.04

0.14

1992

0.09

0.09

0.10

0.09

0.08

0.03

0.09

* 0.00 indicates

securities

gains (losses) that are less than 0.01 percent

of total assets.

Table 1 9
Noninterest I n c o m e as a Percentage of Total Assets
(Insured

commercial

banks

in the Southeast

by consolidated

assets)

Year

All SE
Banks

$0-$25
million

$25-$50
million

$50-$ 100
million

$100-$ 5 00
million

$500 million$1 billion

$1 billion+

1988

1.22

1.40

0.86

1.09

1.06

1.25

1.31

1989

1.17

1.50

0.85

1.05

1.06

1.35

1.23

1990

1.26

1.12

0.91

1.06

1.08

1.12

1.39

1991

1.35

1.67

0.90

1.15

1.17

1.19

1.48

1992

1.42

1.83

0.95

1.00

1.15

1.21

1.62

5
 0


Economic

Review-

S e p t e m b e r / O c t o b e r 1993

Table 2 0
Total N o n i n t e r e s t Expense as a P e r c e n t a g e of Total Assets
(Insured commercial

banks in the Southeast by consolidated

assets)

Year

All SE
Banks

$0-$25
million

$25-$50
million

$50-$ 100
million

$100-$500
million

$500 millionS i billion

$1 billion+

1988

3.61

4.59

3.71

3.59

3.59

3.56

3.58

1989

3.48

4.72

3.64

3.46

3.51

3.62

3.41

1990

3.53

4.54

3.69

3.58

3.45

3.71

3.49

1991

3.72

4.97

3.75

3.72

3.58

3.60

3.74

1992

3.82

4.82

3.80

3.63

3.57

3.71

3.92

Table 21
Percentage Return o n Assets
(Insured commercial

banks in the Southeast by consolidated

assets)

Year

All SE
Banks

$0-$25
million

$25-$50
million

$50-$100
million

$100-$500
million

$500 millionS i billion

$1 billion+

1988

0.82

0.30

0.51

0.81

0.81

0.86

0.87

1989

0.68

0.20

0.64

0.89

0.87

0.55

0.62

1990

0.53

0.03

0.64

0.71

0.83

0.65

0.41

1991

0.66

0.14

0.58

0.75

0.88

0.67

0.60

1992

1.06

0.77

1.01

1.08

1.15

0.97

1.05

Table 22
P e r c e n t a g e Return o n Equity
(Insured commercial

banks in the Southeast by consolidated

assets)

Year

All SE
Banks

$0-$25
million

$25-$50
million

$50-$100
million

$100-$500
million

1988

11.64

2.80

5.48

9.41

10.56

12.85

13.69

1989

9.56

1.79

6.71

9.98

11.06

8.29

9.79

1990

7.39

0.25

6.77

8.00

10.53

7.65

6.47

1991

8.96

1.26

6.07

8.37

11.09

9.70

8.79

1992

13.81

7.13

10.31

11.97

13.92

13.10

14.74

Digitized e r a lFRASERe B a n k of A t l a n t a
F e d for R e s e r v


$500 million$1 billion

Economic

$1 billion+

Review

51

Table 2 3
Adjusted N e t Interest Margin as a P e r c e n t a g e of Interest-Earning Assets
(Insured commercial

banks in the Southeast by state)

Year

All SE
Banks

Alabama

Florida

Georgia

Louisiana

Mississippi

Tennessee

1988

4.19

4.29

4.20

4.82

3.35

4.07

4.00

1989

3.80

3.98

3.73

4.57

2.81

3.85

3.55

1990

3.51

3.97

3.17

4.21

3.07

3.75

3.30

1991

3.71

4.08

3.44

4.09

3.18

4.10

3.82

1992

4.38

4.49

4.37

4.41

4.47

4.45

4.19

Table 2 4
Interest Expense as a Percentage of Interest-Earning Assets
(Insured commercial

banks in the Southeast by state)

Year

All SE
Banks

Alabama

Florida

Georgia

Louisiana

Mississippi

Tennessee

1988

5.66

5.82

5.45

5.75

5.91

5.67

5.77

1989

6.48

6.62

6.35

6.61

6.42

6.44

6.63

1990

6.28

6.25

6.27

6.16

6.24

6.21

6.57

1991

5.23

5.29

5.15

5.34

5.14

5.28

5.28

1992

3.53

3.66

3.42

3.69

3.26

3.66

3.59

Table 25
Tax-Equivalent Interest R e v e n u e as a P e r c e n t a g e of Interest-Earning Assets
(Insured commercial

banks in the Southeast by state)

Year

Alabama

Florida

Georgia

Louisiana

Mississippi

Tennessee

1988

10.49

10.42

10.23

11.11

10.55

10.21

10.51

1989

11.07

11.02

10.86

11.76

10.71

10.80

11.13

1990

10.85

10.70

10.64

11.35

10.53

10.57

11.21

1991

9.84

9.92

9.62

10.39

9.43

9.87

9.89

1992

52

All SE
Banks

8.50

8.65

8.38

8.85

8.25

8.60

8.34




Economic

Review-

September/October

1993

Table 2 6
Loan-Loss Expense as a P e r c e n t a g e of Interest-Earning Assets
(Insured commercial

banks in the Southeast by state)

Year

All SE
Banks

Alabama

Florida

Georgia

Louisiana

Mississippi

Tennessee

1988

0.64

0.32

0.59

0.54

1.30

0.46

0.74

1989

0.79

0.42

0.78

0.58

1.48

0.51

0.95

1990

1.06

0.47

1.21

0.99

1.22

0.61

1.34

1991

0.90

0.55

1.03

0.96

1.11

0.49

0.78

1992

0.59

0.50

0.58

0.75

0.51

0.48

0.56

Table 2 7
Percentage Return o n Assets
(Insured commercial

banks in the Southeast by state)

Year

All SE
Banks

Alabama

Florida

Georgia

Louisiana

Mississippi

Tennessee

1988

0.82

1.16

0.78

1.15

0.03

0.85

0.84

1989

0.68

1.01

0.61

1.10

-0.12

0.79

0.61

1990

0.53

1.03

0.29

0.93

0.21

0.73

0.43

1991

0.66

1.02

0.48

0.87

0.22

0.91

0.77

1992

1.06

1.24

0.87

1.27

1.13

1.11

1.04

Table 2 8
P e r c e n t a g e Return o n Equity
(Insured commercial

banks in the Southeast by state)

Year

All SE
Banks

Alabama

Florida

Georgia

1988

11.64

14.39

12.20

15.76

1989

9.56

12.53

9.54

1990

7.39

13.00

1991

8.96

1992

13.81

Mississippi

Tennessee

0.41

10.91

11.54

14.41

-1.69

9.96

8.29

4.33

11.37

3.15

9.43

5.90

13.29

7.12

9.98

3.35

11.76

10.63

15.83

12.28

14.17

15.76

13.77

13.93

Federal Reserve B a n k of Atlanta




Louisiana

Economic

Review

53

Notes
1. In this study Southeast refers to the six states entirely or partially within the Sixth Federal Reserve District: Alabama,
Florida, Georgia, Louisiana, Mississippi, and Tennessee.
2. The revenue, expense, and profitability figures presented are
generally similar to those displayed in prior bank profitability
studies published in the Economic Review (see Goudreau
1992 for the most recent study). The figures may not be identical because the data have been corrected for reporting errors. Additionally, the interest revenue as a percentage of
interest-earning assets ratio and adjusted net interest margins
may differ from figures reported in previous studies because
of corrections in the treatment of tax-exempt interest income.
3. Six size categories of commercial banks are analyzed in this
study. They are (1) banks with total assets of no more than
$25 million, (2) banks with total assets exceeding $25 million and at most $50 million, (3) banks with total assets
greater than $50 million and no more than $100 million, and
(4) banks with total assets exceeding $100 million, up to
$500 million, (5) banks with total assets exceeding $500 million and at most $1 billion, and (6) banks with total assets
greater than $1 billion.
Only banks that have been open to the public for at least
one full year are included in this study. The ratios displayed
are full-year probability figures based on the mean beginning-,
middle-, and end-of-year balance sheets and income statements. Banks that commenced operations during any particular year will be missing beginning-of-year data and perhaps
more. See Table A for more information about banks included
in this study.
4. A loan-loss provision is a noncash expense item charged to a
bank's earnings when expanding the allowance for possible
bad debt. These provisions are reported on a bank's income
statement. A bank does not set aside funds (cash) in reserve
to cover its loan losses, and an increase in the loan-loss account does not directly cause any change in the allocation of
a bank's assets.
An increase in loan-loss provisions reduces the net value
of the bank's loans on its accounting records and its net income. Increases in provisions will also have a negative impact
on a bank's equity capital by reducing the bank's retained

earnings and may trigger regulatory demands for additional
equity. See Wall (1988, 39-41).
5. Noninterest income is net income derived from fee-based
banking services, such as corporate cash management, check
collection, and consumer annual fees on credit cards, as well
as monthly service charges on deposit accounts. Also included are many new activities, such as fees from participation in
mutual fund commissions, investment advisor fees in merger
and acquisition activities, and securities underwriting fees.
Noninterest expenses are the operating costs of a bank.
They include salaries, equipment rental fees, leases of buildings and equipment, deposit insurance costs, and taxes and
other related expenses. "Consolidated Reports of Income for
Insured Commercial Banks" filed by banks with their primary regulators have three noninterest components. They are
(1) salaries and employee benefits, (2) expenses for premises
and fixed assets, and (3) other noninterest expenses. Salaries
and employee benefits now account for roughly 40 percent of
the total, expenses for premises and fixed assets absorb about
14 percent, and other noninterest costs equal less than onehalf of total noninterest expenses.
6. For example, noninterest income in 1992 for the nation's
largest banks was 24 percent of total revenue. Respective
percentages for the five smaller asset classes, starting with
under-$25 million asset banks, were 15 percent, 12 percent,
11 percent, 14 percent, and 16 percent during the same year.
7. Deposit insurance premiums for commercial banks in 1990
equaled 12 cents for every $100 of deposits. The average assessment for commercial banks for 1991 and 1992 increased
to 21.25 and 23 cents, respectively, per SI00 dollars of deposits.
8. Equity-to-assets ratios for banks in the six asset classifications during the 1988-92 period are displayed in Table B.
Equity-to-assets ratios, with few exceptions, have risen
steadily for each size group during the five years under review. Larger banks maintain equity-to-assets ratios that are
considerably lower than those of smaller competitors.
9. See Table C for a description of characteristics of southeastern banks included in this survey.

Table A
U.S. C o m m e r c i a l Banks, 1 9 9 2
$500 millionSi billion

$0-$25
million

$25-$50
million

$50-$100
million

$100-$500
million

2,513

2,935

2,775

2,520

250

375

Percent of U.S. Banks

22.1

25.8

24.4

22.2

2.2

3.3

Total Assets ($ billions)

40.8

103.4

190.4

480.7

169.6

2,359.4

1.2

3.1

5.7

14.4

5.1

70.5

Number of Banks

Percent of U.S. Total Assets

54



Economic Review-

$1 billion+

September/October 1993

Table B
Equity-to-Total Assets Ratios
U.S. C o m m e r c i a l Banks
(Percent)

Year

All
Banks

$0-$25
million

$25-$50
million

$50-$ 100
million

$100-$ 500
million

$500 millionSi billion

$1 billion+

1988

6.19

9.41

8.82

8.44

7.58

6.72

5.43

1989

6.36

9.71

8.97

8.66

7.72

6.94

5.65

1990

6.38

9.93

9.06

8.73

7.83

7.39

5.64

1991

6.67

9.86

9.19

8.83

7.95

7.16

6.03

1992

7.21

10.01

9.43

9.04

8.33

7.56

6.67

Table C
Southeastern C o m m e r c i a l Banks, 1 9 9 2
$0-$25
million

$25-$50
million

$50-$ 100
million

$100-$500
million

201

474

459

376

39

51

12.6

29.6

28.7

23.5

2.4

3.2

Total Assets ($ billions)

3.4

16.8

31.2

70.4

26.1

213.7

Percent of S.E. Total Assets

1.0

4.6

8.6

19.5

7.2

59.1

Number of Banks
Percent of S.E. Banks

$500 millionSi billion

$1 billion+

References
Brunner, Allan D., and William B. English. "Profits and Balance Sheet Developments at U.S. Commercial Banks in
1992." Federal Reserve Bulletin 79 (July 1993): 649-73.
Goudreau, Robert E. "Commercial Bank Profitability Rises as
Interest Margins and Securities Sales Increase." Federal Reserve Bank of Atlanta Economic Review 77 (July/August
1992): 33-52.
Goudreau, Robert E., and B. Frank King. "Commercial Bank
Profitability: Hampered Again by Large Banks' Loan Prob-

Federal Reserve Bank of Atlanta



lems." Federal Reserve Bank of Atlanta Economic Review
76 (July/August 1991): 39-54.
Kottman, Stacy. "Louisiana Runs against the Current and Will
Underperform the Region in Recovery." Regional Update 5
(October/December 1992): 17-21.
Wall, Larry D. "Commercial Bank Profits: Still Weak in 1987."
Federal Reserve Bank of Atlanta Economic Review 73 (July/August 1988): 28-42.

Economic

Review

55

The Federal Reserve Bank of Atlanta
presents

Banking in a High-Tech World
Managing in a Changing Environment




A conference held at the Swissotel
near Lenox Square
Atlanta, Georgia
April 2 8 - 2 9 , 1994

Topics i n c l u d e
• Mutual F u n d s
• I n t e r e s t R a t e Risk
• P r o p o s e d Regulatory Changes

SlVlarh

Your

For more information call:
Jess Palazzolo, Conference Coordinator
(404) 521-8747

Calendars

New insight into

derivatives

How can you most effectively manage interest rate risk?
What are the risks and returns of mortgage-backed securities?
You can find the answers to these questions
and many more in Financial Derivatives,
a new book that provides insight into the
growing arena of futures, options and swaps.

Financial Derivatives contains 16 indepth articles, written by economists at
the Federal Reserve Bank of Atlanta.

An essential tool for:
• Commercial Bankers
• Investment Analysts
• Portfolio Managers
• Financial Analysts
• Business Professors

Financial
Derivatives Their Uses
New Instruments and

This valuable 242-page reference is only $15, including shipping and handling.
To order, complete the form below. Enclose a check or money order payable to the Federal Reserve Bank of Atlanta
and mail to or include your Visa or Mastercard number and mail or fax to:
Federal Reserve Bank of Atlanta
Public Affairs Department
104 Marietta Street, N.W.
Atlanta, GA 30303-2713

Fax: (404) 521-8050
For more information, call:
(404) 521-8020

O R D E R F O R M : Mail or fax to Federal Reserve Bank of Atlanta • Public Affairs Department • 104 Marietta Street, N.W.
Atlanta, GA 30303-2713 • Fax: (404) 521-8050
I would like to order

(no. of copies) X $15 for a total of $_

Method of payment (check one)
•

1 have enclosed a check or money order

•

Please charge m y credit card

Name
Company
Address

Cardholder Name:

City, State, Zip

Visa/MC #

Country

Expiration Date:

Phone #




Federal Reserve Bank of Atlanta Working Papers

T

he Research D e p a r t m e n t of the Federal R e s e r v e B a n k of Atlanta p u b l i s h e s a w o r k i n g p a p e r series to c o n v e y the research of
staff e c o n o m i s t s and visiting scholars a n d stimulate p r o f e s s i o n a l d i s c u s s i o n a n d exploration of e c o n o m i c a n d financial subjects. T o s u b s c r i b e to the w o r k i n g p a p e r series or to receive single copies of recently p u b l i s h e d p a p e r s (sent automatically to current subscribers), please check the appropriate b o x ( e s ) b e l o w and return the f o r m as indicated.
93-9

93-1

Has the Romers' Narrative Approach
Monetary Policy Shocks?
Eric M. Leeper

93-2

Risk Taking and Failure in De Novo Savings and
Loans in the 1980s
William C. Hunter, James A. Verbrugge, and
David A. Whidbee

93-3

Core Deposits and Physical Capital:
A Reexamination of Bank Scale Economies and
Efficiency with Quasi-Fixed
Inputs
William C. Hunter and Stephen G. Timme

93-4

93-5

93-6

93-7

93-8

•
•

Identified

93-11 Educational Achievement and Economic
Evidence from Taiwan
Ellis W. Tallman and Ping Wang
93-12 Monetary Policy, Interest Rates, and
Budget Arithmetic
Revisited
Marco Espinosa and Steven Russell
93-13 Financial Integration
International Policy
Roberto Chang

The Dynamic Impacts of Monetary
Policy:
An Exercise in Tentative
Identification
David B. Gordon and Eric M. Leeper

of

93-16 Excess Return, Excess Volatility, and Negative
Autocorrelation Caused by Uncertainty
Aversion
and Risk Aversion
Jie Hu

Debt

Please start m y subscription to the W o r k i n g P a p e r Series.
I currently receive other Federal R e s e r v e B a n k of Atlanta publications. (To e x p e d i t e y o u r order, please p r o v i d e the subscriber n u m b e r that a p p e a r s a b o v e y o u r m a i l i n g address: R

•

Inflation:

93-15 Vague Preferences, Noisy Markets, and
Other Parables Concerning the Informational
Role of Prices
Vicente Madrigal and Stephen D. Smith

Tests of

and Sovereign

Growth:

with and without
Coordination

93-14 Towards the Systematic Measurement
Systemic Risk
Hugh Cohen and William Roberds

Managerial Rents and Optimal
Regulatory
Intervention in Troubled Banks
Thomas H. Noe, Michael J. Rebello,
and Larry D. Wall

Private Investment
Negotiations
Roberto Chang

Adjustment
Information

93-10 Liquidity Shocks and Financial Crises
During the National Banking Era
Ellis W. Tallman and Jon R. Moen

Some Implications of Risk Neutrality for
Time Variation in Stock Returns
Ronald W. Best and Stephen D. Smith

Generalized Method of Moments
Forward Rate Processes
Peter A. Abken

Rational Expectations and the Dynamic
of Security Analysts' Forecasts to New
Lucy F. Ackert and William C. Hunter

).

Please send m e a c o p y of the f o l l o w i n g w o r k i n g papers:
• 93-1
• 93-9

(3 9 3 - 2
• 93-10

• 93-3
•93-11

• 93-4
• 93-12

• 93-5
• 93-13

• 93-6
• 93-14

• 93-7
• 93-15

•
•

93-8
93-16

Name
Address
City

State

ZIP

R e t u r n this f o r m to the Public A f f a i r s D e p a r t m e n t , Federal R e s e r v e B a n k of Atlanta, 104 Marietta Street, N . W . , Atlanta,
G e o r g i a 3 0 3 0 3 - 2 7 1 3 . S o m e previously published w o r k i n g p a p e r s are also available. F o r m o r e i n f o r m a t i o n write to this address or call 4 0 4 / 5 2 1 - 8 0 2 0 .






Bulk Rate
U.S. Postage

PAID
Atlanta, GA
Permit 292

Public Affairs Department
104 M a r i e t t a S t r e e t ,

N.W.

Atlanta, Georgia 30303-2713
( 4 0 4 )




5 2 1 - 8 0 2 0

printed co recycled paper


Federal Reserve Bank of St. Louis, One Federal Reserve Bank Plaza, St. Louis, MO 63102