View original document

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



September 21, 1990

Costly Information and the eRA
The Community Reinvestment Act (CRA),
passed in 1977, requires banks and other depository institutions to make an effort to meet
the credit needs of all the communities in which
they are located, including the low- and moderate-income neighborhoods. In recent years,
banks' community reinvestment activities have
come under increasing scrutiny, and CRA enforcement has become more stringent. In this Letter,
we investigate some of the factors that led to
the passage of the law, and the impact of stricter
enforcement standards on commercial banks'
mortgage lending and profitability.

little lending experience in the neighborhood is
likely to have very limited information about the
credit-worthiness of the residents, and the cost of
evaluating applicants from this area may turn out
to be higher than in other neighborhoods where
the bank has more lending experience.

Information, risk, and lending

When the costs of finding profitable opportunities differ sharply across neighborhoods,
banks may find it difficult to make profitable
loans in certain neighborhoods, even though
some good projects may exist in these neighborhoods. Thus, it is possible that bank lending
patterns will differ across neighborhoods even
if other conditions are similar.

Whenever a bank makes a loan, it faces the
risk that the loan may not be fully repaid. Thus,
its expected return is lower than the contractual
rate on the loan. To account for this risk, a bank
might choose to vary its loan rates according to
the credit-worthiness and collateral of the individual borrower, with higher rates for the borrowers that appear to pose the greatest risk. In
the case of mortgage loans, however, banks tend
not to vary their loan rates significantly across
borrowers. Instead, they attempt to offer loans
to only the most credit-worthy borrowers.

Inasmuch as household income prospects and
trends in real estate values may be loosely associated with the ethnic characteristics of neighborhoods, bank lending patterns may appear to
be correlated with these ethnic characteristics.
Many have asserted that these correlations provide evidence that banks have engaged in discriminatory lending behavior. This perception led
to the enactment of laws such as the Equal Credit
Opportunity Act, the Home Mortgage Disclosure
Act, and the CRA.

Identifying credit-worthy borrowers requires the
bank to "invest" in information about potential
borrowers. (Typically, the kinds of information
banks seek include applicants' income and
credit background and the stability of applicants'
employment and residence history.) Presumably,
the more information the bank obtains, the more
effectively it will be able to gauge credit-worthiness. Thus, by investing in information that
enables it to screen applicants better, a lender
may be able to raise the expected return to
mortgage lending.
Obtaining this information and screening applicants is costly. Moreover, these costs may vary
across individuals and neighborhoods. One factor that may influence these costs is the bank's
presence in a given neighborhood. A bank with

eRA enforcement
Under the CRA, banks are required to solicit
loans in low-income areas. They are not required
to lend to bad credit risks, but they must make
an effort to find good lending opportunities in
those areas. As part of their compliance with
CRA-related regulations, banks must have a community reinvestment plan, and must demonstrate
a concerted effort to market bank loan and deposit products to all neighborhoods in their
defined market area.
One important tool available to bank regulators
in enforcing CRA is the power to deny banks'
proposed mergers and acquisitions. Although
this power has not been used widely in the past,
there is a clear trend toward greater attention to
CRA compliance in evaluating such applications.

Regulators sometimes have made approval of the
application contingent on the bank's willingness
to revise its policies to meet CRA guidelines.
In addition, community groups and neighborhood organizations are permitted to file formal
protests against merger and acquisition applications on the ground that applicant banks are
not meeting their obligations under the CRA. In
many instances, these groups have succeeded in
delaying proposed mergers and, in some cases,
obtaining promises of increased lending or other
concessions by banks. Finally, because CRA
compliance ratings assigned by the regulatory
agencies after July 1, 1990, will be disclosed to
the public, heightened public scrutiny will be
brought to bear on depository institutions.

Effects on lending patterns
To determine how CRA affects banks' behavior,
we developed a model of bank lending that takes
into account the possibility that perceived credit
risks may vary across neighborhoods. These perceptions about risk are formed, in part, on the
basis of the bank's information about borrowers
in the area. As the bank increases its presence
in the area, its costs of screening applicants may
fall, making lending in the area more attractive.
In this framework, the bank balances the expected gain from information (lowering default
risk through better screening) with the cost of
obtaining that information (devoting more resources to assessing loans in the neighborhood).
Our results suggest that information effects can
hinderlendingto low-income neighborhoods. If
past experience has not uncovered a large pool
of qualified borrowers in an area, a bank may
maintain only a minimal presence in that neighborhood. With few branches and/or loan officers
in the area, it may be hard to identify new, profitable lending opportunities. Expected returns to
lending (net of search costs) are thus likely to be
lower in those areas where the bank has little or
no presence. This, in turn, makes it less likely
that the bank will increase lending in these areas.
CRA, in this framework, can influence a bank's
lending to residents of low-income neighborhoods by encouraging the bank to search harder
for good borrowers. Because the benefits of additional search under CRA include avoiding
regulatory penalties, the bank may find it worth-

while to increase its presence in low-income
neighborhoods, and in the process, may uncover
more qualified borrowers. (For a full description
of this model, see the Summer 1990 issue of the
Federal Reserve Bank of San Francisco's
Economic Review.)

New institutions
Thus, our model suggests that as a bank obtains
additional information about a given neighborhood, it may find profitable lending opportunities
that can partially offset the costs of complying
with the CRA. Nonetheless, the cost of obtaining
information in neighborhoods targeted by the
CRA still may be relatively high, giving banks an
incentive to find ways to reduce these costs. One
way banks in a number of states have attempted
to reduce costs is to form a multi-bank consortium to find and make loans in disadvantaged
neighborhoods. These institutions, such as the
California Community Reinvestment Corporation,
provide a pool of loan funds from several banks.
The consortium serves the purposes of lowering
the per-bank costs of obtaining information about
low-income neighborhoods as well as spreading
loan risk among several lenders.
Another way banks have sought to reduce
information costs is to work directly with nonprofit community groups, allowing these groups
to perform the initial applicant screening for
CRA-related loans. This approach lowers the cost
of information for the lender by shifting part of
the search and monitoring costs to the community groups. It also increases community group
sensitivity to the credit risk problems faced by
banks when lending in certain areas.

Effort vs. results
Although the CRA is expressed in terms of effort,
recent events may suggest a shift in emphasis.
Community groups cite a lack of lending volume
in low-income neighborhoods as a reason to protest mergers and acquisitions, rather than evidence of insufficient efforts to find good loans.
This shift is understandable, given that it may
be easier to measure results than effort.
Critics of a results-oriented focus maintain,
however, that such an approach could lead to
higher risk. To the extent that credit-worthy borrowers cannot be found, they argue, a resultsoriented approach to CRA would force banks

to lower their credit standards, which, in turn,
could lead to higher default rates. Thus, one
implication of a results-oriented approach is
that bank income could be lower under CRA
than otherwise.
While credit-worthy borrowers exist in CRAtargeted neighborhoods, the relevant issue involves the cost of finding these borrowers. The
critics of a results-oriented approach argue that
since banks did not lend in these neighborhoods
previously, the expected return (not necessarily
the actual return) net of the cost of finding good
borrowers was too low to make these loans profitable. Community groups, on the other hand,
maintain that banks have significantly overestimated the cost of finding credit-worthy borrowers in low-income neighborhoods and that as
banks become more involved in these neighborhoods, as the CRA requires, they will find
profitable lending opportunities.

Social vs. private costs
One important conclusion drawn from our
theoretical model is that below-desired lending
to low-income neighborhoods may arise, in part,
because of a lack of information about opportunities. High costs of gathering the needed
information may keep banks from voluntarily
seeking credit-worthy borrowers in these areas.
CRA represents one mechanism to pressure
banks to address the problems created by costly
information. The law imposes a penalty on banks
if they fail to exert sufficient effort searching for
good loans in low-income areas. In essence, CRA
imposes a tax on banks (in the form of regulatory
penalties) to achieve the social goal of increased
lending to disadvantaged neighborhoods.
The incidence of this tax, however, is an important public policy issue. If there is evidence that
the apparent lack of lending in low-income areas
is due to banks' inaccurate appraisals of the costs
of finding credit-worthy borrowers, then it may be
appropriate to tax banks to achieve the desired
social goal. The recent results-oriented interpretation of CRA seems to be pred icated on the assumption that banks persistently are overestimating the
costs of lending in low-income areas. The evidence on this point is inconclusive, though. Redlining studies generally have failed to demonstrate
conclusively the existence of neighborhood lend-

ing differentials after controlling for relevant
demand and supply factors in housing markets.
To the extent that the cost of obtaining information about lending opportunities to low-income
neighborhoods is indeed high, banks may find
the costs of complying with CRA prohibitive, and
may be induced to close branches in CRA-sensitive neighborhoods in order to change their
defined market to exclude those areas. (Such
closings are difficult, however.) In those cases,
bank customers in those neighborhoods would
bear some of the cost of CRA through reduced
The magnitude of this potential cost is not yet
known. CRA proponents, both inside and outside
banks, argue that CRA loans have performed as
well as other loans. This suggests that banks may
have overestimated the risks of lending to lowincome borrowers. In those cases, the costs of
CRA compliance may be negligible and the bank
may profit from the activity.
However, it is still not clear how costly it will
be to find profitable opportunities as the scale
of CRA lending increases. If these costs rise, the
adverse effect on bank competitiveness will
become important. A policy that requires private
banks to bear the social cost of increasing lowincome lending could put them at a competitive
disadvantage with respect to other institutions
that do not face CRA responsibilities.

Sticks vs. carrots
Our research indicates that information costs
can make it more difficult for banks to lend to
borrowers in low-income neighborhoods. CRA
uses a "stick" to pressure banks to search for
additional qualified borrowers in these neighborhoods. In the process, banks bear the cost
of this social policy. If the costs of compliance
become large, it may be necessary to consider
alternative ways to surmount information problems without putting banks at a competitive
disadvantage. For example, banks could be
rewarded for good CRA performance (perhaps
through expanded powers). Alternatively, loan
guarantee programs could be used to encourage
a more even geographic distribution of lending.

Jonathan A. Neuberger

Ronald H. Schmidt
Senior Economist

Opinions expressed in this newsletter do not necessarily refiect the views of the management of the Federal Reserve Bank of
San Francisco, or of the Board of Governors of the Federal Reserve System.
Editorial comments may be addressed to the editor (Barbara Bennett) or to the author.... Free copies of Federal Reserve
publications can be obtained from the Public Information Department, Federal Reserve Bank of San Francisco, P.O. Box 7702,
San Francisco 94120. Phone (415) 974-2246.

Research Department

Federal Reserve
Bank of
San Francisco
P.O. Box 7702
San Francisco, CA 94120