FRBSF Economic Letter

2004-16 | June 25, 2004

Has the CRA Increased Lending for Low-Income Home Purchases?

Liz Laderman

When Congress enacted the Community Reinvestment Act (CRA) in 1977, its main goal was to address concerns that some banking institutions were not fully meeting the credit needs of qualified potential borrowers, particularly those in low- and moderate-income (LMI) and minority neighborhoods of inner cities. Since then, debate has continued over the need for, and the effectiveness of, the CRA.

This Economic Letter discusses some of the research on whether the CRA has helped alleviate any shortfall in LMI home purchase lending. Overall, the literature suggests that, while the effects of the CRA versus the effects of other developments are not easy to separate, access to credit for the groups and neighborhoods targeted by the CRA has improved; in addition, it suggests that, for most banks, LMI home purchase lending has become as profitable as other home purchase lending.

What were the CRA’s goals for LMI lending?

The CRA states that each federal bank or thrift supervisory agency must “use its authority when examining financial institutions to encourage such institutions to help meet the credit needs of the communities in which they are chartered….” The CRA applies to banks and thrifts and their finance and mortgage company affiliates; it does not apply to credit unions and independent finance and mortgage companies. Under the CRA, if a bank’s record of meeting the credit needs of its entire community, including LMI neighborhoods, is unsatisfactory, the bank may face supervisory penalties.

The CRA’s goal, in effect, was to encourage lending to qualified borrowers in LMI neighborhoods to mitigate a perceived shortfall of such lending relative to lending to qualified borrowers in other neighborhoods. In LMI neighborhoods, median family income is less than 80% of that for a broader geographic area, either a metropolitan statistical area (MSA) for urban neighborhoods or the non-MSA area of the state for rural neighborhoods. The regulation that implements the CRA legislation specifies that a bank’s record of meeting credit needs also includes its record of lending to LMI individuals, as well as to LMI neighborhoods. An LMI individual may or may not live in an LMI neighborhood, but has an income that is less than 80% of the median individual income for the broad area in which the individual lives.

The CRA regulation states that it is anticipated that banks can fulfill their CRA obligations with loans “on which the banks expect to make a profit.” For an economist, at least, therein lies a puzzle. If LMI lending was profitable, why did banks need the CRA to prod them to make those loans? One possible reason is prejudicial discrimination. Most LMI neighborhoods have a relatively high proportion of minority residents. To the extent that personal prejudices limited a bank’s lending to certain individuals or neighborhoods, the bank would have missed profitable lending opportunities.

Canner and Passmore (1996) suggested an economic foundation for the CRA: imperfect information—in this case, about LMI neighborhoods and individuals. LMI home purchase lending may have been limited insofar as banks believed that there were important, but unknown, differences between it and other home purchase lending. Thus, the CRA may have given banks an incentive to get information that may have revealed previously unrecognized profitable LMI lending opportunities. Of course, such efforts entail costs, which must be taken into account in assessing the net profitability of CRA-related lending.

Has the CRA helped alleviate any shortfall in LMI home purchase lending?

Most of the studies examine data beginning in 1993 or later, largely because that was the first year the home purchase loan data collected under the Home Mortgage Disclosure Act (HMDA) included lending by independent mortgage companies.

The evidence is consistent with a narrowing of any gap between LMI and other home purchase lending during the mid-1990s. Specifically, it suggests that LMI home purchase lending increased more than other home purchase lending during this period. Avery et al. (1999) report that the number of home purchase loans to low-income (below 50% of area median income) and moderate-income (below 80% of area median income) borrowers for properties in MSAs increased 37% and 29%, respectively, between 1993 and 1997, while lending to middle-income (below 120% of area median income) borrowers increased 16% and lending to high-income (at least 120% of area median income) borrowers increased 18% during the same period. Likewise, the number of home purchase loans to residents of low- and moderate-income MSA neighborhoods increased 43% and 32%, respectively, while lending to residents of middle-income and high-income neighborhoods rose 23% and 17%, respectively.

Evidence on whether the CRA per se contributed to the LMI increases being greater than the non-LMI increases is somewhat mixed but tends to favor the view that the CRA did play a role. In support of the view that other developments accounted for the increased access to credit for LMI neighborhood home purchasers, Gunther (2000) found that the LMI neighborhood home purchase loans of institutions covered by the CRA (“CRA lenders”) did not increase faster than the non-LMI neighborhood home purchase loans of CRA lenders between 1993 and 1997, while the LMI neighborhood home purchase loans of institutions not covered by the CRA (“non-CRA lenders”) did increase faster than their non-LMI neighborhood home purchase loans.

Avery, Calem, and Canner (2003) offer some indirect evidence on home purchase lending in LMI neighborhoods that suggests that the CRA did have an effect. They examined housing statistics that likely are positively correlated with home purchase lending. They compared levels and changes in housing outcomes in census tracts just above and just below the 80% income threshold used to determine the LMI status of a neighborhood. They found that, at the time of the 1990 census, census tracts with median family income equal to 75%–80% of median MSA family income had lower homeownership and higher vacancy rates than census tracts with 80%–85% of median MSA family income. However, the LMI tracts had more favorable changes over the 1990s than the other tracts, and, by the 2000 census, homeownership and vacancy rates for the two types of tracts differed little from each other. Because the CRA would have focused on the LMI tracts and not the slightly higher income tracts, the authors suggest that at least part of the improvement in outcomes in the LMI tracts may have been due to the CRA.

Moreover, Apgar and Duda (2003) found that, between 1993 and 2000, the LMI neighborhood plus LMI individual home purchase loans of CRA lenders did increase faster than the non-LMI neighborhood, non-LMI individual home purchase loans of CRA lenders. And by focusing on banks’ “assessment areas,” Apgar and Duda found further evidence in support of the view that the CRA did encourage LMI home purchase lending. Assessment areas are the geographic regions that regulators scrutinize most closely when examining banks for CRA compliance. (Generally, a bank’s assessment areas are where the bank has branches or deposit-taking automated teller machines or where it originates or purchases a substantial portion of its loans.) Therefore, one would expect that, if the CRA were instrumental in encouraging LMI home purchase lending, the highest incidence of LMI home purchase lending would be in banks’ assessment areas. Indeed, the authors found that CRA lenders operating in their assessment areas have a higher share of LMI home purchase loans to total home purchase loans than do either CRA lenders outside of their assessment areas or non-CRA lenders.

While the results in these studies are consistent in some respects with a role for the CRA in narrowing any gap between LMI and other home purchase lending, in one particular respect they call that role into question. If the CRA did help narrow the gap between LMI home purchase lending and other home purchase lending during the 1990s, why didn’t it succeed before the 1990s in effectively eliminating the gap?

Changes related to LMI access to credit

Three changes in the late 1980s and the 1990s may help explain a delay in the CRA’s effectiveness. First, in 1989, the CRA was amended to require public access to CRA examination evaluations and performance ratings. This likely helped motivate banks to comply with the CRA in order to avoid adverse publicity. Second, and perhaps more importantly, in 1995, the CRA evaluation process increased the emphasis on actual lending and decreased the emphasis on banks’ documentation of their efforts to assess community needs. Third, advances in computer and financial technology during the 1990s likely reduced imperfect information problems that may have impeded LMI lending. During this period, credit evaluation techniques and data improved with the increases in computer capacity, computer speed, and accessibility of large stores of financial and demographic information. Once imperfect information problems were sufficiently reduced, LMI lending could grow at a relatively quick pace.

A prominent role for technological change in encouraging LMI lending does not preclude a role for a strengthened CRA. For example, the existence of a strong CRA may have given financial market innovators sufficient incentive to use technological advances for CRA-relevant applications, knowing that, with so many banks subject to the CRA, economies of scale could be realized. Similarly, some observers credit the CRA with having provided the impetus for the development of an improved infrastructure for the financing of affordable housing construction, an improvement that would have been cost effective only on a large scale.

Has LMI home purchase lending been profitable?

Evidence suggests that, recently, for most banks, the profitability of LMI home purchase lending has become comparable to that of other home purchase lending. Meeker and Myers (1996) report that over three-fourths of the banks and thrifts that responded to the authors’ 1994 survey indicated that their institution’s LMI home purchase lending was not as profitable as their other home purchase lending. However, more recent evidence supports the view that, for most banks, LMI home purchase lending is as profitable as other home purchase lending. In a 1999 survey by the Federal Reserve Board (2000), 56% of responding banks reported that the profitability of their LMI home purchase and refinance loans in their CRA assessment areas was about the same as the profitability of their other home purchase and refinance loans. The latter survey may more fully reflect the effects of the computer and financial market innovations of the 1990s. The survey results also may have improved because banks had more experience with LMI home purchase lending by 1999, especially in light of the 1995 change in the regulatory enforcement of CRA that increased the emphasis on actually making the loans.

Consistent with comparable profitability between LMI and other home purchase loans for most banks, results in Canner, Laderman, Lehnert, and Passmore (2003) suggest banks are not subsidizing their LMI home purchase borrowers in the form of interest rates for LMI borrowers that are lower than they would be absent the CRA.


One of the CRA’s goals was to encourage LMI home purchase lending in order to meet more fully the home purchase credit needs of potential LMI borrowers and to do so with profitable loans. LMI home purchase lending has increased more than non-LMI home purchase lending, and the available evidence suggests that the CRA likely did contribute to a narrowing of any gap between LMI and non-LMI home purchase lending. In addition, it appears that whatever LMI home purchase loans the CRA has spurred, those loans have become as profitable as other home purchase loans for most banks.

Liz Laderman

[URLs accessed June 2004.]

Apgar, William, and Mark Duda. 2003. “The Twenty-Fifth Anniversary of the Community Reinvestment Act: Past Accomplishments and Future Regulatory Challenges.” FRB New York Policy Review (June), pp. 169–191.

Avery, Robert, Raphael Bostic, Paul Calem, and Glenn Canner. 1999. “Trends in Home Purchase Lending: Consolidation and the Community Reinvestment Act.” Federal Reserve Bulletin 85, pp. 81–102.

Avery, Robert, Paul Calem, and Glenn Canner. 2003. “The Effects of the Community Reinvestment Act on Local Communities.” Unpublished manuscript. Board of Governors of the Federal Reserve System.

Board of Governors of the Federal Reserve System. 2000. “The Performance and Profitability of CRA-Related Lending.” Report by the Board of Governors of the Federal Reserve System, submitted to the Congress pursuant to section 713 of the Gramm-Leach Bliley Act of 1999.

Canner, Glenn, Elizabeth Laderman, Andreas Lehnert, and Wayne Passmore. 2002. “Does the Community Reinvestment Act (CRA) Cause Banks to Provide a Subsidy to Some Mortgage Borrowers?” Finance and Economic Discussion Series 2002-19, Board of Governors of the Federal Reserve System.

Gunther, Jeffery. 2000. “Should CRA Stand for ‘Community Redundancy Act?’” Regulation 23(3) pp. 56–60.

Meeker, Larry, and Forest Myers. 1996. “Community Reinvestment Act Lending: Is It Profitable?” FRB Kansas City Financial Industry Perspectives, pp. 13–35.

Opinions expressed in FRBSF Economic Letter do not necessarily reflect the views of the management of the Federal Reserve Bank of San Francisco or of the Board of Governors of the Federal Reserve System. This publication is edited by Sam Zuckerman and Anita Todd. Permission to reprint must be obtained in writing.

Please send editorial comments and requests for reprint permission to
Research Library
Attn: Research publications, MS 1140
Federal Reserve Bank of San Francisco
P.O. Box 7702
San Francisco, CA 94120