February 24, 2009

CRA: A Framework for the Future

Governor Elizabeth A. Duke

At the Revisiting the CRA Policy Discussion, Washington, D.C.

Good afternoon.  I am pleased to be here to discuss the future of the Community Reinvestment Act (CRA).  As many of you probably know, I joined the Federal Reserve Board last year after having spent many years as a community banker.  My experience gives me both a practical and theoretical interest in today's discussion of CRA's future. 

When the Federal Reserve Bank of Boston celebrated the 30th Anniversary of CRA's enactment, it did so by hosting a forum in October 2007 that focused not just on the law's contributions to making financial services more readily available in low- and moderate-income communities, but also on the challenges of applying the law in a financial landscape that is vastly different from the one that existed in 1977.  It was at that gathering of researchers, regulators, bankers, nonprofit practitioners, and community advocates familiar with the CRA that Congressman Barney Frank challenged the Federal Reserve Bank of Boston to collect the best thinking about how the CRA might be improved for the future.

The Bank took this challenge to heart and, together with the Federal Reserve Bank of San Francisco, invited an impressive group of experts to contribute to the creation of a publication serving as the center of discussion at today's forum, Revisiting the CRA:  Perspectives on the Future of the Community Reinvestment Act.  This volume presents a variety of perspectives from many who have been closely involved in the development and implementation of CRA policy over the years.  As one might imagine, there is no shortage of good ideas contained in these pages including suggestions for statutory as well as regulatory changes.  While many divergent views are expressed in this publication, its most striking feature is an expression of a shared understanding that access to financial services for all, including low- or moderate-income communities and individuals continues to be an important goal for policymakers.

In my remarks today, I will highlight what I consider to be the most compelling reasons for making changes to the CRA at this time, discuss many of the themes and issues raised in this publication, and suggest a framework for how, as policymakers concerned with ensuring the full participation of every individual in our economic recovery, we might weigh the various policy options before us.

CRA:  A Model for Government Intervention
Being among experts today, I need not revisit how CRA came into being.  Suffice it to say that Congress, concerned with the worsening condition of many older, urban neighborhoods in the 1970s, wanted to address the lending practices of some financial institutions that took deposits from households in their local communities but made loans and investments outside those same communities.  In addition to this disinvestment, some lenders engaged in redlining, a practice based sometimes on race, at other times on income, and in still others on both factors.  These practices motivated Congress to enact the CRA and, in so doing, it clarified an affirmative obligation for banks and thrifts to serve the credit needs of their entire communities, including low- and moderate-income areas.  At the time, Congress justified this obligation by citing the special privileges conferred upon depository institutions including federal deposit insurance.  By focusing the attention of regulators on this obligation and requiring that a financial institution's performance under the CRA be considered part of its application to merge or acquire another institution, the law provided a strong incentive to meet community credit needs while leaving regulators with a great deal of flexibility to improve upon the implementing regulations as markets evolved.

The most significant changes to the CRA regulations were in response to an executive order issued in 1993 by President Clinton that directed banking agencies to make regulations more performance-oriented.  The regulations created several evaluation methods based on a bank's size and business strategy.  In particular, they created separate lending, investment, and service-test evaluation criteria for large banks.  The revisions helped to stimulate increased lending and investment activity in low- and moderate-income areas.  In fact, some studies suggest that these changes to the regulations coincided with an increase in annual lending commitments from $ 1.6 billion in 1990 to $103 billion in 1999.1  Research that directly tested for the effects of CRA, by comparing mortgage lending volumes and neighborhood outcomes in neighborhoods just above and just below CRA's income threshold, shows that the law has led to additional lending and has had favorable effects on homeownership rates, home values, and vacancy rates.2

CRA was written in a simpler time for banks.  By and large, a local bank's focus, even that of a larger one, was local.  Over time, however, it has become apparent that changes in the structure of the financial industry since CRA's enactment pose a significant challenge to the law as it is written.  The notion of serving a local community has been challenged not only by industry consolidation that has resulted in the creation of very large banks that operate across the nation, but also by advancements in technology that allow banking services to be delivered without traditional branch networks.  The article by Avery, Courchane, and Zorn noted that, by 2007, the 25 largest depository institutions operated nearly 40 percent of all retail banking offices, up from just 10 percent in 1987.  During that same period, their share of deposits more than doubled to nearly 55 percent, while their share of mortgages soared nearly threefold to 67 percent.3  Given merger activity in recent months, consolidation continues to present challenges to local communities as the largest financial institutions grow even larger.  At the same time, many financial institutions have perfected lending through loan production offices, agents, brokers, and electronic means, making the notion of a local lending area based on deposit-taking branches and automated teller machines seem outmoded.

Perhaps even more striking is the dramatic increase in lending from institutions other than the CRA-regulated financial institutions.  As securitization grew, the share of mortgage originations by non-CRA-regulated institutions also grew:  in 1990, only 17 percent of mortgage lending was originated by institutions not covered by the CRA.  This share compares with a peak of 40 percent in 1993 and an average around 30 percent thereafter.4  Moreover, nonbank financial service providers who offer a range of products, such as payday loans, check cashing, remittance services, and other services, have proliferated in recent years and target lower- income areas in particular.  These services are offered at a very high cost, and yet serve needs that are not fully met by regulated financial institutions.  While the trend in mortgage lending has abated very recently due to the subprime crisis, the prevalence of alternative financial services has not, and it is clear that CRA's coverage and, arguably, its impact, have diminished as a result.

After having worked with CRA over the years, I am still surprised at the misperceptions that persist about it.  One widely held misperception is that CRA is only about mortgage lending to low- and moderate-income borrowers in lower-income neighborhoods.  As a former community banker, I know that CRA's impact is just as important in meeting the needs of small farms and businesses and, as such, it serves as a valuable catalyst for job creation in both urban and rural areas across the country.  This point is particularly noteworthy at a time when mounting job losses are adding to the woes of consumers and exacerbating the problems in housing and mortgage lending.  The most recent misperception promulgated by many who either do not know much about the law or don't like it, is that the CRA caused the subprime mortgage meltdown.   

Our recent analysis of CRA-related lending found no connection between CRA and the subprime mortgage problems.  In fact, the Board's analysis found that nearly 60 percent of higher-priced loans went to middle- or higher-income borrowers or neighborhoods, which are not the focus of CRA activity.  Additionally, about 20 percent of the higher-priced loans that were extended in low- or moderate-income areas, or to low- or moderate-income borrowers, were loans originated by lenders not covered by the CRA.  Our analysis found, in fact, that only 6 percent of all higher-priced loans were made by CRA-covered lenders to borrowers and neighborhoods targeted by the CRA.  Further, our review of loan performance found that rates of serious mortgage delinquency are high in all neighborhood groups, not just in lower-income areas. 

These conclusions were supported by research conducted by the Federal Reserve Bank of San Francisco and published in the Revisiting the CRA volume under discussion today.  Moreover, an analysis of foreclosure rates in that study found that loans originated by CRA-covered lenders were significantly less likely to be in foreclosure than those originated by independent mortgage companies.5  Clearly, claims that CRA caused the subprime crisis are not supported by the facts.

The question of the moment, then, is not whether the current crisis was caused by the CRA, but what can be done to make CRA a more effective regulatory incentive going forward as we face an unprecedented set of community needs in the wake of the current foreclosure crisis and the resulting economic slowdown.  Lower-income neighborhoods that were once stable thanks to the efforts of community development practitioners and their banking partners now suffer from the negative effects of high foreclosure rates, including increased numbers of vacant and abandoned property, a loss of equity in their own homes, and higher rates of crime and arson.  Infrastructure in these communities is eroding:  retailers, service providers, and other contributors to the economic health of these neighborhoods are departing. 

Communities with lower levels of personal and neighborhood assets, including many minority communities targeted for higher-priced subprime mortgages, have been particularly hard hit.6  As credit has tightened, less capital is available for these communities.  The pool of loan and investment capital is shrinking at the same time that philanthropic dollars are challenged by investment losses.  To make matters even more challenging, state and local governments face extraordinary budget deficits and have fewer resources to offer these communities.

Given this context, it is no surprise that some members of Congress have expressed an interest in considering measures to improve the supervision of financial institutions, including the role of CRA in providing financial products and services to lower-income communities.  The Federal Reserve Banks of Boston and San Francisco have offered a valuable place to start the discussion of what should come next.  Revisiting the CRA offers more detailed background information about CRA's origins and the developments in the banking industry that I have alluded to here.  Some of the contributing authors recommend revisions to the current regulations, others advocate for the expansion of CRA's coverage, and still others suggest that we start all over again with the creation of a new model for providing financial products and services to underserved communities. 

A Framework for Change
There is no shortage of good thinking in this volume.  And while I am not prepared to pick and choose among the many options presented in it today, I would like to discuss several in the context of some principles that I think provide a useful framework for thinking about any new regulatory initiative.  Framing the issue may be particularly helpful in trying to imagine CRA's future in a new financial environment. 

First, I believe it is important to be clear about the problem that we are trying to solve.  When CRA was enacted, the concern was that lower-income communities did not have access to capital and financial services.  While concern about credit access has begun to resurface in a tightening credit environment, I would hesitate to limit the focus of a new CRA in this way.  Access to credit in lower-income areas has been less of a problem in recent years than has been access to loan products that were fairly priced, well-underwritten by lenders, and well-understood by consumers.  The analysis that I cited earlier regarding the targeting of lower-income and minority communities indicates that we continue to have a problem that warrants regulation.  Whether, as Stella Adams suggests in her article, race should be explicitly considered in evaluating a financial institution's CRA performance or, as Lawrence White suggests in his piece, stronger enforcement of the existing fair lending and antitrust laws make CRA unnecessary, these issues deserve serious debate.  Perhaps, as Larry Lindsey suggests, we should think of financial products and services as a public good and regulate them accordingly. 

In any case, it is important that, first, we clarify the problem that needs to be addressed.  Then, we can address whether the solution requires statutory change or whether it can be successfully resolved through revisions to the regulations.

Second, policymakers need to identify who is in the best position to solve the problem identified.  Bill Apgar argues in his paper that an uneven regulatory structure encouraged mortgage companies and non-bank subsidiaries to engage in subprime lending that banks bypassed because of fair lending and CRA requirements.  He argues that all lenders should have a CRA obligation.  For years, policymakers have hesitated to expand the coverage of CRA because of the nexus the law created between a financial institution's obligation to its community and its access to deposit insurance and other banking system benefits.  Changing that nexus, it has been argued, could make CRA look more like a tax, which could be more easily attacked politically, than a quid pro quo.  While the recent expansion of access to the Federal Reserve's discount window might be seen as an opportunity to expand coverage to new entities, it is not clear to me that this nexus alone should be the basis for expanding CRA's coverage.  The concern Apgar raises over the unintended impact of uneven regulation may, in itself, be reason enough to extend CRA coverage to providers of all consumer credit and financial services.

This concern leads me to my third principle for developing a new CRA structure.  Any new regulatory initiative should be transparent.  As a former banker, I fully appreciate both the need for regulations that are designed to achieve their goals and the need for those goals to be clearly stated.  As a former president of the American Bankers Association, I advocated reductions in the regulatory burden.  Regulatory burden refers not just to the number of regulations that an industry must follow, but also the quality of those regulations.  Much of the motivation for the CRA regulatory amendments in the 1990s stemmed from the vagueness of the original 12-assessment-factor approach.  The revised regulations provided the clarity necessary to spur the increases in lending and investment activity that I referred to earlier.  They provided clearer standards by which financial institutions could organize their activities.  From a consumer perspective, the fact that Congress amended the CRA statute in 1989 to make evaluations public provided the transparency necessary to help create a dialogue between banks and community advocates. This dialogue contributed to an increased number of public/private partnerships that were uniquely successful in addressing the economic and community development needs of lower-income communities.

A fourth principle, closely related to the need for transparency, is that policymakers should balance the benefits and costs of regulation by taking care to tailor laws and regulations to meet their stated goals as effectively and efficiently as possible.  The documentation requirements should take into account the size and resources of the institution, and they should not overwhelm the activities themselves.  There is still room to improve CRA in this respect.  A great deal of justification is required to demonstrate that certain activities qualify for consideration under the regulations.  Much of this is related to the need to demonstrate a benefit to an institution's assessment area.  In order to simplify the regulation's approach to meeting the credit needs of lower-income areas, it may be necessary to consider a different way of measuring impact.  Moreover, measuring performance for any institution must take a broad view of community needs, balancing quantitative measures of performance with consideration of the quality of the credit, investments, and services extended.  Or, as the introductory piece in Revisiting the CRA, by Olson, Chakrabarti, and Essene, suggests, the story of CRA may be one of a regulation that began by measuring process, evolved to a system of measuring outputs, but perhaps, in the future, needs to measure outcomes in lower-income communities.

Finally, any new CRA regulation should be flexible.  One of CRA's strengths has been the flexibility that the consideration of a bank's performance context provides.  The regulatory agencies' ability to redesign CRA regulations over the years kept it relevant as the financial markets changed.  For example, recent revisions recognized the difficulty in meeting the credit needs of rural communities that do not have clearly delineated neighborhoods with differing income levels.  It also offered added incentives for financial institutions to provide assistance in disaster areas.  These provisions may not have been possible if Congress had dictated the exact terms of the CRA evaluation that regulators should apply in determining an institution's CRA rating, and, as a general matter, I believe that Congress was wise not to be overly specific in the CRA.

The future health of communities depends on each of us doing our part to focus attention on the needs of those communities that have been affected by the combined impact of foreclosures and a weak economy.  At the encouragement of Congressman Frank, the Federal Reserve Banks of Boston and San Francisco were inspired to broaden policy discussions by inviting CRA experts to contribute to a very thoughtful volume on the future of CRA.  In formulating a new approach to CRA, I would recommend keeping the most effective feature of the law--its flexibility.  Any new regulatory structure should also be clear about the problem we are trying to solve, it should determine who is in the best position to solve the problem, and it should be transparent and designed to ensure that community benefit is maximized without placing excessive regulatory burden on financial institutions.  If we follow these principles, I am sure that, together, we can determine the most effective way to improve upon a regulation that has had demonstrated success in making credit available to low-and moderate-income communities, and we can further determine the most effective way to strengthen its role as we meet the challenges ahead.


1. National Community Reinvestment Coalition (September 2007), CRA Commitments, www.community-wealth.org/_pdfs/articles-publications/cdfis/report-silver-brown.pdfReturn to text

2.  See www.federalreserve.gov/pubs/feds/2008/200861/200861pap.pdf and www.chicagofed.org/cedric/files/2003_conf_paper_session5_canner.pdfReturn to text

3.  Robert B. Avery, Marsha J. Courchane, and Peter M. Zorn (2009). “The CRA within a Changing Financial Landscape,” Revisiting the CRA: Perspectives on the Future of the Community Reinvestment Act, pp. 34-50. Return to text

4.  Avery, Courchane, and Zorn. Return to text

5.  Elizabeth Laderman and Carolina Reid (2009). “CRA Lending During the Subprime Meltdown,” Revisiting the CRA: Perspectives on the Future of the Community Reinvestment Act, pp. 115-133. Return to text

6.  Lower-income and minority communities were targeted for higher-priced subprime mortgages and, as a result, have experienced high rates of foreclosures.  An analysis of 2005 lending indicates that 40 percent of lending in minority communities (those where more than half of the population was minority) were higher-cost subprime loans, compared with 23 percent of loans made elsewhere.  Similar discrepancies were found in areas where the median income was below 80 percent of the average median income.  Lei Ding, Janneke Ratcliffe, Roberto Quericia, and Michael Stegman (Spring 2008), “Neighborhood Patterns of High Cost Lending: The Case of Atlanta,” Journal of Affordable Housing & Community Development Law, pp. 193-217. Return to text

Last Update: February 24, 2009