The Region

Suspending Disbelief: Low Incomes Do Not Mean High Defaults

Top of the Ninth

Gary H. Stern | President, 1985-2009

Published June 1, 1994  | June 1994 issue

This column is based on remarks presented last year at a conference on race, poverty and housing policy at the University of Minnesota Humphrey Institute of Public Affairs.

The striking disparity revealed in the data collected through the Home Mortgage Disclosure Act (HMDA) has been a subject of great interest in recent years. HMDA data has shown significant disparities between mortgage application denial rates for white applicants and those for minorities.

The interpretation of these data has been, and will continue to be, the focus of considerable scholarly and political debate. What is undeniable is that minority applicants are being denied credit with significantly greater frequency than non-minorities. Much of the debate has been shaped by the statistical analysis of the HMDA data performed by the Federal Reserve Bank of Boston and others. From a review of the Boston study and other work in this area, it appears to me that "old style" discrimination is largely non-existent. In other words, clearly qualified applicants of all races are approved for mortgage credit and clearly unqualified applicants of all races are rejected. But the statistical analyses also suggest that imperfect white applicants are more likely to have their loan applications approved than are comparable minority applicants.

A variety of reasons have been advanced to explain this phenomenon. Some have argued that the existence of this disparity is proof that discrimination still exists in society in general and in the banking industry in particular. Others have argued that there is no discrimination evidenced by bank lending decisions in this area and that, in fact, bank lending practices are consistent with long-term default risks associated with loan applicants. Neither of these explanations can be readily tested in a statistically rigorous manner. Still other commentators have suggested that the HMDA data indicate that disparate lending decisions are being made in banks despite policies to the contrary. If this explanation is true, it would appear to offer the promise that both the banks and the regulators should be able to implement corrective action.

Going the extra mile

Without trying to examine the merits of these explanations in detail, I would suggest that a significant part of the problem evident in the mortgage application data may be a lack of personal commitment to address the circumstances of low-income, minority borrowers. Let me state the question in the following manner: Do denial rates for minorities and low- income home buyers reflect the greater risk associated with making loans to low-income and minority borrowers? Or do these denial rates reflect, rather, commitment and costs? Do, in fact, the denial rates for low-income and minority borrowers reflect an unwillingness to go the "extra mile" for those less fortunate?

I think I can illustrate what I mean. How can the Federation of Appalachian Housing Enterprises (FAHE) in Berea, Ky., have a default rate of far less than 1 percent on mortgages it writes? Lenders would all agree that is an enviable rate. And these loans are made even when the families have average incomes of less than $6,000. In fact, this non-profit lender in the Appalachian region of Kentucky and Virginia has had only a single foreclosure in the past eight years. They have found that the most stable borrowers have been single parents with exceedingly low incomes—exactly the opposite of the conventional wisdom. Based on this experience, certainly it is a mistake simply to presume that poor people cannot be expected to be responsible borrowers. We must be careful to avoid accepting this or other myths about behavior.

How can FAHE do this? The answer, at least in part, is that FAHE, as a lender, does not deny its responsibility for the families for whom it is the only realistic hope for a mortgage. FAHE could just walk away from the family in trouble and ultimately stop making loans to families with the same high-risk characteristics—low-paying jobs, bad credit histories, vulnerabilities. But they don't. They work with them. They work through the medical bills when the father breaks his leg and loses eight weeks of work, or when the mother battles pneumonia and loses time from work. They know that when back-to-school time rolls around the mortgage payment might be late, and that at Christmas, times are going to be a little tight.

FAHE and only a handful of other organizations accept the responsibility of going the extra mile—some would say, of suspending disbelief. They accept the responsibility of seeing that poor does not necessarily equate with irresponsible. Most importantly, they accept the notion that we all have something at stake when that family at risk becomes that family in crisis.

To be intellectually honest, we have to recognize that FAHE is a not- for-profit organization. It does not pay taxes. It does not have the obligations to earnings and returns to shareholders that profit-making institutions like banks do. These are important distinctions, because they affect the terms and conditions under which FAHE lends. And they may have significant implications for public policy in this area, and what banks are required or can be reasonably expected to do.

What do denial rates really mean?

Is a bank engaging in discrimination when denial rates for minorities so greatly outpace the rate for white applicants? Perhaps; the study by the Federal Reserve Bank of Boston certainly suggests that is the case. But it may also be a lack of commitment, or abdication of responsibility, that allows the denial of loan applications from minorities to be high. Indeed, it would not be surprising if the emphasis heretofore placed on process in implementing the requirements of the Community Reinvestment Act (CRA) had contributed to a lack of commitment.

Obviously, we cannot easily deal with applications that are denied based upon some discriminatory premise. But discrimination rarely takes the "Get out of my bank, we don't lend to people like you" form. It is more insidious and subtle and will be sometimes impossible to detect.

I would suggest that the larger piece of the denial rate puzzle may well be income related, employment related and credit related—those families (borrowers) that I spoke of earlier. And it is here that indifference, that abdication of responsibility, may be the culprit.

Lest I sound excessively negative, there are things that are working, and in my estimation they are working precisely because they are examples of the commitment to community and responsibility that I mentioned earlier.

The Pine Ridge and Rosebud Indian reservations in South Dakota together have arguably the lowest incomes in the nation. But operating in these two places are two remarkably successful lending programs. One is operated in conjunction with a bank and one is an independent nonprofit organization. On the Rosebud Sioux Reservation, the Sicangu Enterprise Center has to date made 140 loans, with the assistance of the Farmers State Bank in Mission, S.D. Some of these loans are as small as $200, but they provide critically needed capital for micro-enterprises—home businesses—for the Lakota people. Eight peer groups on the Rosebud Reservation have borrowed from this program and have a spotless record of repayment.

Working independently of a bank, the Lakota Fund on the Pine Ridge Reservation of South Dakota has made over 200 loans, some as small as $200. They, too, employ the peer lending concept, in which there is collective responsibility for the borrowers' success.

In Great Falls, Mont., the Neighborhood Housing Services (NHS) program has helped over 120 families, who would not otherwise be able to get mortgages, move into new homes. In 13 years of serving families with incomes as low as $7,000, the Great Falls NHS has had only two defaults. Using grant funds from the Affordable Housing Program of the Seattle Federal Home Loan Bank, they provide down payment and closing cost assistance to very low-income families. Their experience flies in the face of conventional wisdom that if a family cannot come up with a down payment it won't be able to make its monthly payment either. Like FAHE, Great Falls NHS find their most stable borrowers are single parents with children—sometimes the families at greatest risk are the families most likely to persevere.

There are other examples just as compelling, but they do not represent mainstream lending in this country or in the Ninth District. The simple truth is that existing market mechanisms do not reach, or at least do not reach well, the people at greatest risk. We could add examiners and try mightily to root out ever more examples of discrimination, but will we really serve the needs of the community? I am doubtful, and I am doubtful because we may not be getting at the primary reasons for the wide disparity between the denial rates for whites and minorities.

As I have suggested, we must start with commitment and accept personal responsibility. And that commitment must be shared by bankers, their regulators, community leaders in general and borrowers as well.

The premise—and performance—of CRA

At the request of the Clinton Administration, the bank regulatory agencies are in the process of revising the regulatory standards which implement the Community Reinvestment Act. While CRA has been much vilified by both the banking industry and community groups, I find its underlying premise hard to dispute. Simply stated, CRA requires that financial institutions, in recognition of their utilization of federal deposit insurance to attract deposits, demonstrate that they serve the convenience and needs of their communities including the "continuing and affirmative obligation to help meet the credit needs of the local communities in which they are chartered."

One of the objectives of the revisions to CRA is to place more emphasis on performance and less on process. In striving to achieve this goal, I believe that we should heed the lessons from the examples that I have cited. A regulatory structure that allows financial institutions a variety of alternatives to address the needs of their communities through their lending, investments and location decisions is essential. In addition, the examples suggest a prominent role for nonprofit organizations whose mandates differ significantly from those of commercial banks.

Indeed, if we are serious about addressing the problems of access to credit and homeownership for low-income, minority residents, we may well want as a matter of public policy, to channel significant resources to the nonprofit sector and find ways to establish or further strengthen profit and nonprofit partnerships. And we certainly will want to include enhanced training programs and job opportunities as a central part of a solution to these ills.