Discrimination Testing Heads Closer to Home Loans

In June, the Federal Reserve Board's Consumer Advisory Council voted overwhelmingly to urge the board to fund an extensive pilot program of testing to detect discrimination in mortgage lending. The action serves as a further indication that credit discrimination is emerging as a critical compliance issue. For financial institutions, the council's decision represents a very substantial step in the direction of intensified enforcement of the Fair Housing Act. This fall the governors will consider the council's recommendation. Nearly all the governors attended the June meeting of the council, an advisory group representing consumer, academic, and financial industry interests. Background. Testing is an investigative technique used for many years by civil rights groups and some government agencies to check for discrimination and other illegal behavior in the sale and rental of housing. Pairs of individuals matched in all significant respects except one-generally race-approach businesses to seek housing or credit. The results are assessed to see if both receive comparable treatment. Civil rights advocates have long suggested that testing should be used widely to detect racial discrimination in credit. When I was deputy comptroller of the currency in the late 1970s, the banking agencies evaluated proposals to adopt testing. At that time, the agencies decided the difficulties of attempting testing outweighed arguments in favor of it. Points against testing. There are five main arguments against testing: (1) It is expensive, and may be a waste of money unless there is reason to believe there is widespread credit discrimination in the banking system-a finding not supported to date by examinations. (2) There are great practical-and possibly legal-difficulties in equipping testers with fictional profiles that can stand up to the credit checking process. (3) It raises ethical concerns. For example, is it fair to the lenders involved? (4) It may erode the essentially cooperative relationship between institutions and regulators. (5) It may not produce reliable or meaningful results. Why support grows. While such arguments have dissuaded the agencies in the past, there is reason to believe that the situation has changed. First of all, several highly publicized studies have led some observers to conclude that there may be widespread discrimination in mortgage credit. These analyses range from the Atlanta Journal-Constitution's Color of Money articles to research by the Federal Reserve Bank of Boston. These studies have found patterns in Home Mortgage Disclosure Act data that suggest great disparities in the amount of credit extended in predominantly minority areas versus other areas, even when differences in factors like income and property values are held constant. These studies, far from conclusive, have convinced many civil rights groups as well as members of Congress that a serious discrimination problem exists. A second factor is a result of the Financial Institutions Reform, Recovery and Enforcement Act. FIRREA greatly expanded the Home Mortgage Disclosure Act, adding new requirements that institutions it covers must collect and disclose not only where they make their home loans by census tract, but also information on denied applications and on the race, sex, and income of each applicant. These revisions make it possible, for the first time, to compute denial rates readily by race and sex of applicants, holding income constant. Analyses based on the new HMDA data, of course, will still fall far short of proving whether discrimination exists. Income is not a proxy for creditworthiness, and the new data will not show such factors as debt load that may explain denial patterns. But it is the surface appearance bankers must be concerned with. If new studies echo the patterns revealed by the earlier ones, they will undoubtedly create intense political pressure on the agencies for tougher nondiscrimination enforcement. …