Memorandum to the Chairman
Committee on Banking, Housing, and Urban Affairs
United States Senate

From: John E. Silvia, Chief Economist
Linda Lord, Chief Counsel
Wayne Leighton, Senior Economist
Dina Ellis, Counsel

July 19, 2000

Section 713 of the Gramm-Leach-Bliley Act (P.L. 106-102) requires the Federal Reserve Board to conduct a comprehensive study of the Community Reinvestment Act of 1977 focusing on 1) default rates; 2) delinquency rates; and 3) the profitability of CRA-related loans. That report was released by the Federal Reserve on July 17, 2000. Despite the acknowledged limitations of the report and its methodology (discussed below), several results are clear.

• By every measure, CRA loans are not as profitable as non-CRA loans.
  
  
• No institution reported that CRA lending was more profitable than non-CRA lending for home mortgage, refinance, home improvement, or small business loans.
  
  
• For home purchase and refinance lending, three times as many institutions reported their CRA-related loans not profitable as compared to non-CRA-related loans.
  
  
• One out of three large institutions report that CRA lending in the home mortgage and refinance markets is not profitable.
  
  
• Three times the value of CRA home improvement loans are not profitable as compared to non-CRA loans.
  
  
• Delinquency rates for CRA loans in the home purchase and refinance market are twice that for non-CRA loans.
  
  
• Among all institutions, about 40 percent of CRA special lending programs are not profitable. For large institutions, 58 percent report that their CRA special lending programs are not profitable. This is inconsistent with the safety and soundness requirements of CRA.

Although better than anecdotal information, the report falls short of being a rigorous, data-based scientific study. In its favor, it may be described as an initial exploratory effort, pointing toward a more comprehensive effort. Among the report's deficiencies:

• Although not required in the statute, survey data is limited to only one year, 1999. This period was characterized by an on-going economic expansion with strong employment growth and low and relatively stable inflation and interest rates. This economic background likely would produce the most favorable survey results for lending, CRA or otherwise. In a more difficult economic environment, the quality of CRA lending would be more adequately tested.
  
  
• Five hundred institutions were surveyed, of which only 143 institutions voluntarily responded. These institutions account for about one-half of the assets of all U.S. banking institutions. Less than a quarter, i.e. approximately 35 institutions, provided quantitative responses. Only subjective qualitative responses were taken from the remaining institutions. Moreover, the value of the data is compromised by its voluntary nature, raising the statistical problems inherent in any self-selected sample.
  
  
• Acknowledging the limited quantitative data, the report's authors instead gave emphasis to "qualitative results" in the report. Such "qualitative" analysis does little to contribute to a dispassionate factual review of CRA and its results. CRA discussion to date has not suffered from a lack of "qualitative results."

Each of these limitations tend to bias the report toward results favorable to CRA.

Respondents provided $570 billion in home purchase and refinance loans in 1999, of which 10 percent were CRA-related; $12 billion in home improvement loans, of which 18 percent were CRA-related; $117 billion in small business loans, of which 50 percent were CRA-related; and $13 billion in community development loans. As described below, when compared to non-CRA loans, CRA-related loans were less profitable in all categories.

• In the home purchase and refinance market, unprofitable lending is three times greater with CRA loans as compared to other loans. Only 6 percent of non-CRA lending in this category was non-profitable, compared to 18 percent for CRA lending.
  
  
• When questioned on relative profitability, no banks stated that CRA loans were more profitable than non-CRA loans for home mortgage, refinance, and home improvement lending. About one out of every five banks stated their CRA loans in this market were less profitable. Two out of every five large banks reported these CRA loans to be less profitable.
  
  
• Delinquency rates for CRA loans are twice that for non-CRA loans in the market for home purchase and refinance (1.57 percent v. 0.79 percent).
  
  
• Small business lending was unique, in that the volume of CRA loans was approximately equal to non-CRA loans. The profitability of CRA loans to small business was slightly less than that for non-CRA small business loans. The study's definition of a small business loan was expansive. The study defined a CRA-related small business loan as any small business loan made within the financial institution's CRA assessment area to (1) a firm with revenues of $1 million or less (regardless of neighborhood income) or (2) a firm in a low- or moderate-income neighborhood (regardless of firm size). Accordingly, the study defined a non-CRA small business loan as only a loan made to a small business outside of the institution's CRA assessment area.

  CRA Special Lending Programs

• A large percentage of CRA special program loans are not profitable. About 39 percent of these loans are not profitable on a per program basis and 44 percent on a per dollar basis. For large institutions, 58 percent of large banks report that their CRA special lending programs are not profitable. This appears to violate the safety and soundness requirements of CRA, which require that CRA lending is to be "consistent with the safe and sound operation" of the financial institution.
  
  
• According to the report, compared to smaller institutions in the sample, large- and medium-sized institutions report a higher percentage of special program loans that are unprofitable and have higher delinquency rates.
  
  
• Financial institutions report that for their "special loan programs" they offer reduced interest rates and fee waivers or reductions for about 47 percent of the programs. They also report that they "alter their customary underwriting standards for a large majority of their special lending programs." The most frequently cited underwriting variances are lower down payments, higher debt-to-income ratios, and the acceptance of alternative measures of credit quality. These lower standards would seem to be reflected in profitability measures.
  
  
• According to the report, delinquency rates are higher on a per program dollar basis than on a per program basis. The study interprets this result as "suggesting that larger programs have higher delinquency rates."
  
  
• According to the report, "[o]btaining either a satisfactory or outstanding CRA rating is a reason mentioned for about 75 percent of the [CRA special lending] programs." (p. xiii)

Return to Chairman's Page