What Can You Do with CRA Data?

The Federal Reserve Board has published a paper that explores what can and cannot be learned from CRA data. Not surprisingly, the authors, Glenn Canner and Raphael Bostic, both economists with the Board of Governors, find that uses of CRA data are limited by the nature of the data.

Their article contains findings that may help banks with self-analysis and comparisons of their performance to that of other banks. It is useful to have ready a sense of what you can do with your data and how to approach it. Canner and Bostic have given you a list. Their findings will be useful in how you use your data for self analysis.

First, the data are more limited in scope than HMDA data. The data are based on loans made, so there is no measurement of demand. Second, it is aggregated into categories which limits the flexibility of analysis. You are working with totals rather than application data.

Reporting of loan location can obscure the area where the loan proceeds are actually used. Canner and Bostic conclude that the potential for misinformation is not large because small businesses tend to have few offices. Address information based on post office boxes can be more misleading. Their analysis found a pattern of loan concentration in post office geographies because many small businesses - particularly small farms - use post office boxes rather than a street location. They found that of the census tracts for which loans were reported, 31% of the small businesses and 58% of the small farm loans reported were census tracts with post box services.

Because there is no demand information, the data does not provide information that helps to measure the industry's responsiveness to demand or to need. Higher loan volumes reported may relate to need or may be the result of other factors. Further obscuring this question are factors such as underwriting standards in a community, profitability factors and similar credit issues.

Finally, the data show only a limited piece of the total lending picture. The data reporting is based on loans originated. It does not put this into the context of loans outstanding. Nor does the data compare terms of loans.

Clearly what is not reported can be important to understanding the lending dynamics in your community and the successes of your CRA program. What this comes down to then, is the familiar formula of outreach and credit needs assessment. Banks still need to know their community.

Copyright © 1998 Compliance Action. Originally appeared in Compliance Action, Vol. 3, No. 3, 3/98