Community Reinvestment Act Proposed Changes to Federal Reserve System

Jennifer J. Johnson, Secretary
Board of Governors of the Federal Reserve System
Washington, DC 20551 Reference: Docket No. R‐1386

Dear Sirs,

GeoDataVision has been working with community and regional banks as well as community organizations with respect to the Community Reinvestment Act (CRA) since 1994. In our capacity as consultants we have worked with more than 700 banks helping them to delineate and map their Assessment Areas, analyze their CRA performance and understand their communities’ need for credit services. Our experience gives us an in‐depth understanding of the Regulation, how it has been applied and how it can be improved to be more effective. The following are our suggestions for improving the regulatory implementation of the CRA.

• Assessment Area delineation: Changes in banking practices and technology have made the implicit assumption that banks primarily derive their deposits from the neighborhoods surrounding their branches obsolete. Today, widespread use of the Internet to attract deposits and remote deposit capture technology, increased use of brokered deposits for deposit gathering and developments in the secondary market, not only for mortgages but for government guaranteed loans (e.g., SBA small business loans) translate into a geographic area for funds much wider than the traditionally defined branch oriented approach. The Regulation needs an alternative for banks that employ non‐traditional funding sources. Regulators rightly have touted the “flexibility” inherent in the application of the CRA, but have taken a rigid and outdated approach to the delineation of Assessment Areas. We suggest that banks be given an alternative Assessment Area delineation method that corresponds to the geographic dispersion of the majority of their deposits or declare themselves to be a “brokered deposit bank” in cases where the majority of deposits are raised from brokers in which case the bank will be allowed to delineate “focus” areas based on where the majority of their lending activity is located.
• Reporting requirements: In 2005 the Regulation was revised to exempt Small banks and Intermediate‐Small banks from reporting requirements under the Regulation. We believe this was a profound mistake for several reasons. First, many community banks not only stopped reporting their CRA‐related lending, they stopped collecting and monitoring their CRA responsibilities altogether. We have spoken with many CRA officers who want to collect and monitor their CRA activity but who don’t receive support from management because it is another cost that is not mandated. The relaxed requirement actually made the job of CRA officers more difficult.

  Second, one of the most important and relevant measures of CRA performance and community credit needs is the credit market data reported under CRA and HMDA. With more than 90% of banks not required to report their CRA small business loans a valuable measurement of local credit needs has been removed making it more difficult to ascertain community needs and to compare bank performance to peers. While large banks may control a substantial majority of the small business loan market nationally, in many local communities they are not dominant. Therefore, an information vacuum exists regarding small business lending activity by local community banks and examiners and bankers are denied an important insight into significant local credit markets. We recommend that all lenders who are required to perform under CRA should also be required to report under CRA. This requirement ironically will make the job of CRA officer easier to fulfill and it will allow more fair and accurate comparisons of lending performance between peer lenders.

• Small Business and Small Farm loans reported: The definition of a small business loan is limited to loans of $1 million or less and has remained unchanged for many years (small farm loans are limited to $500 thousand). We suggest that the size be adjusted to reflect inflation at least every 5 years.
  

  Also, the definition of “renewed” loans is limited to loans whose maturity is adjusted at renewal. We point out however, that a substantial volume of small business lending occurs in the form of lines of credit that are “renewed” annually. However, much of this activity is excluded from data reported because many banks extend lines of credit that are secured by UCC filings and employ demand notes. Consequently, the note maturity is not changed and the “renewal” is not a “renewal” according to the Regulation. This results not only in the omission of much small business lending activity, but also distorts the data because some lenders will be reporting small business line of credit renewals while many others won’t (because of the demand note feature). Examiners rightly insist on data integrity when examining banks. The same concern should be expressed for the market data. A large inconsistency caused by a technical idiosyncrasy undermines the integrity and value of the market data and makes comparison between lenders misleading.

  The structure of reported lending under CRA does not allow the identification of lenders below the county level of activity. This is very different from the file structure reported under HMDA which includes information about lender identity for every loan record reported including census tract information. This is significant because many Assessment Areas for community banks consist of only parts of counties. It is impossible to development market share and market rank comparisons on small business lending below the county level. This would add invaluable insight into the “performance context” that is the backbone of CRA performance analysis. We recommend that the file structure of lending activity reported under CRA be modified to parallel the HMDA file structure. There simply is no good reason not to report this important information. The “micro‐ data” file already contains the necessary information.

  
  
• Community Development: There is no national database of community development loans except for the data released in the Aggregate and Disclosure files. However, that data does not identify the location of Community Development loans thereby severely limiting the usefulness of the data. If banks are required to report the location of small business and small farm loans, why are they not required to identify the location of their Community Development loans? This would be an invaluable help in determining community needs and comparing lenders community development lending performance. The requirement would have negligible cost but would have significant benefits.

  There is no requirement to report any Community Development investments including qualified community development donations. This means the only alternative for comparing banks is to take the CRA performance evaluations of banks collected over different time periods. The result is a very labor intensive project with very limited benefit. If the Regulation requires Community Development Investment it ought to require the reporting of such data so that community needs may be more readily recognized and comparisons made among lenders. Any bank held accountable for performing under CRA abhors an information vacuum in which valuable community data is missing and performance context standards are difficult to ascertain. More Market Information Derived from CRA lenders is a needed resource that makes performance analysis more easy and reliable. Serious CRA officers appreciate this information because it helps them put their Community Development investment activity into its “performance context”.

• Performance Standards: We recommend that the composite performance evaluation ratings distinguish “high satisfactory” performance from “low satisfactory” performance as is done in the lending tests. This allows examiners to notify banks and the public about a bank that is marginally performing its CRA responsibilities. Moreover, we suggest that two consecutive “low satisfactory” composite ratings result in a notice of impending “needs to improve” and a third consecutive composite performance rating of “low satisfactory” would mandate a “needs to improve” rating.
• Performance Context Data: The Agencies use certain demographic data available only from certain proprietary sources. Specifically, the Agencies use “small business demographics” and “family income demographics” purchased from certain vendors. This effectively gives the vendors a monopoly on the data because it is not available from any other source. Moreover, the reliability of the data is highly questionable. For example, the small business demographic data vary significantly from the Census data as published in the “County Business Patterns” database. Although the databases differ in measuring business size (the proprietary source measures by gross annual sales volume and the Census data is based on number of employees) the total number of businesses should be approximately the same. But when we have compared the data in examiner tables to the Census business counts we have discovered variances of more than 200% in the number of business entities in the same market! The family income demographics data is not transparent in terms of how it was derived and the vendor is in the position to charge a very high price because of the monopolistic position the Agencies have effectively provided by the use of the data. We suggest that the Agencies use a more transparent database from the public domain such as the family income demographic database developed by the Department of Housing and Urban Development.

Respectfully submitted by GeoDataVision
Len Suzio, President