Quantitative Performance Metrics for the Community Reinvestment Act: How Much Reinvestment Is Enough?

ABSTRACT

Since the passage of the Community Reinvestment Act (CRA) in 1977, regulators have grappled with the question of how best to evaluate a bank’s performance in meeting the credit needs of its communities. This article contributes to the debate on how to determine a bank’s CRA rating by presenting an analysis of which activities are currently reported as fulfilling a bank’s CRA obligation. Using data on mortgage, small business, and community development lending, investment, and service activities from performance evaluations (PEs) released in 2011 and 2016 for all banks in California, the article answers three questions. First, what are the inconsistencies in what is reported across PEs, and how do they complicate efforts to develop a single metric of CRA activities? Second, how do banks’ CRA-motivated loans and investments vary by markets and economic cycles? Third, to what extent are these loans and investments aligned with the intent of CRA? The results suggest that regulators should focus on reorienting the exam toward giving credit for the loans and investments that promote community development, rather than moving to a single metric based on dollar volumes that could incentivize banks to do less—or even worse, to do harm.

KEYWORDS:

• Community Reinvestment Act
• banking
• regulation

Acknowledgments

Thank you to Lei Ding and Susan Wachter for coordinating this special issue on CRA, and to the participants at the Federal Reserve Bank of Philadelphia Symposium on CRA, as well as two anonymous reviewers, for helpful and constructive feedback. I would also like to thank Sam Giffin, Julie Mendel, Adrian Napolitano, Amelia Baum, and Beatriz Sofia Stambuk-Torres for their excellent assistance and in reviewing and coding the PEs.

Disclosure Statement

No potential conflict of interest was reported by the author.

Notes

1. The full ANPR can be found at https://www.occ.gov/news-issuances/news-releases/2018/nr-occ-2018-87a.pdf

2. Data on mortgage and small business lending are available through the Home Mortgage Disclosure Act and the Federal Financial Institutions Examination Council, but these comprise only a share of banks’ activities under the CRA.

3. The Community Reinvestment Act, Public Law 95–128, Title VIII.

4. The Federal Deposit Insurance Corporation Improvement Act of 1991 expanded the disclosure requirements by requiring the federal banking agencies to disclose the data that support their conclusions.

5. In general, the regulatory agencies jointly develop CRA regulations and guidance and exam procedures, promoting a uniform vision and minimizing the possibility that depository institutions will modify their charters to shop around for a more permissive regulator. To ensure consistency in answering questions about the CRA, the regulators also periodically issue interagency interpretive letters that address policy issues related to the law’s implementation.

6. Regulators adjust the thresholds for inflation, but as of 2018, large banks were those with assets over $1.284 billion, whereas intermediate small banks were those between $321 million and $1.284 billion.

7. Zinman (2002) notes that the addition of small business lending reflected an evolution of CD practice and a shift from focusing on housing to concerns over economic development.

8. No bank can receive an overall rating of satisfactory or outstanding if it does not receive a rating of at least satisfactory on its lending test; conversely, any bank that obtains an outstanding rating on the lending test is guaranteed an overall rating of at least satisfactory.

9. For both neighborhoods and individuals, income categories are defined as follows: low income is less than 50% of MSA median family income; moderate income is at least 50% but less than 80% of MSA median income; middle income is at least 80% but less than 120% of MSA median income; and upper income is at least 120% of MSA median income.

10. All of the exam procedure guidelines for the different bank types can be found online at https://www.fdic.gov/news/news/financial/2006/8cep_otherexam.pdf

11. Examination procedures for the intermediate small banks can be found at https://www.ffiec.gov/cra/pdf/isbank.pdf

12. Initially, I attempted to use data science techniques to scrape these PDFs for the desired data. Unfortunately, because the information was often presented differently in the PEs (e.g., sometimes as text in a narrative with the numbers written out, and sometimes in a table) it was impossible to write a code that could process multiple PEs, and the level of error in the generated data was high.

13. Nationally, 54% of banks evaluated under CRA in 2011 and 2016 were small banks, compared with 43.7% in California. In contrast, my sample is skewed toward large banks (21%) compared with the national distribution (12.8%). The share of intermediate small banks is similar (31.6% nationally compared with 31.3% in California) as is the distribution of CRA ratings (8% outstanding, 89% satisfactory, and just 2% needs to improve).

14. Because the presence of a CD partner is critical to a bank’s CRA activities, I wanted choose a state with a significant number of CD corporations, CD financial institutions, and innovative CD initiatives underway. However, it would be very interesting to replicate this analysis in regions with less CD capacity to see how findings may vary.

15. I also spoke with three CRA examiners from the Federal Reserve to inform myself about the review process; however, I did not do so in a systematic way across regulators. Additional research is needed to understand how examiners are trained in CRA compliance as well as whether there are differences in how the three regulatory agencies treat the exams.

16. For example, if the mortgage lending evaluation covers 18 months, I divide the total value of mortgage loans by 18, and then convert it to a 2-year annualized value by multiplying by 24. If within the same PE the investments are evaluated over 36 months, I use 36 instead of 18 to calculate a monthly amount of CRA activity.

17. One bank also received a satisfactory rating despite the PE indicating that examiners identified “Illegal credit practices inconsistent with helping to meet community credit needs [including] substantive violations of Section 1024 of Regulation X, which implements Section 8 of Real Estate Settlement Procedures (RESPA).”

18. Another example of this arbitrary benchmarking was the examiners’ evaluation of a bank’s loan-to-deposit (LTD) ratio for small banks. Examiners compare a bank’s LTD to that of peer institutions, but in the vast majority of PEs concluded that the ratio was “reasonable.” However, the reported ratios varied widely. For example, a ratio of 72 was compared favorably with ratios of 76 and 78, whereas in another case, a ratio of 72 was compared favorably with ratios of 89 and 105.

19. Further, although CRA does not specifically focus on the racial distribution of mortgage loans, in San Francisco, 45% of loans in LMI tracts went to non-Hispanic White borrowers and 33% to Asian borrowers, with only 3.2% going to Hispanic borrowers and less than 1% to Black borrowers. The overall population of San Francisco in 2017 was 41% non-Hispanic White, 34% Asian, 15% Hispanic, and 5% Black. The remaining share of HMDA loans does not include any information about borrower race or ethnicity.

20. For an excellent discussion of the relative advantages of a standard-based over a rule-based approach for CRA, see Barr (2005).

21. Aggregate data on the total volume of CRA activities do not exist, but in 2015, the FFIEC reported that banks made $87 billion in CD loans and $54 billion in small business loans in LMI neighborhoods. This does not include banks’ substantial investments in LIHTC or New Markets Tax Credits each year, nor the myriad of other programs that banks support as part of their CRA portfolio.

22. In 2015, the enacted budget allocated $3.1 billion to the Community Development Block Grant Program. Data accessed May 3, 2017, at https://portal.hud.gov/hudportal/HUD?src=/program_offices/comm_planning/about/budget/

Additional information

Notes on contributors

Carolina K. Reid

Carolina K. Reid is an associate professor in the Department of City and Regional Planning at University of California Berkeley, and the Faculty Research Advisor at the Terner Center for Housing Innovation.