June 2004 Profitwise News and Views

Bank Expansion Decisions and CRA Performance ¹
by Robin Newberger and Anna Paulson

In early 2004, JP Morgan Chase and Bank One announced plans to create the second largest banking organization in the country. Before the deal was completed, the primary regulator for the acquiring bank, in this case the Federal Reserve, along with state banking agencies and the Department of Justice, had to evaluate the effect of the proposed merger on competition, the financial condition of the new entity and the competence of the managerial resources of the applicant. In addition, bank regulators had to consider each bank's record in providing credit to low- and moderate-income neighborhoods and individuals, according to provisions of the 1977 Community Reinvestment Act (CRA). Community groups also weighed in with their assessment of the consolidation based on CRA considerations. The merger could have been delayed substantially if community groups had identified sufficient grounds for objection.

While the size of the institutions might heighten publicity around this particular merger, the regulatory and public scrutiny is standard for all bank mergers and acquisitions. In this environment, the question has often been asked whether banks strategically increase their share of lending to low- and moderate-income individuals and neighborhoods in anticipation of merging with or acquiring other institutions. While most studies of the Community Reinvestment Act focus on its impact on bank profitability and efficiency,² this article considers whether linking CRA enforcement to the merger review process itself encourages banks to make lending decisions that facilitate the approval of consolidation plans. This possibility creates a near-term concern about "window-dressing" that could mask the actual role a bank plays in a given community. It also has implications regarding the longer-term commitment of banks to meet credit needs throughout their assessment areas once the consolidation process is finished, or more generally, when consolidation does not play a major role in the banking industry.

Empirical analysis helps shed light on this matter. The basic question is straightforward: to what extent does the proportion of lending to low- and moderate-income individuals and neighborhoods (hereafter abbreviated as "CRA lending")³ in one year influence the probability that a bank makes an acquisition in the following year? In other words, do banks increase CRA lending prior to implementing consolidation plans? The analysis isolates the effect of CRA lending by holding other possible influences fixed, such as key bank characteristics (size of assets, capital/asset ratio, bank holding company status) and the calendar year (this captures the effect of other banking trends over the period). The inclusion of these other variables in the analysis disentangles the impact of CRA lending from other possible determinants of acquisitions.

The analysis shows that banks with higher levels of lending to low- and moderate-income neighborhoods do have a greater likelihood of making an acquisition compared to other banks. The higher the percentage of the institution's mortgage originations in a given year that are directed to low- and moderate-income individuals or neighborhoods, the greater the probability that the institution will acquire another bank in the following year.

The impact is both statistically and economically significant. If we were to compare two similar banks, one whose share of CRA loans is in the bottom quartile of all banks (i.e., up to 25 percent of banks do less CRA lending), and a second whose share falls in the third quartile (i.e., a quarter to just under a half of banks do more CRA lending), the second bank would be 8 percent more likely to make an acquisition than the first. To put this in context, the average bank would have to grow its assets by $252 million, or 43 percent, to show a comparable increase in the probability of acquiring another bank.

One might be concerned that it is not CRA lending, but some other characteristic that explains the association between CRA lending and the likelihood of making an acquisition. This possibility is ruled out by a second statistical model that controls for individual bank characteristics. These results show that if a particular bank were to increase its share of CRA lending, the likelihood that that bank makes an acquisition would also increase. These results are particularly relevant for considering the experiences of banks which occasionally take part in the acquisition process. These are the institutions whose decisions about how much CRA lending to do are most likely to be influenced by the possibility of facing the regulatory and public scrutiny associated with an acquisition.

To convince ourselves that these findings are due to the link between CRA enforcement and the merger review process, three additional analyses can be done. The first compares the share of CRA lending in each year studied. Since public and regulatory scrutiny of a bank's CRA record became more intense over the 1991 to 1995 period, CRA lending in later years should have a greater impact on the likelihood of acquisitions. The results show CRA lending in 1990 and 1991 do not have a significant impact on the probability of an acquisition in the following year, but the effect of CRA lending is significant and increasing from 1992 to 1994.

The second analysis compares CRA lending according to asset size. Given that public and regulatory scrutiny is more intense for big banks that make acquisitions, the share of CRA lending at larger banks should exhibit a greater impact on acquisition behavior than smaller banks. Again, the results show that the relationship between CRA lending and the probability of future acquisitions is highest for banks in the top two quartiles of the asset size distribution.

Finally, the results show that CRA lending has no statistical or substantive impact on the likelihood of being the target of a merger. These banks typically face less attention from regulators and the public in the application process.

Linking CRA lending to public and regulatory scrutiny may improve the effectiveness of the Community Reinvestment Act, at least during periods of consolidation in the banking industry. Provisions in the Gramm-Leach Bliley Act, passed in 1999, suggest that CRA may remain an important strategic consideration for banks for some time. As is the case for acquisitions, banks that wish to expand their activities into insurance and/or security underwriting will need to seek regulatory approval, and regulators are required to consider the bank's record of CRA lending in granting it. The findings from this analysis make it important to consider the impetus for banks to focus on CRA lending if consolidations and expansions into new activities become less prevalent. As regulators, community groups and bankers assess whether the trend towards consolidation will continue, understanding the relationship between CRA lending and bank interactions with the public and regulators will be essential for shaping policies to ensure that the credit needs of the broadest spectrum of society are met.

Notes:
¹ This article summarizes research by Anna Paulson, Raphael Bostic, Hamid Merhan and Marc Saidenberg. Their findings appear in "Regulatory Incentives and Consolidation: The Case of Commercial Bank Mergers and the Community Reinvestment Act" (PDF, 382KB).

² Studies have shown that, while lending to low-income individuals and minorities generates greater defaults, greater return volatility, higher operating costs, and increased charge-off rates, lenders are compensated for this and generate similar rates of return compared to banks that do not specialize in loans to low- and moderate-income individuals. Other researchers have also found that banks that specialize in lending to low- and moderate-income individuals are as profitable as other banks.

³ "Lending" refers to home mortgage originations because public data are only available for mortgages over the period studied. The period of acquisitions studied is 1991 to 1995. The information on CRA lending and bank characteristics is taken from 1990 to 1994 (to lag the data on acquisitions by one year).

Robin Newberger is a research analyst in the Consumer Issues Research group of the Consumer and Community Affairs division of the Federal Reserve Bank of Chicago. Ms. Newberger is currently working on research related to the savings behavior of low- and moderate-income people in Chicago. She holds a B.A. from Columbia University and a Masters in Public Policy from the John F. Kennedy School of Government at Harvard University. She is a holder of the Chartered Financial Analyst designation.

Anna Paulson is the manager of the Consumer Issues Research group of the Consumer and Community Affairs division at the Federal Reserve Bank of Chicago. She was a member of the Finance Department at the Kellogg School of Management at Northwestern University for six years prior to joining the Fed. Her research focuses on how households cope with risk and incomplete financial markets. In 1998-99, she was a National Fellow at the Hoover Institution at Stanford University. Her current research focuses on the financial assimilation of immigrants, the dynamics of entrepreneurship and the impact of risk on household decisions. She holds a Ph.D. in economics from the University of Chicago, and a B.A. in economics from Carleton College.