Assessing the effects of Community Reinvestment Act agreements on bank lending behavior in the 1990s

This dissertation assesses the effectiveness of Community Reinvestment Act (CRA) agreements during the 1990s. CRA agreements are non-binding memorandums of understanding between banks covered by the CRA and community based organizations. CBOs use CRA agreements to negotiate increased community reinvestment mortgage lending by the bank. Recent research has found that CRA agreements increase the availability of residential mortgage credit. However, methodological limitations have either (1) prevented a quantification of the effects of CRA agreements or (2) when the research has quantified the effects, it has failed to secure the causal connection between increased CRA lending activity and the existence of CRA agreements.; Based on six-years of micro-loan data from 25 states and using an analysis of variance (Anova), the research finds that banks that negotiated CRA agreements with CBOs were substantially more involved in all eight “community reinvestment” markets—defined here as applications from and originations to, low-and-moderate income (LMI) persons and LMI census tracts, and applications from and originations to, minority persons and predominantly minority neighborhoods. On average, the metropolitan statistical areas with CRA agreements had LMI market shares that were between 1.0 and 2.0 percentage points larger thanks to the CRA agreements. These are mortgages that “but for” the CRA agreement would not have happened.; The dissertation research also finds that negotiated agreements signed by New Jersey Citizen Action and fifteen lenders active in New Jersey did not appreciably increase those lenders' probabilities of approving Black and Hispanic LMI applicants. This is despite the fact the CRA agreements negotiated by NJCA included below-market interest rates and reduced fees. On the other hand, the agreements appear to have played a substantial role in boosting these lenders' involvement in the Black and Hispanic mortgage credit markets.; The policy implications of the research are twofold. First, the findings argue strongly against the recently passed “sunshine provisions” in the current CRA regulations that require banks and CBOs to report to regulators on their CRA agreement activity. Second, the findings suggest that the relatively modest regulator framework put in place by the CRA and the Home Mortgage Disclosure Act could be transferred to other markets with positive social outcomes.