Sumit Agarwal, Efraim Benmelech, Nittai Bergman, and Amit Seru write,

We use exogenous variation in banks’ incentives to conform to the standards of the Community Reinvestment Act (CRA) around regulatory exam dates to trace out the effect of the CRA on lending activity. Our empirical strategy compares lending behavior of banks undergoing CRA exams within a given census tract in a given month to the behavior of banks operating in the same census tract month that do not face these exams. We find that adherence to the act led to riskier lending by banks: in the six quarters surrounding the CRA exams lending is elevated on average by about 5 percent every quarter and loans in these quarters default by about 15 percent more often. These patterns are accentuated in CRA-eligible census tracts and are concentrated among large banks. The effects are strongest during the time period when the market for private securitization was booming.

Robert B. Avery and Kenneth P. Brevoort wrote,

We rely on two empirical approaches. In the first approach, which focuses on the CRA, we conjecture that historical legacies create significant variations in the type of lenders that serve otherwise comparable neighborhoods. Because not all lenders are subject to the CRA, this creates a quasi-natural experiment of the CRA’s effect. We test this conjecture by examining whether neighborhoods that have historically been served by CRA-covered institutions experienced worse outcomes. The second approach takes advantage of the fact that both the CRA and GSE goals rely on clearly defined geographic areas to determine which loans are favored by the regulations. Using a regression discontinuity approach, our tests compare the marginal areas just above and below the thresholds that define eligibility, where any effect of the CRA or GSE goals should be clearest.

We find little evidence that either the CRA or the GSE goals played a significant role in the subprime crisis. Our lender tests indicate that areas disproportionately served by lenders covered by the CRA experienced lower delinquency rates and less risky lending. Similarly, the threshold tests show no evidence that either program had a significantly negative effect on outcomes.

It was from David Weinberger that I first heard the suggestion that facts do not settle disputes, because for every fact there is an equal and opposite fact.

Note that the Agarwal paper does not refer to the prior Avery paper, so that there is no discussion of the opposite conclusions arrived at by the two approaches.