Consumer protection regulation creates a level playing field for vulnerable categories of consumers against potential abuses by service providers. Legislators around the globe regulate service providers in sectors such as insurance, financial services, and health services. These rules are often implemented in the aftermath of salient abuses against consumers, and in times of pressure from public opinion and the media (Zingales, 2015; Barth, Caprio, and Levine, 2012). Despite their aim, these rules change service providers' incentives in ways that legislators might not envision, and that could ultimately harm vulnerable consumers (Glaeser and Shleifer, 2003; Gargano, Rossi, and Wermers, 2016).  

In a recent paper (D’Acunto and Rossi, 2016), we document the unintended consequences of the Dodd-Frank Wall Street Reform and Consumer Protection Act (Dodd-Frank) in the realm of mortgage origination. We find that, after Dodd-Frank, credit in the middle of the distribution by loan amount decreased by 15%, whereas credit to the top of the distribution increased by 21%. Dodd-Frank changed lenders’ origination behavior because it increased the expected costs of originating mortgages. Lenders faced the higher expected costs of origination and the costs of compliance before the date of Dodd-Frank’s execution, when banks had to be compliant. This announcement effect of regulation is often overlooked by lawmakers (e.g. see D’Acunto, Hoang, and Weber (2016)).

After Dodd-Frank, lenders reduced the originated loans below the mean of the distribution by amount, and increased the originated loans above the mean. This shift in lending towards larger loans should have been more pronounced for larger lenders for at least three reasons. First, large loans cannot be sold to GSEs. Large lenders find it less costly to keep originated mortgages on their balance sheets, because they have a larger amount of deposits. Second, large lenders can offer a larger set of financial services than smaller institutions, and hence acquiring wealthy customers is more profitable to them than to smaller institutions. Third, large lenders operate in several counties, and hence can more easily move their origination to cater to the demand for larger loans. Consistent with these three arguments, we find that the change in mortgage origination behavior increased monotonically with the size of lenders after Dodd-Frank.

In all our analyses, we keep constant the demand-side characteristics of the housing market at the individual-applicant and county levels, including applicants’ income, race, as well as county-level house prices and share of properties foreclosed. We also assess a set of competing demand-side explanations. Anecdotal evidence suggests that after the financial crisis, wealth has polarized, that is, households in the mid-range of the income distribution have moved toward the left tail or the right tail of the distribution. This wealth polarization argument predicts that our results should be stronger in counties with a higher share of middle-class households before the crisis. Instead, our supply-side interpretation predicts that the results should be stronger in counties with a lower share of middle-class households before the crisis, because in those counties lenders would find more wealthy households after Dodd-Frank. The evidence is consistent with the second prediction. The results are also similar for counties with a high or low share of government employees, and counties with a high or low share of households investing in stocks.

To tackle the identification problem, we instrument the current share of large banks that operated in each US county with the pre-crisis share. This variation cannot have been determined by the financial crisis, or by any other shock that affected counties differentially after the crisis. At the same time, there is inertia in the geographic presence of banks across counties. Demographic characteristics are balanced across counties sorted by the share of large banks. This instrumental-variable strategy confirms our baseline results.

Our results speak to the debate about the costs and benefits of regulating economic activity. Proponents of regulation aim to help vulnerable consumers. But regulators often neglect the changes in the incentives of private organizations, which react based on their own objective function. In the case of Dodd-Frank, many middle-class households did not obtain cheaper mortgages, but were cut out of the mortgage market altogether.

Francesco D’Acunto is Assistant Professor of Finance at the R.H. Smith School of Business, University of Maryland. Alberto Rossi is Assistant Professor of Finance at the R.H. Smith School of Business, University of Maryland.