The Impact of Regulatory Stress Tests on Bank Lending and its Macroeconomic Consequences

Abstract

We use an expansive regulatory loan-level data set to analyze how the portfolios of the largest US banks have changed in response to the Dodd-Frank Act Stress Test (DFAST) requirements. We find that the portfolios of the largest banks, which are subject to stress-testing, have become more similar to each other since DFAST was implemented in 2011. We also find that banks with poor stress-test results tend to adjust their portfolios in a way that makes them more similar to the portfolios of banks that performed well in the stress-testing. In general, stress-testing has resulted in more diversified bank portfolios in terms of sectoral and regional composition. However, we also find that all the large banks diversified in a similar way, creating a more concentrated systemic portfolio in the aggregate. Finally, we analyze the effects of stress-testing and portfolio sensitivity to macroeconomic scenarios on credit supply. Our findings indicate that banks that experience worse results in the stress tests cut lending relative to their peers and specifically in loans that are most sensitive to the stress-test scenarios. At the borrower level, firms that rely more on credit from banks with poor stress-test results are not able to substitute lost funding and therefore face a larger reduction in credit and cut back investment. These results highlight a macroprudential effect of stress-testing: Credit growth is curtailed during a credit expansion in those banks holding a portfolio that is more sensitive to stressful scenarios. Hence, these banks are expected to be in a more resilient position at the onset of a downturn.