Understanding bank supervisors' risk assessments

A recurring criticism of U.S. bank supervisors is that their standards vary procyclicly with banking and economic conditions. The 2010 reforms of supervisory standards for bank capital adequacy and liquidity (Basel III) directly address procyclicality in supervision and its effects on credit cycles. We revisit the question of procyclicality in bank supervisors’ standards and find mixed support for the Basel III reforms. Using data on bank supervisors’ safety and soundness assessments of all U.S. FDIC-insured banks between March 1985 and December 2010, as well as information on banks’ financial and macroeconomic conditions, we develop a model of supervisors’ risk assessments of banks: the Ratings Rule Model. We use the Model to examine the relationships between changing risk assessments of banks by their supervisors, bank conditions and economic conditions. Specifically, we estimate the marginal effects of changes in explanatory variables of the Ratings Rule Model on banks’ likelihood of receiving high (low) supervisory ratings. We next analyze the marginal effects and test for significant changes in effects between stressful and non-stressful periods for banking markets. We find evidence that supervisors’ standards for capital adequacy under pillar II of the Basel Accord have been procyclical in the past – becoming more stringent during periods of banking market stress and less stringent during non-stressful periods. In addition, these changes in supervisory risk tolerances for equity capitalization appear to have had a greater impact on risk assessments of sound, well-managed banks than on weak, poorly managed banks. We find, however, that supervisory standards for capital adequacy did not become more stringent during the current financial crisis (2007–2010). Finally, supervisors’ attitude toward other categories of risk – asset quality, earnings strength and liquidity – appear to be somewhat countercyclical.