Bank capital, lending, and regulation: A meta-analysis

Abstract

We collected over 1600 estimates on the relationshipbetween bank capital and lending and construct 40 vari-ables to capture the context in which these estimatesare obtained. Accounting for potential publication bias,we find that a 1 percentage point (pp) increase in capi-tal (regulatory) ratio results in around 0.3 pp increase inannual credit growth, while changes to capital require-ments cause a decrease of around 0.7 pp. Using Bayesianand frequentist model averaging, we show that the rela-tionship between bank capital and lending changesover time, reflecting the post-crisis period of increas-ingly demanding bank capital regulation and subduedprofitability. We also find that the reported estimatesof semi-elasticities are significantly influenced by theempirical approach chosen by researchers. Our findingssuggest that the literature fails to provide policymakerswith reliable estimates of the effects of capital regulationon bank lending, and our study offers insights that couldhelp guide future research.