The bank capital-competition-risk nexus – A global perspective

Abstract

Empirical studies of banking risk, be it at the institution or sector level, typically focus on either the relationship of competition to risk or bank capital adequacy to risk, but only a subset of studies integrate the two. Lack of integration entails potential bias arising from omission of relevant control variables, and accurate assessment of the interrelations is particularly important in the light of the introduction of a regulatory leverage ratio alongside risk-adjusted capital adequacy in Basel III, as well as macroprudential surveillance and policy which seeks to forecast, assess and control risk at a sectoral level. To advance the literature, we provide estimates for the relation between capital adequacy, bank competition and four measures of aggregate bank risk for different country groups and time periods. Our modelling approach uses control variables that capture aspects of banks’ business models that contribute to financial stability, aggregated to the level of the banking sector. We use macro data from the World Bank’s Global Financial Development Database over 1999–2015 for up to 112 countries globally. We contend that use of macro data means our results are of particular relevance to regulators undertaking macroprudential surveillance, because such data gives a greater weight to large systemic institutions than the more commonly-used bank-by-bank data. Results largely support “competition-fragility”, i.e. a positive relation of competition to risk controlling for capital; both capital measures controlling for competition are significant predictors of risk, but signs vary across risk measures; the leverage ratio is just as widely relevant as the risk-adjusted capital ratio; and there are some differences in results between advanced countries and emerging market economies. Finally, we find competition drives capital ratios lower in a Panel VAR.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 44 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.