84th International Atlantic Economic Conference

October 05 - 08, 2017 | Montreal, Canada

Market discipline on bank risk-taking: How important are liquidity relative to capital requirements?

Saturday, 7 October 2017: 9:40 AM
Colleen Baker, Ph.D. , University of Oklahoma, Norman, OK
Christine M. Cumming, PhD , Economics, Rutgers University, New Brunswick, NJ
Julapa Jagtiani, Ph.D. , Supervision, Regulation & Credit, Federal Reserve Bank of Philadelphia, Philadelphia, PA
The level and composition of capital and liquidity at large banking institutions have changed substantially as a result of new Basel III capital and liquidity requirements. The changes in liquidity in particular appear to have influenced observed bond spreads when included in a simple OLS regression, while changes in capital levels have a muted impact. Moreover, capital and liquidity levels appear to have had little impact on bond spreads in the immediate pre-crisis period.

The two major elements of regulatory and nonregulatory bank liquidity measures are high quality liquid assets (HQLA) and retail deposits. Under Basel III, sixty percent of HQLAs must meet a narrow definition of high quality: assets with sovereign risk and high market liquididty. Retail deposits are seen as very slow to run and therefore nearly stable. In earlier work, we demonstrated that large banking organizations have increased both HQLAs and retail deposits since the financial crisis.

We consider several capital measures, including both new Basel III and pre-existing Basel I/II capital measures as well as SRISK and its components, intended to capture tail or stress risks.

Since market discipline is an important complement to regulatory efforts to promote financial stability, how much attention do bond markets, normally thought to be quite credit-sensitive, pay to banking organization capital levels and liquidity? Is the attention consistent over time? These are likely complex questions. Recent theoretical and empirical work on the relationship between solvency (capital adequacy) and liquidity suggests a systematic, but nonlinear relationship between them. We examine potential specifications of influences on bond spreads that incorporate possible interactions between capital and liquidity and nonlinearities in those interactions.