Bank Value Creation, Intermediation and Managerial Ability

Abstract

Research in banking assumes that the production function is either parametric or deterministic in nature. This thesis develops a novel method for the parametrization of efficient frontiers, generalized frontier analysis (GFA), which uses artificial neural networks and the theory of asymmetric loss functions from forecasting to relax these assumptions. Results show that GFA can validly parametrize cost- and shareholder value efficiency (SHVE). This thesis also validates the SHVE concept by demonstrating that SHVE scores are more informative than managerial ability in explaining bank value creation.

Moreover, the work examines conflicting theoretical predictions about the relations between opacity, fragility and bank intermediation. These are disentangled using measures of bank opacity and liquidity creation from the recent literature. Since available opacity and liquidity creation proxies may not be mutually exclusive, there is the danger of obtaining trivial regression coefficients. Therefore this thesis focuses on intermediation quality, which is operationalized using efficient frontiers. Results show that both opacity and fragility improve the intermediation quality of banks.

Finally, this thesis investigates whether and how bank intermediation activity and managerial ability are related. This thesis hypothesizes, and the data supports, the notion that more able bank managers are both better liquidity creators and more avid risk takers. In addition, the interaction of liquidity creation and managerial ability in crises has thus far not been addressed. While empirical studies suggest that liquidity creation may increase during crises, theory predicts that it may be optimal for banks to reduce intermediation. Analysis shows that more ably managed banks reduce intermediation and risk during the financial crisis as hypothesized.

Overall, this thesis contributes to the banking and efficiency literatures by shedding light on heretofore unaddressed questions regarding the intermediation, managerial ability, shareholder value efficiency and opacity of banks and by developing a new method for the parametrization of efficient frontiers.