Are banks risk-averse or risk-neutral investors?

In this study, we verify how to construct the utility maximization problem for banks in their asset allocation. We consider two utility maximization problems for the risk-averse and risk-neutral banks to determine the optimal lending ratios to represent banks' optimal asset allocation. We apply the mean- variance utility for the risk-averse problem and impose the risk constraint for the risk-neutral problem to obtain the optimal solution. In order to validate the model, we investigate how the optimal lending ratios derived from the two models fit the actual bank lending ratio through calibration. Statistical tests for the calibration results do not indicate a significant difference in the model fitting between the risk-averse and risk-neutral models. Hence, this enables us to use both models to describe banks' asset allocation.(c) 2023 Elsevier B.V. All rights reserved.

Automated summary

This study verifies the construction of utility maximization problem for banks in their asset allocation. It considers both risk-averse and risk-neutral utility maximization problems, determining optimal lending ratios as the representation of banks' optimal asset allocation. The mean-variance utility is applied to the risk-averse problem and a risk constraint is imposed on the risk-neutral problem to obtain the optimal solution. Through calibration, the study investigates the fitting of the optimal lending ratios derived from the models to the actual bank lending ratio. The statistical tests indicate no significant difference in model fitting between the risk-averse and risk-neutral models, enabling the use of both models to describe banks' asset allocation.