Loan Monitoring, Competition, and Socially Optimal Bank Capital Regulations

Abstract

Listen

Translate article

INTRODUCTION It is widely believed that the Federal Deposit Insurance Corporation's (FDIC's) flat-rate deposit insurance premium structure encouraged excessively risky bank lending and thereby contributed to the spate of U.S. bank failures during the 1980s and early 1990s. Several reforms have already been introduced to curb this incentive for bank moral hazard behavior. For instance, the 1991 FDIC Improvement Act (FDICIA) mandates the imposition of risk-adjusted capital standards and deposit insurance premiums and early bank closure policies that would reduce the financial burden on the FDIC.(1) However, the fundamental problem in implementing any reform is that the regulator may be less informed than the bank about the quality of the bank's loan portfolio. Several authors have therefore analyzed the design of incentive-compatible regulations which would lead the bank to truthfully reveal its loan quality. But there is little consensus about the existence and nature of such regulations.(2) Furthermore, the regulator's problem is not over if incentive-compatible regulations can be designed. The regulator must next choose one scheme, from an array of incentive-compatible regulations, which will maximize social welfare. However, the welfare implications of regulatory design under asymmetric information have been relatively unexplored in the literature.(3) This article has two main objectives. First, it examines the existence of incentive-compatible regulations in a framework that explicitly incorporates two important features of banking activity: monitoring services provided by the bank to its clients, and potential collusion between the bank and the borrower (through side-payments) to extract a regulatory subsidy. Second, it analyzes the nature of socially optimal incentive-compatible regulations under full information and asymmetric information. A bank in our model performs two intermediation functions. First, the bank provides liquidity services to its depositors for an implicit fee. In equilibrium, this fee equals the value of the services provided. Second, the bank monitors its borrower's project to improve its chance of success.(4) Such monitoring is costly and unobservable by outsiders. Therefore, the market and the regulator make inferences about the bank's optimal level of monitoring which influence the competitive loan rate charged, as well as the regulatory capital requirement and deposit insurance premium actually imposed on the bank. In equilibrium, these inferences are correct. This paradigm yields four main results. First, under full information, banks have an interior optimum capital structure even in the absence of capital regulations. This arises due to the tradeoff between the costs of monitoring and the liquidity service fees foregone and the monitoring benefits earned through increased equity financing. Bank equityholders have a larger residual claim on the loan's payoff and thus have an incentive to increase monitoring effort to improve the loan's probability of success. Optimally, banks use greater equity financing for higher quality loans which are more sensitive to bank monitoring. This result offers a new and empirically testable rationale for the existence of depository institutions. It also provides an explanation of regulatory policy introduced by FDICIA whereby banks with high risk-adjusted capital standards are subject to less regulatory supervision since they are believed to be safer.(5) Second, under full information, social welfare is maximized if the regulator does not impose any capital requirements, when the loan and liability markets are both competitive. There are currently two opposing views in the literature on the effects of capital requirements on bank loan portfolio risk. Koehn and Santomero (1980) and Kim and Santomero (1988) argue that stricter capital requirements would not necessarily lower bank risk, whereas Keeley and Furlong (1990) and Yoon and Mazumdar (forthcoming) support conventional wisdom that they could. …