Optimal employment contracts in a principal-agent relationship

Abstract

Two unique characteristics of modern Japanese labor management practice are the prevalence of lifetime employment and of biannual bonuses. In contrast these practices are relatively infrequent in the United States. Many of the explanations offered so far attribute the difference to cultural, and/or institutional differences in the two countries. In this paper, we present a different point of view; one that provides an economic rationale for various practices. The model developed is based upon the framework of the principal-agent relationship. The principal (employer) offers to the agent (employee) a contract (which specifies contract length, wage rate and profit sharing schedule) so as to maximize the expected value of a stream of future profits. The agent in turn accepts this contract if and only if he can make himself better off in doing so. If the appropriate level of effort can be specified and enforced by the principal, there is no ‘agency problem’. The resulting outcome is said to be ‘first best’. On the other hand, in many situations, the principal is not able to enforce the effort level (due to, for example, monitoring costs or lack of observability). Thus, in the presence of this agency problem, the appropriate level of effort is chosen, for any given contract, by the agent in such a way that his expected utility is maximized. The optimal contracts in this framework are said to be ‘second best’. The main focus in the paper is on the second best case. We find that the optimal contracts depend upon various factors such as labor productivity, disutility of effort, risk aversion, alternate market opportunity, transaction cost, demand conditions and discount rates. By using numerical examples obtained by dynamic programming, we illustrate that any form of employment policy can be optimal for the appropriate situations. In addition we estimate the agency cost (or the incentive gap) between the first best and second best solutions.