Testing between Competing Models of Real Business Cycles

Abstract

This paper tests a real business cycle model with efficient long-term labor contracts (the efficient long-term contract model) against a standard real business cycle model (the intertemporal substitution model). In the former model, employment and real wages are determined by bilateral dynamic bargaining between firms and workers. In the latter model, employment and real wages are determined instead by the dynamic optimization of households within the competitive market framework. We estimate each model using aggregate Japanese data. Our results show that the data are consistent with the efficient long-term contract model, but are inconsistent with the intertemporal substitution model. A business cycle phenomenon is typically found in time-series data on employment and real wages. According to the data of almost all developed countries, employment fluctuations are considerably greater than wage fluctuations. Many recent economic studies have attempted to explain this phenomenon. Most of these studies fall into the equilibrium business cycle theory in which business cycles are the result of individual agents optimizing in a competitive environment. Among the most famous early examples of such a theory are the rational expectations, general equilibrium models of Lucas (1972) and Barro (1976) which stress the role of nominal shocks in the presence of imperfect information. More recently, the predominant equilibrium business cycle theory has been the real business cycle model of Kydland and Prescott (1982), Long and Plosser (1983) and King and Plosser (1984) which emphasizes the importance of real shocks to production technology. Real business cycle theory considers a model economy populated by a single infinitely-lived individual with given initial resources, production possibilities and tastes. The representative individual chooses a preferred consumption-production plan and the resulting allocation is Pareto optimal. The implications of real business cycle theory are thus derived from the Pareto optimal equilibrium allocation as calculated from the planning problem of a social planner or representative agent. A policy implication of this theory is that monetary policy has no significant effect