The role of labor market rigidities in explaining United States macroeconomic behavior.

摘要

This work investigates the role that labor market rigidities play in the data series implied by dynamic stochastic general equilibrium (DSGE) models of the US economy. This work questions and evaluates the increasing reliance by the macroeconomic literature on the Calvostyle wage rigidity first introduced by Erceg et al. (2000). The baseline case of Calvo wage stickiness is analyzed and compared with the alternative labor market rigidities of staggered relative contracting (Fuhrer and Moore 1995a, 1995b), sticky information (Mankiw and Reis 2001a, 2002), and search and matching frictions (Trigari 2004).; The models are compared by their abilities to match the stylized facts and empirical US evidence as characterized by the impulse response functions (IRFs) resulting from a structural vector autoregression. The focus here is on policy shocks identified with the short-run restrictions of Christiano et al. (1999) and technology shocks identified with the long-run restrictions of Shapiro and Watson (1988) and Blanchard and Quah (1989). Following the approach of Christiano et al. (2001, 2005), a subset of the parameters under each specification of the model is estimated by minimizing the distance between the empirical and model IRFs. This work extends the current literature by nesting these alternative labor market models for the first time in a dynamic, nonstationary general equilibrium model that includes technology growth and can allow for a permanent increase in the level of productivity.; This work offers several conclusions involving the role that alternative labor market rigidities play in the data series implied by DSGE models of the US economy. First, the sticky information framework analyzed here improves on the performance of the flexible wage version of the model when matching impulse responses to a monetary policy shock, but cannot improve on the flexible wage version when matching the responses to a long-run technology shock. Second, the relative staggered contracting specification performs well in matching the empirical responses of a policy shock and a technology shock when output considerations are included directly into contract negotiations. Third, both the right-to-manage and Nash bargaining specifications lose some of their ability to explain labor market behavior and unemployment when extended and nested in a more general, standard DSGE model. With the introduction of a procyclical vacancy equation, these search and bargaining frameworks are able to offer a potential improvement over alternative frameworks through the inclusion of an extensive margin (number of workers) in the labor market. These results suggest that the source of the labor market friction has important implications for the behavior of aggregate macroeconomic variables and thus for the conduct of monetary policy.