An important goal of macroeconomic theory is to provide microfoundations for the behavioral equations used in general equilibrium models. Agents’ behavior should be based on optimization subject only to tastes and to technological constraints. This goal extends to price and wage-setting behavior. Fixing prices (or wages) and not revising them in response to shocks should also be optimal given technological constraints that make revision costly.
The “new neoclassical synthesis” in macroeconomics (a phrase coined by Goodfriend and King, 1997, henceforth NNS) has put the endogenization of price and wage setting on hold. The NNS involves using plausible exogenous nominal rigidities, with agents constrained to set prices or wages in advance, in order to explain the quantitatively important and highly persistent fluctuations of macroeconomic aggregates in response to monetary shocks that we observe in the data.
Recently, Chari, Kehoe and McGrattan (2000) established that nominal price rigidity is not sufficient to explain this persistence. In their model, firms that are allowed to adjust their prices make large adjustments that rapidly neutralize the effects of monetary shocks. In earlier work, Ball and Romer (1990) showed that nominal price rigidity cannot be explained as an equilibrium outcome in standard models with constant returns to scale in production and a competitive labor market. In such a setting, firms’ marginal costs are highly procyclical, so they have a strong incentive to adjust their prices in response to a change in aggregate demand, even if other firms keep their prices constant.
The loss in profits from not adjusting prices outweighs the fixed costs of adjusting prices (such as menu costs) unless the latter are implausibly high, and nominal price rigidity is not a Nash equilibrium. Ball and Romer (1990) also showed that with the introduction of real rigidities, which are features that make firms’ marginal costs insensitive to aggregate output, nominal price rigidity is an equilibrium for small adjustment costs. In a related paper, Jeanne (1998) showed that nominal price rigidity combined with real rigidities could generate substantial persistence in dynamic general equilibrium models. These results suggest a close connection in models of nominal price rigidity between persistence and nominal price rigidity as an equilibrium.
Huang and Liu (2002) showed recently that a model with households that rent differentiated labor to firms and set their nominal wages in advance can generate more persistence than an equivalent model with nominal price rigidities. Real rigidities are not necessary in their model: this is important since the empirical plausibility of real rigidities is controversial. Given the link between persistence and supporting nominal price rigidity as an equilibrium, Huang and Liu’s results beg the following question. Can nominal wage rigidity more easily be supported as an equilibrium outcome than nominal price rigidity?
This paper answers the question in the affirmative by developing a simple model of wage determination by monopolistically competitive households. It shows that nominal wage rigidities can be supported as a Nash equilibrium with modest fixed costs of adjusting wages, without introducing real rigidities. It also shows that the adjustment costs that are required remain modest even if labor supply is inelastic, if labor types are highly substitutable in production, and if there are decreasing returns to labor in production. A dynamic extension to the basic model with overlapping wage contracts is used to show that very small frictions are sufficient to generate sizeable nominal wage rigidities as a socially optimal outcome.
The paper is structured as follows. The following (second) section surveys recent papers that address the related questions of persistence and of supporting nominal rigidities as an equilibrium. The third section sets up a simple model of wage determination by monopolistically competitive households that is closely related to the textbook model of Romer (2000). The fourth section addresses the model’s calibration, calculates the size of fixed costs necessary to sustain nominal wage rigidity as a Nash equilibrium, and investigates the robustness of the results to changes in the model’s parameter values. The fifth section addresses the question of the optimal length of wage contracts as a function of the size of the fixed cost of adjusting the nominal wage. The sixth section concludes.
Download
PDF Ebook Nominal Wage Rigidity as a Nash Equilibrium
