Long-run monetary non-neutrality, menu costs and demand-supply interactions: The lessons of some agent-based simulations

Résumé

This paper investigates whether two possible phenomena, the presence of fixed adjustment costs (menu costs) related to firms’ price adjustment and demand-supply interactions are able to serve as theoretically and empirically plausible explanations for the empirical evidence supporting the violation of long-run monetary neutrality. An agent-based menu cost model is developed and calibrated to reproduce some important empirical stylized facts about price changes, which are revealed using a micro-level dataset from the U.S. It is shown that it is possible to come up with model variants, in which the presence of menu costs causes monetary shocks to have permanent real effects, but if firms are assumed to be hit by idiosyncratic productivity shocks, long-run monetary neutrality holds even in the presence of menu costs. Idiosyncratic productivity shocks are necessary for the model to produce realistically large price changes. However, the presence of demand-supply interactions, i.e. a positive feedback from the output gap to potential output leads to long-run monetary non-neutrality even in model variants with good empirical fit. In the fullyfledged calibrated model variant, around one quarter of a typical monetary shock is absorbed by real output in the long run. This suggest that monetary policy may have substantial long-run real effects. Two limitations of long-run expansionary monetary policy are pointed out. First, its effectiveness decreases with the size of the monetary shock. Second, if price adjustment is asymmetric because of the presence of trend inflation, there is an intermediate range of the shock size, within which negative monetary shocks are more effective in the long run than positive ones.