Empirical Comparison of Sticky Price and Sticky Information Models

Mankiw and Reis (2002) have revived imperfect information explanations for the short run real effects of monetary policy. This paper contrasts their sticky information model with the standard sticky price model. First, I utilize a theoretical relation between aggregate prices and unit labor cost that allows me to leave unspecified household preferences, wage setting and money demand. Second, I introduce a modeling approach that allows me to nest the sticky price and the sticky information models within a single empirical framework. Third, I propose a single-step estimation method that provides consistent estimates of adjustment speeds and reliable confidence bands that enable me to reject flexible prices. Finally, I use the approach to carry out an empirical specification analysis of multiple structural models. An empirical comparison favors the sticky price explanation over the Mankiw-Reis model.