Essays on monetary policy, sticky prices and sticky information

Many studies recently on monetary policy analysis extensively use the sticky price model of price adjustment based on Calvo's (1983) model in a New Keynesian Macroeconomic framework. This price setting model, however, has been criticized for producing implausible results regarding inflation and output dynamics. Mankiw and Reis (2002) propose the sticky information Phillips curve as an alternative to the sticky price model.; In the first study of this dissertation, I replace the sticky price Phillips curve with the sticky information model of price adjustment in a New Keynesian macroeconomic framework. The dynamic systems resulting from these models are solved and impulse responses of output, inflation and interest rates for different policy alternatives by assuming a cost-push shock to the economy are obtained and compared. The results show that impulse responses of the variables to a unit cost-push shock are hump-shaped and the maximum deviations from steady states occur at some quarters after the shock in the sticky information model as observed in data. On the other hand, those responses for the sticky price model are not hump-shaped and the maximum deviations occur immediately with the shock contrary to data. Both models exhibit enough inflation persistence to a cost-push shock, however variables are more persistent in the sticky information model. Both models also exhibit similar short-run tradeoffs between inflation and output.; In the second study, I investigate optimal monetary policy implications of the sticky information model within a New Keynesian macroeconomic framework. The model is solved for optimal policy, and welfare implications of three alternative monetary policy regimes: unconstrained policy, price-level targeting and inflation targeting, are compared when there is a shock to the economy. The results for a cost-push shock illustrate that optimal policy depends on the degree of price stickiness and the persistence of the shock. Inflation targeting is the optimal policy if prices are flexible enough or the shock is persistent enough. However, for a demand shock, inflation targeting emerges as the best policy for all values of the price stickiness and the shock's persistence. When the volatility of nominal interest rate is taken into consideration, the results indicate that inflation targeting is the best policy, in the sense that it results in smaller welfare loss and volatility of nominal interest rate, if prices are sticky enough and the persistence of the shock is large enough. However, price-level targeting might be preferable to inflation targeting if prices are more flexible.; Finally, in the third study, I develop a structural model of inflation by combining two different models of price setting behavior: the sticky price model of the New Keynesian literature and the sticky information model of Mankiw and Reis. In the basic framework of the Calvo model, I assume that there are two types of firms. One type of firm chooses its prices optimally through forward-looking behavior---as assumed in the sticky price model. It uses all available information when deciding on prices. The other type of firm sets its prices under the constraint that the information it uses is "sticky"---as assumed in the sticky information model. It collects and processes the information necessary to choose its prices optimally with a delay. This leads to the sticky price-sticky information (SP/SI) Phillips curve that nests the standard sticky price and sticky information models. This structural model is estimated by a non-linear instrumental variables (GMM) method. The results show that both the sticky price and sticky information models are statistically and quantitatively important for price setting and that sticky price firms make up the majority of the firms in the economy. The resultant SP/SI Phillips curve models inflation better than either the sticky price or sticky information models. The results are robust to alternative...