The Empirical Relevance of Simple Forward- and Backward-looking Models: A View from a Dynamic General Equilibrium Model

Recent research have provided evidence that backward-looking models fit the data well while purely forward-looking models seem to be inconsistent with data. Consequently, many recent papers in the monetary policy rule literature have used "hybrid" models, which contain both backward- and forward-looking components. In this paper, I demonstrate that a dynamic general equilibrium model with flexible prices and forward-looking properties cannot account for the empirical findings, i.e. that backward-looking behavior seems more important than forward-looking behavior, and that backward-looking models fit the data better than purely forward-looking models. The results also show that the equilibrium model cannot replicate the estimated high weight on backward-looking behavior on US data for the hybrid model.