Surprising Comparative Properties of Monetary Models," working paper

In this paper we investigate the comparative properties of empirically-estimated monetary models of the U.S. economy. We make use of a new research tool—a data base of models designed for such investigations. We focus on three representative models: The Taylor (1993a) model, which has already participated in several model comparisons, and two well-known examples of the current generation of new Keynesian models, the Christiano, Eichenbaum, Evans (1995) model and the Smets and Wouters (2007) model. Although the three models differ in terms of structure, estimation method, sample period, and data vintage, we find surprisingly similar economic impacts of unanticipated changes in the federal funds rate. However, the models have different policy implications regarding the appropriate policy response to other sources of economic fluctuations. Optimal simple policy rules that set the interest rate in response to inflation, the level and growth rate of output, and the lagged, interest rate are not robust, that is, the rule that is optimal in one model performs quite poorly in the other models. Then, using the model data base, we show that a rule with no interest rate smoothing and no response of the interest rate to the growth rate, as distinct from the level, of output, delivers a more robust stabilization performance