Abstract:
We find that macroeconomic uncertainty plays a significant role in U.S. monetary policy. First, we construct a measure of uncertainty as felt by policymakers at the time of making their rate-setting decisions. This measure is derived from a real-time, Bayesian estimation of a small monetary VAR with time-varying parameters. We use it to calculate the probability of being in a high-uncertainty regime. Second, we estimate a monetary policy reaction function that, apart from macroeconomic uncertainty, includes Greenbook forecasts, revisions of those forecasts, and a measure of stock market volatility. Using data for the period 1969 - 2008, we find that policymakers set an interest rate that is significantly lower in a high-uncertainty regime, compared to a low-uncertainty regime.
