Deviations from rules-based policy and their effects

Abstract

Rules-based monetary policy evaluation has long been central to macroeconomics. Using the original Taylor rule, a modified Taylor rule with a higher output gap coefficient, and an estimated Taylor rule, we define rules-based and discretionary eras by smaller and larger policy rule deviations, the absolute value of the difference between the actual federal funds rate and the federal funds rate prescribed by the three rules. We use tests for multiple structural changes to identify the eras so that knowledge of subsequent economic outcomes cannot influence the choice of the dates. With the original Taylor rule, monetary policy in the U.S. is characterized by a rules-based era until 1974, a discretionary era from 1974 to 1985, a rules-based era from 1985 to 2000, and a discretionary era from 2001 to 2013. With the modified Taylor rule, the rules-based era extends further into the 1970s and there is an additional rules-based period starting in 2006. We calculate various loss functions and find that economic performance is uniformly better during rules-based eras than during discretionary eras, and that the original Taylor rule provides the largest loss during discretionary periods relative to loss during rules-based periods.