Test of the Taylor Rule and Policy Implications

Abstract

The Taylor rule [CRCSPP, 1993; MS, 1998; MPR, 2001] describes the behavior of the Fed in determining the federal funds rate and is given by: R 1⁄4 πþ 0:5 GAP þ 0:5ðπ π*Þ þ R* where R, π GAP, π*, and R* are the federal funds rate, the inflation rate, the output gap, the inflation target, and the real interest rate. Taylor assumed a 2% inflation target and a 2% real interest rate and suggested that the Fed would raise the federal funds rate by 0.5 percentage points if actual GDP is greater than potential GDP by 1.0 percentage point or if the inflation rate is greater than the inflation target by 1.0 percentage point. Judd and Rudebusch [FRBSFER, 1998] indicated that the Taylor rule is a useful guide to characterize major elements of US monetary policy. Kozicki [FRBKCER, 1999] presented the strengths and weaknesses of the Taylor-type rules. Mehra [JEB, 2001] suggested the inclusion of the long bond rate in the Taylor rule. Bullard and Schaling [FRBSLR, 2002] argued that the stock market performance should not be considered by the Fed while Rogobon [FRBSLR, 2003] maintained that US monetary policy responded to the stock market performance. Chadha, Sarno, and Valente [IMF Staff Papers, 2004] found that exchange rates and asset prices may be considered in the Taylor-type rules for the US, UK, and Japan. Osterholm [AEL, 2005] showed that the Taylor rule as a tool for U.S. monetary policy seemed to be questionable. Tchaidze [IMF Staff Papers, 2005] indicated that a simple Taylor rule may misspecify the monetary policy reaction function if time-varying inflation targets are considered by policymakers. This paper attempts to examine whether the Taylor rule may be confirmed empirically based on the simplified equation with specific values for π* and R* suggested by Taylor: R 1⁄4 1þ 1:5p þ 0:5 GAP. The sample consists of quarterly data during 1987.Q3 to 2005.Q3 with a total of 74 observations. This period coincides with the Greenspan era. Quarterly data for actual GDP and potential GDP Atl Econ J (2007) 35:121–122 DOI 10.1007/s11293-006-9056-4