According to its official mandate, the Federal Reserve sets the federal funds rate to achieve a dual goal of price stability and maximum sustainable employment. Since the global crisis erupted, debate has been raging over the Federal Reserve's conduct of monetary policy over the period 2002-2006. For example, Taylor (2007) criticises the Federal Reserve for departing from its usual conduct of monetary policy after 2001 and suggests it kept the federal funds rate too low between 2002 and 2006 (see Figure 1). On the other hand, Bernanke (2010) justifies the policy of the Federal Reserve on the grounds that the risks of deflation and high unemployment were particularly pronounced at that time.
Figure 1. The actual federal funds rate versus the prescriptions from an estimated Taylor rule. The estimation period is 1985:Q1 - 2001:Q4. The rule responds to inflation and output growth and features some interest rate smoothing.
In a recent paper (Groshenny 2010), I look at what would have happened to inflation and unemployment if the Federal Reserve had followed a Taylor rule. I use a structural macroeconomic model that describes how the federal funds rate affects inflation and unemployment to perform a counterfactual experiment where the central bank adjusts the policy rate strictly according to a Taylor rule.1
Figure 2 illustrates the implementation of the counterfactual scenario. Black lines represent actual data. The top-left panel shows the estimated deviations from the Taylor rule (monetary policy shocks). In line with Taylor (2007), we see that a string of large negative deviations from the Taylor rule occurred between 2002 and 2006. The counterfactual paths (green lines) of inflation, unemployment and the federal funds rate are generated by turning off these deviations over the period 2002:Q1 - 2006:Q4 indicated by the pink shaded area. The dark shaded areas mark the NBER recessions. We see that a strict implementation of the Taylor rule would have caused a large drop in inflation (year-on-year rate of change in the GDP deflator), with a trough in 2004:Q1. From 2004 to 2005, inflation would have been on average 150 basis points lower than its historical path. The unemployment rate would have increased substantially until 2004:Q2 and would have been roughly 150 basis above its historical path for three years.
Figure 2. The counterfactual endogenous paths (green) of inflation, unemployment and the federal funds rate are generated by turning off the estimated monetary policy shocks over the period 2002:Q1 - 2006:Q4.
Figures 3 and 4 show how the distribution of inflation and unemployment would have evolved over time if the Federal Reserve had strictly followed the prescriptions from the estimated Taylor rule. In 2003:Q4 and 2004:Q1 the probability of inflation exceeding 1% would have been close to zero. In addition, the probability of unemployment greater than 8% in 2004:Q2 would have been over 90%.
Figure 3. Evolution of the distribution of counterfactual inflation over time. The vertical red lines denote the historical inflation rates.
Figure 4. Evolution of the distribution of counterfactual unemployment over time. The vertical red lines denote the historical unemployment rates.
These results are in line with Bernanke's (2010) testimony. They suggest that the deviations from the Taylor rule between 2002 and 2006 reduced the risk of deflation and high unemployment materially, and were thereby consistent with the pursuit of the dual mandate.
The views expressed in this article are those of the author only and do not necessarily reflect the views of the Reserve Bank of New Zealand.
Bernanke, B (2010), “Monetary policy and the housing bubble”, Speech at the Annual Meeting of the American Economic Association, Atlanta.
Gertler, M, L Sala and A Trigari (2008), “An estimated monetary DSGE model with unemployment and staggered nominal wage bargaining”, Journal of Money, Credit and Banking, 40(8),1713-1764.
Groshenny, N (2010), “Monetary policy, inflation and unemployment: In defence of the Federal Reserve”, CAMA Working Papers 37.
Taylor, JB (2007), “Housing and monetary policy”, NBER Working Paper 13682.
1 The model combines the current workhorse for monetary policy analysis, the New Keynesian model, with the search and matching model of the labour market developed by Diamond et al. (2008) which have shown that such a model fits the postwar US macro data reasonably well. The Taylor rule responds to inflation and output growth and features some interest rate smoothing. The model is estimated over the period 1985:Q1 - 2001:Q4, i.e. after Volcker's disinflation and before the period criticised by Taylor (2007).