سیاست های پولی ذخیره فدرال و غیر خطی بودن قاعده تیلور

We propose and estimate a generalized Taylor rule for the monetary policy of the US Federal Reserve (Fed) to find out how the Fed funds rate is sensitive to changes in inflation and output gap variables in the post war period. We find that Fed's monetary policy has only reacted significantly to changes in inflation when they were between approximately 6.5–8.5%. However, the policy stance change on these changes was relatively small. The findings suggest that the US Fed has been too averse to change from its current monetary policy stance, and that it has not reacted noticeably to changes in the US economic activity, as measured by the output gap. The generalized functional form for the monetary policy rule suggests that similar non-linearity exists in the directional change of the Fed rate.

Over the years, US policymakers have given primary importance to the achievement of price and output stability in the US economy. The primary policy instrument in the hand of the US Federal Reserve (Fed) is the Federal funds rate (Fed rate). Using this instrument, the Fed tries to stabilize the output and inflation of the US economy (Taylor, 1993). The seminal paper on US monetary policy was by Taylor in 1993. He showed that movements in the Fed rate are driven by movements in inflation and output gap variables. Following Taylor (1993), the US monetary policy rule has been investigated further by Clarida et al., 2001, Clarida et al., 1998, Clarida et al., 2000, Judd & Rudebusch, 1998, Kozicki, 1999, Orphanides, 1998, Rudebusch, 2002 and Taylor, 1999 using a linear framework and going back for the sample period as far as 19492. The above-mentioned studies claim that the US Fed rate reacts differently to changes in the inflation and output gap variables, depending on who chairs the US Federal Reserve and their views on current macroeconomic conditions. They estimated the Taylor rule in the sub-sample periods of Miller and Burns (using the sample period from the March quarter of 1970 to the June quarter of 1979), Paul Volcker (using the sample period from the September quarter of 1979 to the June quarter of 1987), and Alan Greenspan (using the sample period from the September quarter of 1987 to the June quarter of 2006) using linear methods. The reason for performing policy analysis over different regimes is to avoid the potential for a non-linear response of the Fed funds rate to output gap and inflation gap variables. If such non-linearity exists in the Taylor rule, and if we use a linear framework, then the ordinary least square estimates are biased. Moreover, the results are inconsistent and the hypothesis tests are meaningless. Such a rule will misguide the economy. Very few studies have shown the non-linearity of the Taylor rule. However, these studies have imposed restrictions on their policy rule and have contradictory findings. For example, Dolado et al. (2003) showed that the US monetary policy can be characterized by a non-linear rule after 1983, but not before 1979. They used a Generalized Method of Moments (GMM) type estimation to estimate a reduced form equation derived from the Central Bank's loss minimization problem to model the policy rule. They did not do the analysis for the whole sample considered in the study but only for part of the Burn-Miller and the Volcker–Greenspan regimes. Osborn et al. (2005) showed, using the framework of Hamilton (2001), which parameterizes non-linear relationships, that the Fed operated a non-linear monetary policy rule during the pre-Volcker period (1960–1979) but not the post-Volcker period. Similarly, Dolado et al. (2005) did not find any evidence of non-linearity in the Taylor rule for the US, but they did find non-linearity in the monetary policy rule of France, Spain, Germany and the ECB. In this paper, we adopt a different perspective to examine whether or not non-linearity exists in the US Fed monetary policy rule; that is, whether the US Fed has reacted non-linearly, depending on the levels of inflation, output gap, and the lagged Fed rate variables rather than who runs the Fed, as previous studies have done. If it has reacted non-linearly, we also examine the nature of such non-linearity. The other subsidiary questions that we attempt to answer are: • Does the US Fed target a particular inflation range, and if it does, what is that target range? • Is the US Fed more sensitive to the inflation range than the output growth target for its monetary policy rule? We propose a non-linear Taylor type rule in a generalized framework, which shows that Fed responses vary with the levels of inflation and output gap variables. Neglecting such a rule leads to a non-linear Fed reaction function, which prior studies have viewed as being due to the monetary regime effect (i.e., who chairs the Fed) rather than the level effect of inflation, for instance. These studies have therefore estimated the Taylor rule in the sub-samples only, to avoid similar non-linearities in the Fed's reaction function. The rest of the paper is organized as follows. In the next section, we discuss the Taylor rule and its forward-looking and backward-looking versions in detail. In Section 3, using a linear framework, we show that the US monetary policy is seen as depending on the monetary regime—masking the level effect of variables like inflation on the Fed's fund rate. Section 4 provides an analytical model for our proposed non-linear monetary policy rule for a Central Bank that unmasks the true relationship between the variables and the Fed's fund rate. The estimation procedure for our proposed monetary policy rule and the estimation results are discussed in Section 5. In the last section, we provide some concluding remarks.

In this paper, we argue that the US Fed conducts its monetary policy by looking at the current macroeconomic conditions, no matter which regime holds the chair of the US Fed. We find that the Fed only reacts significantly to changes in inflation when they are between about 6.5% and 8.5%, or when the expected inflation is in the range of 8-10%. The policy stance change on these changes was relatively small but non-linear. Unlike the Taylor rule, the US Fed overwhelmingly depends on the prevailing Fed rate for its stance on its future rate. The findings suggest that the US Fed has been too averse to changes from its current monetary policy stance, and that it has not reacted noticeably to changes in the US economic activity, as measured by the output gap since 1960.