The Federal Reserve's response to aggregate demand and aggregate supply shocks: evidence of a partisan political cycle.

• Author: Gamber, Edward N.

1. Introduction

   The orthodox approach to modeling the Fed's response to economic conditions is to employ a monetary policy reaction function. Most reaction functions are estimated by regressing a policy indicator, possibly the federal funds rate or a monetary aggregate, on variables that describe the state of the economy, such as unemployment, inflation, and growth in output.(1) Under certain conditions which we discuss below, estimated coefficients from a reaction function provide information about the Fed's monetary policy priorities. Moreover, extending the model to include election and partisan dummy variables may provide information about the Fed's response to political pressures.

   There are two problems with interpreting the coefficients in a standard reaction function as the weights that the Fed attaches to its policy objectives. First, the monetary policy decision may not be represented accurately by an aggregate reaction function (one that models the Fed as a whole) because there are twelve voting members on the FOMC. Chappell, Havrilesky, and McGregor [10] have recently addressed this issue by employing a "disaggregated" set of reaction functions.

   The second problem with interpreting reaction function coefficients as the weights the Fed places on its policy objectives is that this interpretation implicitly assumes structural stability of the underlying macroeconomy. Abrams, Froyen, and Waud [1, 31] acknowledge this shortcoming when they state, ". . . coefficients from estimated reaction functions . . . do not provide direct information on policymaker utility functions. Rather than being the solution to an unconstrained optimization dependent only on policymaker preferences, reaction functions are the output of a constrained maximization, where the constraints are the reduced-form equations that characterize the economy. [Therefore] a finding of instability in policy response would not necessarily indicate that policy formation was subject to political or other pressures." Nearly all studies employing reaction functions, however, avoid this problem by assuming that the structural parameters of the economy are stable over the time period estimated. Recently, for example, Chappell, Havrilesky, and McGregor [10, 185] state, "Under the assumption of a stable macroeconomic structure, estimated reaction function coefficients reveal information about the weight the Fed attaches to the various goal variables."

   Empirical evidence, however, does not support the assumption of a stable macroeconomic environment. In particular, evidence suggests that there is a unit root in GDP and thus, deviations from trend are not purely transitory fluctuations driven by aggregate demand.[2] Output fluctuations are, to some extent, a function of permanent aggregate supply shocks. Reaction function studies that assume a stable macroeconomic environment implicitly assume that all deviations from trend are aggregate demand driven.[3] If, indeed, both aggregate demand and aggregate supply shocks are important determinants of economic fluctuations, then prior interpretations of reactions functions may be invalid.

   To illustrate the importance of allowing for this type of structural change in the underlying model (i.e., aggregate supply shocks), consider the following case: Suppose the Fed is faced with a decrease in output. If the decrease in output is generated by a negative aggregate demand shock, the accompanying reduction in inflation allows the Fed to initiate an expansion and trade an increase in inflation for an increase in the level of output. If the decrease in output is generated by a negative supply shock, however, the Fed is faced with the dilemma of increasing inflation further to achieve an increase in output or decreasing output further to achieve a reduction in inflation.

   The problem with failing to properly identify the source of the shock is most evident in the context of studies of the political business cycle. These studies generally test whether the Fed is pressured to ease monetary policy prior to elections or during Democratic administrations. However, since aggregate demand and aggregate supply shocks may elicit different responses from the Fed, a change in the Fed's response to output may not be due to political pressure but, instead, to a change in the type of shock generating the movement in output. Thus, failure to identify the nature of these shocks may lead to incorrect conclusions about the Fed's response to political pressures.

   The purpose of our paper is to identify aggregate demand and aggregate supply shocks using the method developed by Blanchard and Quah [8] and to measure the Fed's response to each of these shocks. In addition, we measure the difference in the Fed's response to these decomposed aggregate shocks over pre- and post-election periods, and during Democratic and Republican administrations. Thus, while Chappell, Havrilesky, and McGregor [10] disaggregate the Fed's policy vote, we disaggregate the Fed's policy objectives.

   Our results suggest that the Fed responds countercyclically to both aggregate demand and aggregate supply shocks. We find that the Fed responds with greater vigor, however, to aggregate demand shocks. Further, we find that the Fed responds to all shocks with greater enthusiasm during Democratic administrations, particularly during pre-election periods. Thus, we find evidence of a partisan election cycle in the response of monetary policy to aggregate demand and aggregate supply shocks.

   In section II, we present a model of Federal Reserve behavior. We use this model to illustrate how monetary policy can be expected to react to aggregate demand and aggregate supply shocks. In section III, we describe the data and present the method that we employ to identify aggregate demand and supply shocks. In section IV, we present the results of estimates of the reaction function for the period 1955 through 1992. Section V summarizes and concludes the paper.

2. A Model of Fed Behavior

   We adopt the usual convention and assume that the Fed minimizes the disutility associated with deviations in output growth and inflation from their target levels.[4] This assumption implies a quadratic functional form for the Fed's loss function (L):

   L = [Alpha][([y.sup.*] - y).sup.2] + [[Pi].sup.2] (1)

   where [y.sup.*] denotes the targeted natural rate of output growth, [y.sup.*] - y denotes deviations from the target level of output growth, [Pi] denotes deviations from the target inflation rate of zero, and [Alpha] represents the relative importance of the inflation and output goals.[5]

   The Fed minimizes equation (1) subject to the structure of the economy. For simplicity, we assume a general form for the structure of the economy:

   y = y(P, [[Epsilon].sup.D], [[Epsilon].sup.S]), [y.sub.1] [greater than] 0, [y.sub.2] [greater than] 0, [y.sub.3] [greater than] 0, (2)

   [y.sup.*] = [y.sup.*]([[Epsilon].sup.S]), [Mathematical Expression Omitted], (3)

   [Pi] = [Pi](P, [[Epsilon].sup.D], [[Epsilon].sup.S]), [[Pi].sub.1] [greater than] 0, [[Pi].sub.2] [greater than] 0, [[Pi].sub.3] [less than] 0, (4)

   where P represents the Fed's aggregate demand policy instrument, and [[Epsilon].sup.D] and [[Epsilon].sup.S] are aggregate demand and aggregate supply shocks, respectively. The subscripts denote partial derivatives. The signs on the partial derivatives suggest that the Fed can manipulate output and the inflation rate but it cannot manipulate the natural rate of output. That is, the target level of output growth is a function of the aggregate supply shock only. This model is a general description of a Fischer [14] type contracting model where the existence of fixed nominal contracts allows the Fed some short-term stabilization powers.

   The Fed chooses P to minimize the loss function, equation (1), subject to the constraints imposed by the structure of the economy, equations (2) through (4). This minimization yields the following first order condition:

   [Alpha][y.sub.1]([y.sup.*] - y) = [[Pi].sub.1][Pi]. (5)

   To determine...