Some Thoughts on Practical Stabilization Policy

This paper argues that the primary objective of monetary policy should be long run price stability or at least a low average rate of in°ation. But there is also a welfare improving role for monetary policy in helping the economy adjust to nonpolicy shocks. This gives rise to a fundamental tension in the conduct of monetary policy. Understanding this tension is central to interpreting the qualitative properties of actual monetary policy, for evaluating its e®ects and for thinking about alternative institutions that would lead to better monetary policy. Is there a core of practical macroeconomics that we should all believe? Given my limited space constraints, I won't try to discuss a top ten list of eternal macro truths. Nor will I discuss the importance of modelling macro phenomena using quantitative general equilibrium models. Instead I will approach the question from the perspective of stabilization policy. So de ̄ned, my answer to the question is: Yes, there is a core of practical macro. And as regards stabilization policy, most of it has been learned in the past twenty eight years as the profession has collectively struggled with the task of separating the wheat from the cha® of Milton Friedman's Presidential address (Friedman 1968). The net result has been a fundamental shift in methodology and core beliefs about stabilization policy. The critical change in methodology was the switch to thinking about stabilization policy as a game theoretic problem, rather than a control theory problem. The change in substance involved a fundamental shift in the profession's views about the cyclical roles of ̄scal and monetary policy. In sharp contrast to the views that prevailed in the early 1960's, there is now widespread agreement that counter cyclical discretionary ̄scal policy is neither desirable nor politically feasible. Practical debates about stabilization policy revolve almost exclusively around monetary policy. To an extent that would have been virtually impossible to predict thirty years ago, these debates are grounded in a number of widely agreed upon propositions. First, excluding anticipated in°ation e®ects, monetary policy is neutral in the long run. At the very best, the long run Phillips curve is neutral. More likely it slopes the wrong way, so that high rates of in°ation are correlated with low growth rates. Second, persistent in°ation is always a monetary phenomenon. Third, monetary policy is not neutral in the short run. Fourth, most aggregate economic °uctuations aren't due to monetary policy shocks. A corollary of the ̄rst two propositions is that the primary objective of monetary policy should be long run price stability or at least a low average rate of in°ation. A corollary of the last two propositions is that there is a welfare improving role for monetary policy in helping the economy adjust to non-policy shocks. This is because the kinds of models that have any chance at all of accounting for how the economy responds to a monetary policy shock imply that a k% rule for money growth is not optimal. Taken together these corollaries point to a fundamental tension in the conduct of monetary policy. How can we achieve the second objective without compromising the ̄rst? Understanding this tension is central to interpreting the qualitative properties of actual monetary policy, for evaluating its e®ects and for thinking about alternative institutions that would lead to better monetary policy. To be concrete: the Fed tries to achieve the ̄rst objective of monetary policy, a low rate of in°ation, by slowing down the growth rate of money when in°ation starts to rise. Indeed, there are substantial pressures on policymakers to engage in preemptive strikes against in°ation even before actual in°ation moves, say when the unemployment rate approaches the NAIRU du jour. On the face of it, such behavior is di±cult to understand. If persistent in°ation is