How effective is fiscal policy: Guest post from Roger Farmer (crosspost at CT)
Roger Farmer, professor of economics at UCLA, has sent a response to my post on the fiscal multiplier, which is over the fold. I’ll make some substantive points in comments, but I’d like to start by saying that this is a good example of a discussion to which blogs are ideally suited. Contributions from people like Roger who have something important to say, but not the time or inclination for a regular blog, make it even better.
John. You ask for responses from DSGE theorists to the question; how effective is fiscal policy? Let me provide a personal response.
In the opening paragraph of your post you assert that: “The biggest theoretical issue in macroeconomics is ‘what causes unemployment’.” I prefer the following characterization of the state of our subject:
There are two unanswered “big theoretical issues” in macroeconomics. The first is; what causes large sustained shocks to aggregate demand of the kind we saw in 1929 and again in 2007? The second is; why does high unemployment persist?
Distinguishing these two puzzles is central to arriving at an understanding of the crisis that finds most western economies operating with high and apparently sustainable unemployment rates.
There is an active discourse in the blogosphere that characterizes the debate as a battle between a group of classical fiscal policy deniers and a rearguard group of Keynesian realists who carry the truth of the Master. In this characterization of the intellectual discourse the evidence is plain: fiscal policy matters and it matters in a big way. If stubborn government ministers would recognize this simple truth, a large and sustained fiscal boost by leading economies would solve all of the world’s problems.
I consider myself to be equally entitled to the label ‘Keynesian’ as the leading proponents of fiscal expansion, but I do not share the consensus view of self proclaimed Keynesians that a large fiscal expansion is the solution to our dilemma.The slavish devotion to remedies to the Great Depression that were proposed more that eighty years ago smacks of religion; not science. As Keynes himself said on leaving the Bretton Woods convention in 1946, “I was the only non-Keynesian in the room”.
Writing in 1968, Axel Leijonhufvud made the distinction between Keynes and the Keynesians. That distinction is useful because it clarifies the idea that the formulation of the General Theory that was popularized by Samuelson in the third edition of his undergraduate textbook misses the key idea of Keynes’ masterpiece; that high unemployment is an equilibrium phenomenon, not a temporary deviation from a classical ideal state. Samuelson’s interpretation is, to quote Joan Robinson, “bastard Keynesianism”.
In my introductory paragraph, I raised two questions. Let me begin with the second question first: Why does high unemployment persist? Samuelsonian Keynesians see the issue as one of deficient nominal aggregate demand. Unemployment is high because nominal prices and nominal wages are too high. We can wait for nominal wages and prices to adjust, or we can boost demand through monetary or fiscal policies taking existing wages and prices as given. Since the money interest rate is at its lower bound, monetary policy will not work. That leaves fiscal policy to boost demand.
What’s wrong with that analysis? It suffers from the uncomfortable fact that the General Theory does not contain an explanation of unemployment that is consistent with rational behavior by profit seeking firms in the labor market. There is no coherent theory of unemployment in the General Theory; there is simply an assertion that households are not on their labor supply curves. That lacuna pervades all of the new-Keynesian models developed before the 2007 crisis and many of those developed since. In almost all of these models there is no unemployment since households are always on their labor supply schedules. A recession is, to quote James Tobin, “a sudden attack of contagious laziness”.
There is a group of DSGE models that account for unemployment by combining search theoretic models of the labor market with inter-temporal choice by maximizing households. These models are surveyed in a piece I wrote for a special issue of Macroeconomic Dynamics that is available here. In these models there is a coherent account of the idea that there may be a continuum of steady state unemployment rates, and unlike bastard Keynesian analysis, the problem is real not monetary. That is a disturbing thought to those who would boost demand by purely monetary means since it suggests the possibility that we may end up with 8% unemployment and high inflation simultaneously.
Nothing I have said so far is inconsistent with the notion that fiscal policy is the answer to our dilemma. That is where we would be wise to ask ourselves the first question posed in my introductory paragraph. What causes large sustained shocks to aggregate demand? The answer to that question is contained in a second strand of research that I have been engaged in both individually, here, and jointly here. Competitive unregulated financial markets cannot be expected to find the “right” relative price of capital except by chance, for the reasons expounded more than thirty years ago in a research agenda that was initiated at the University of Pennsylvania. Demand shifts occur when rational forward looking agents, all of whom have rational expectations of future prices, revise their beliefs of what will occur in the future. Agents are rational. Markets are not.
This brings me back to the correct policy response. Monetary policy has two dimensions. One is the expansion of central bank balance sheets to increase the money supply with the goal of boosting nominal demand. The second is a change in the composition of the balance sheet with the goal of influencing the relative price of assets, with the goal of boosting real aggregate demand.Willem Buiter has called the first policy; quantitative easing and the second; qualitative easing.We need more qualitative easing.We may need a bigger government sector; but that is a logically separate question from the issue of how to cure the unemployment problem.
Will a big fiscal expansion cure the unemployment problem? Probably. But the only time it was unambiguously successful in the US was during WWII when government went from 15% of the economy to 50% in the space of two years.We could put all of the unemployed people into the army. I do not think that is the right response. Alternatively, we could give private companies the incentive to employ more people by moving the economy from a high unemployment equilibrium to a low unemployment equilibrium. That solution would require bloggers, politicians, journalists and policy makers to consider some new ideas.