Labor Markets and Monetary Policy
49th Annual Winter Institute
St. Cloud State University
St. Cloud, Minnesota
March 3, 2011
In the 1970s, the United States experienced high inflation and high unemployment simultaneously. That experience made clear that accommodative monetary policy may not always be an appropriate response to high unemployment. In the intervening thirty years, macroeconomists have engaged in a careful reconsideration of the exact role of monetary policy. The main conclusion of this research is that the primary role for monetary policy is to offset the impact of what economists term nominal rigidities — that is, the sluggish adjustment of prices and inflation expectations to shocks.
With that conclusion in mind, my slides define the natural rate of unemployment to be the unemployment rate u* that would prevail in the absence of any nominal rigidities. To offset nominal rigidities, monetary policy accommodation should track the gap between the observed unemployment rate and u*. The challenge for monetary policymakers is that u* changes over time and is unobservable.
In this speech and the accompanying notes, I ask two questions. First, what can policymakers learn about the current level of u* from the aggregate data on unemployment and vacancies? Second, what other data can be useful in informing policymakers about the current level of u*?
In answering the first question, I apply the canonical Diamond-Mortensen-Pissarides model (for which the three won the Nobel Prize in Economic Science in 2010) to the aggregate data on unemployment and vacancies. I find that the model and data do not provide a definitive measure of u*. Unemployment insurance benefits have increased since December 2007. Firms expect higher taxes and higher input prices. It is possible that these changes in labor market conditions may have been sufficiently large to generate a big increase in u* in the past three years. However, it is also possible that low job creation is largely attributable to nominal rigidities that are generating low demand. In this case, u* will have changed little since December 2007. In the notes that accompany the slides, I show that the resulting possible range for u* is large: from as low as 5.9% to as high as 8.9%.
In terms of the second question about auxiliary information sources, I proceed more heuristically. The basic intuition is that u* is low if high unemployment is being generated by low demand. Hence, I look for information about the current level of demand. I find that business surveys and the current data on inflation both point to u* being low relative to the current unemployment rate.
I conclude that the current accommodative stance of monetary policy is appropriate. However, the Federal Open Market Committee will need to remain vigilant to the possibility of changes in the gap between the unemployment rate and u*.
1 I am speaking for myself today, and not for others in the Federal Reserve or on the Federal Open Market Committee.