Skip to main content

2010, No. 37
Posted 2010-12-27

Unemployment and the Role of Monetary Policy

by Brett W. Fawley and Luciana Juvenal

The most recent U.S. business cycle contraction reached its trough in June 2009 according to the National Bureau of Economic Research. Yet, labor markets—at least when measured by the unemployment rate—have yet to show significant improvement. The Federal Reserve has a dual mandate that includes both maximum sustainable employment and price stability: Should Federal Reserve policymakers feel that they must act to reduce unemployment? Or is current unemployment beyond the reach of monetary policy?

Macroeconomics emphasizes that the primary means to reduce the unemployment rate is to increase the growth rate of aggregate demand (setting aside job retraining and similar labor market programs). Since the 1960s, analysts often have referred to the relationship between the growth rate of gross domestic product (GDP) and the unemployment rate during the recovery from a business cycle trough as "Okun's law." Derived from historical relationships, this rule of thumb suggests that the unemployment rate will fall by 1 percentage point during each year that the growth rate of GDP exceeds the growth rate of potential output by 2 percentage points. Other analysts, however, have argued that Okun's law is misleading: Policymakers cannot exploit this relationship because it depends crucially on inflation expectations not increasing following expansionary policy actions. If inflation expectations increase rapidly following a shift toward expansionary policy, the law's relationship vanishes. Further, the law depends on estimates of the growth rate of potential output, which often are highly uncertain.

Although the Fed's dual mandate includes "maximum sustainable employment," these words presumably refer to that portion of unemployment that monetary policy actions have some power to affect: cyclical unemployment—that is, unemployment caused by the diminished demand for workers resulting from a downturn in the business cycle. A second theme, also dating from the 1960s, has been revived recently to question the efficacy of monetary policy to combat current unemployment: structural unemployment (mismatches in the labor market between the skills needed by firms and those possessed by prospective employees). Structural unemployment is one of the two types of unemployment that monetary policy cannot be expected to influence. The other type is frictional unemployment, which refers to workers (voluntarily and involuntarily) changing jobs and the time required to locate better matches between workers and jobs. Batini et al. (2010) suggest that 1.75 percentage points of the current unemployment rate may be attributable to unusually large skill mismatches. Kocherlakota (2010) offers an even higher estimate of 2.5 percentage points. In the recent recession, an additional factor has been the extension of unemployment benefits from 6 months to 99 weeks. Elsby, Hobijn, and S¸ahin (2010) argue the extension has increased the unemployment rate by 1.8 percentage points above where it otherwise would be.

The chart compares the unemployment rate and average duration of unemployment for all months since 1948, highlighting the months since the March 1991 business cycle trough. Note that observations for the past three recoveries (following business cycle troughs in March 1991, November 2001, and January 2009) lie at the upper edge of the scatter diagram; further, the current recovery is conspicuous for its high unemployment rate and duration. The data suggest that the extent of structural unemployment during economic downturns has increased since 1991. Identifying the causes of this phenomenon is an active research area. One hypothesis is that an increasingly rapid pace of technological change erodes worker skills more rapidly than in the past, and that the erosion becomes evident primarily during downturns when separated workers seek jobs with new employers. The increasing duration of unemployment is worrisome because studies suggest that long periods of unemployment reduce the likelihood that a worker will ever find new stable employment.

Do the chart's data also suggest that monetary policy since 1991 might have become less effective in reducing unemployment during cyclical recoveries? Perhaps, but the picture is not clear. Labor productivity increased rapidly during the two previous recoveries but not in the current recovery, reinforcing arguments that inadequate aggregate demand may be the culprit behind this recovery's persistently high and long-duration unemployment.


Elsby, Michael; Hobijn, Bart and S¸ahin, Ays¸egül. "The Labor Market in the Great Recession," in David H. Romer and Justin Wolfers, eds., Brookings Papers on Economic Activity: Spring 2010. Washington, DC: Brookings Institution, 2010, pp. 1-48.

Kocherlakota, Narayana. "Back Inside the FOMC." Presented in Missoula, MT, September 8, 2010.

Batini, Nicoletta; Celasun, Oya; Dowling, Thomas; Estevão, Marcello; Keim, Geoffrey; Sommer, Martin and Tsounta, Evridiki. "United States: Selected Issues Paper." IMF Country Report No. 10/248, International Monetary Fund, July 2010.

© 2010, Federal Reserve Bank of St. Louis. The views expressed are those of the author(s) and do not necessarily reflect official positions of the Federal Reserve Bank of St. Louis or the Federal Reserve System.