Academic journal article Economic Review - Federal Reserve Bank of Kansas City

A Closer Look at Jobless Recoveries

Academic journal article Economic Review - Federal Reserve Bank of Kansas City

A Closer Look at Jobless Recoveries

Article excerpt

A Closer Look at Jobless Recoveries

By Stacey L. Schreft and Aarti Singh

Most analysts believe the U.S. economy is now recovering from the last recession. This belief is bolstered by the fact that the economy has grown a little more than 2 percent in the past year. Yet businesses have continued to shed workers, prompting The New York Times to dub this "the worst hiring slump in 20 years." Market analysts and economists have a different name for what is happening. They call it a "jobless recovery."

The only other jobless recovery in postwar U.S. history occurred following the 1990-91 recession. In the early years of that recovery, forecasting models based on data from past business cycles predicted that the observed pickup in output would be accompanied by employment growth. Those forecasts were consistently wrong and left policymakers puzzled by businesses' continued trimming of payrolls.

Today, 12 years later, policymakers are again trying to understand the unexpected joblessness of a recovery. Since this is the second consecutive jobless recovery, the possibility that employment will be stagnant in most future recoveries must be considered. This possibility makes understanding the behavior of employment in recoveries, especially jobless recoveries, a priority. Understanding employment, one of the most important variables considered in evaluating economic activity, may enable policymakers to more accurately forecast the pace and strength of recoveries and to develop more effective policy responses.1

This article takes a closer look at jobless recoveries and finds that they have many common features that distinguish them from the typical recovery. Section I presents evidence that the current recovery is indeed jobless and remarkably similar to the recovery from the 1990-91 recession. Section II delves deeper, examining changes in the pattern of employment over time that distinguish the jobless recoveries. The evidence suggests that firms relied more heavily on just-in-time employment practices in the jobless recoveries, substituting more flexible labor inputs-temporary and part-time workers and overtime-for less flexible labor inputs. Sections III and IV consider the implications of just-in-time employment for the pattern of job loss and the overall joblessness of recoveries.


Typically, when output starts growing again after a recession, employment also starts to grow. When employment growth does not resume, the recovery is said to be jobless. This is the definition of a jobless recovery used in this article. It takes the descriptor "jobless" literally: If the growth rate of employment in a recovery is not positive, then the recovery is deemed to be jobless.2

This section first discusses how the joblessness of a recovery can be assessed and then shows that the current recovery, like the previous one, has been jobless. If output growth has also been weak in these recoveries, then some weakness in employment growth would be expected. The section thus goes on to consider whether the recent recoveries were jobless beyond what can be associated with any observed weakness in output growth.

Examining jobless recoveries

Business cycles consist of periods in which economic activity contracts and periods in which it expands. The trough of a cycle marks the date at which a contraction ends and an expansion begins. In the early part of an expansion, the economy is growing, but only making up for the decline in activity that occurred during the preceding recession. This article refers to this phase of an expansion as the recovery.3 Different macroeconomic variables recover at different rates. Output, for instance, recovered within two quarters in the average postwar business cycle.

To assess the joblessness of a recovery, it is first necessary to know when the preceding recession ended. Although the trough of the 2001 recession has not yet been determined, this article follows most analysts in taking it to be December 2001. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.