What goes around comes around: Jobless recoveries nothing new

New research by W. P. Carey School of Business faculty finds that jobless recoveries have been with us far longer than most experts think. In fact, sluggish job growth has followed U.S. recessions since at least 1950. And there are ways employers can adjust their hiring and retention practices to better weather this phenomenon.

The 2001 recession Americans have been struggling through doesn't feel like it's over from a jobs perspective — a situation similar to the 1991 recession 10 years earlier. As the United States struggles with its second "jobless recovery" in the last 15 years, new research may shed some light on the dynamics involved in rebounding from recessionary times.

In a recent research paper, W. P. Carey School of Business economics professor Richard Rogerson and doctoral student Kathryn Koenders propose a fresh approach to this challenging topic.

Rogerson and Koenders note that one of the accepted facts of the business cycle is that employment and output have a tight correlation, even though employment lags output by about one quarter. However, the apparent slow growth of employment during the recovery periods after the last two recessions challenges this characterization.

Typically, average hours worked in manufacturing will decline once a recession takes hold because overtime hours are lopped off. Then, once the recession eases, overtime hours are added back on until, at a certain point, workers need to be hired again to meet demand.

For instance, within the manufacturing sector, the average weekly hours of production workers was 39.8 hours in 1970, according to the Bureau of Labor Statistics — a low figure in this half of the century. The hours worked in 1991 were just 40.4 hours and 40.3 for the most recent recession in 2001, all considered on the lower end for average hours worked each week.

As part of their fresh understanding of this phenomenon, the authors point out that the two most recent recessions in the U.S. have a key commonality — both followed unusually long expansions. Spurred by this observation, they propose "an economic mechanism that links the speed at which employment increases during the recovery from a recession to the length of the expansion preceding the recession."

This mechanism underlines the manner in which organizations seek to shed unneeded labor, according to the research, and specifically there is an assumption that inefficiencies with regard to the use of labor emerge over time within an organization.

"Eliminating these inefficiencies (a process we refer to as reorganizing) requires scarce organizational resources that must be diverted away from current production... and we show that reorganization will be postponed to periods in which production is relatively low," write Rogerson and Koenders.

The paper also concludes that, after a long expansion, obviously many more organizations have postponed their reorganization efforts. Since such reorganization leads to the elimination of jobs (over an extended period of time) it gives rise to a longer period in which the economy is shedding labor. Thus, this delays the date at which new employment begins to create a recovery period.

The paper was based on the study of eight U.S. recessions in the post-1950 period. Rogerson and Koenders argue that all three recoveries from recessions that followed long expansions exhibit the pattern of a long delay in achieving employment. "This finding contrasts sharply with the characterization that it is the two most recent recoveries that are distinct," they write.

Rogerson says the research may offer some suggestions for policymakers.

"I think it represents some caution about interpreting economic cycles," Rogerson says. These cycles are relevant not only to the U.S. economy, even though the paper zeroed in on the United States. Rogerson acknowledges that the dynamics discussed would likely apply to any Western industrialized nation.

Certainly, the forces the paper talks about should be present in any such nation, although one factor, labor market institutions, might offer the potential for differences.

— Richard Rogerson, professor of economics

Rogerson says different labor market institutions may have inherently different influences about how firms make decisions, but notes this would be interesting to track in future papers.

The straightforward economic mechanism that Rogerson and Koenders argue may be relevant for understanding the different behaviors of the labor market across business cycles stresses two key effects. First, it says that "internal organizational dynamics" are affected by aggregate shocks. Second, it stresses that the situations of organizations affect "the manner in which the economy responds" to aggregate shocks.

"In periods of high economic activity, organizations postpone structural changes to take advantage of current opportunities. But once an organization begins the process of restructuring, it is less likely to hire workers and more likely to release workers. These effects suggest the possibility that long expansions will be followed by recoveries in which employment starts to increase much later than output," according to the paper.

The way the economy is generally understood to work is that during the business cycle, the movement within the labor market reflects either "cyclical" or "structural" effects. In a market-driven economy, one effect may be more dominant than the other, yet both are present. Cyclical changes in labor demand, for instance, are supposed to be temporary.

During a recession most companies would see their revenue reduced and profit margin shrunk during a recession. To minimize the losses, it is expected that firms will cut costs, which often means a reduction in personnel.

Once economic growth resumes, workers who became unemployed during the recession are recalled or perhaps find new jobs as firms gear up for more production. As well, output and employment growth tend to be most strong during the initial stages of an economic expansion.

The second type of labor market turnover widely understood to be accepted is what economists call "structural" job losses or gains. Structural unemployment happens when new technologies lead to new processes that make hands-on labor less relevant. Or, perhaps new types of goods and services replace existing products. Industries have always taken advantage of innovations that have reduced their demand for labor, with the result that fewer workers are needed to produce the same goods at the same output.

As an example, consider the decade 1992 to 2002. The number of motor vehicles produced in the U.S. grew from 9.7 million to 12.5 million, while the number of workers in the industry declined by about 5 percent at the same time, according to data from the Census Bureau.

Rogerson and Koenders acknowledge the economic theories on the business cycle that have been around since the work of Burns and Mitchell in the mid-1940s. Burns and Mitchell were among the first to measure the business cycle and argued that these cycles have striking similarities to each other in many ways.

But Rogerson says any theory that links the speed of the recovery in employment to the length of the preceding expansion "would seem to be a viable candidate for explaining the anomalous behavior of employment following the last two recessions."

Yet the paper points out that since the recession of 1969-1970 also followed an unusually long expansion, a slow recovery of employment should have been observed. The paper argues that when viewed from this new perspective, "the behavior of employment in the 1970 recovery is in fact very similar to the behavior of employment in the recoveries of 1991 and 2001" and yet is different from the behavior of employment in the other post-World War II recoveries.

Rogerson says although the research focuses on the implications for business cycles, it is believed the model "may also prove useful for examining plant and firm level dynamics more generally."

From a practical and helpful standpoint, the authors also point out a possible solution for organizations struggling with firing and hiring decisions. During periods when an organization finds itself with too many workers, but expects that over time these same workers will eventually be needed, it may be optimal to keep them on staff, particularly if they have valuable organizational capital. In this way, any decreases in employment can occur through attrition, which might produce a "pattern of relatively stable separation rates coupled with an extended period of very low hiring rates."

Rogerson and Koenders caution their work is suggestive. But Rogerson says he is enthused to think of this work being developed further, both by Koenders and others who will even more rigorously analyze the data and the mechanism they have developed.