OLMIS

by Phoebe Colman
Published Oct-27-2011

Job Flow Trends in Recession and Recovery The View From the Fourth Quarter The Long View: Job Flow Dynamics and Economic Change

Business Employment Dynamics (BED) data enhance understanding of labor market changes by revealing information about the flow of jobs in Oregon's economy from quarter to quarter. The publication of data for the fourth quarter of 2010 makes it possible to analyze job flow patterns before, during and after the most recent recession.

BED data provide information that can help explain variations in net job change. The low point of the 2008 recession in terms of net job growth came in March 2009 when the Oregon economy lost 47,524 jobs. This loss resulted from a combination of high gross job losses and low gross job gains. In other words, the distance between gains and losses reached a maximum at this point (Graph 1).

In the remaining quarters of 2009, increasing job gains and decreasing job losses combined to close this distance. By March 2010, gross job gains and gross job losses were nearly equal, with gains slightly higher. This net job gain of 63 jobs in the first quarter of 2010 was the first positive net change since the recession began. Over the next two quarters the improving trend continued, peaking with a net gain of 11,368 jobs in the third quarter of 2010.

BED data for the fourth quarter of 2010 show a decline in net job growth from the previous quarter. Although net growth is still positive, it has decreased from 11,368 jobs gained to 2,457 jobs gained. The reason for this decreased growth is that gross job gains increased only slightly in the fourth quarter. At the same time, gross job losses which had been steadily decreasing in previous quarters, increased from 82,215 to 91,616.

Comparing BED data for Oregon with national data for the fourth quarter of 2010 reveals a reversal of the trends that emerged in the third quarter. While the net job growth in Oregon decreased from September to December 2010, the net job growth for the U.S. as a whole increased from 151,000 to 563,000. This growth was the result of a gross job gain of almost 7.0 million jobs combined with a gross job loss of just under 6.4 million jobs (Table 1).

How do seasonally adjusted BED data from the fourth quarter of 2010 compare with gains and losses observed at the same point in previous years? While trends for Oregon and the nation have fluctuated somewhat between improvement and decline in recent quarters, a review of fourth quarter data provides a different perspective on how the job market has fared since the recession began.

Data from the fourth quarter of 2007 reveal that from September to December of that year, gross job gains slightly exceeded gross job losses for both Oregon and the nation, producing a net job growth of 0.2 percent of total employment in both cases.

One year after the recession began, the situation was markedly different. In December 2008, gross job losses far outpaced gross job gains both for Oregon and the U.S. (Graph 2). In Oregon, the net job loss of 2.8 percent corresponded to a loss of 40,168 jobs, the second-largest quarterly net loss for Oregon in the BED data series. The largest occurred in the following quarter, when the Oregon economy shed 47,524 jobs, or 3.4 percent of private-sector employment. The nation as a whole fared somewhat better than Oregon in terms of gross job losses, with a net loss of 1.6 percent representing 1.8 million jobs nationwide in the fourth quarter of 2008. Parallel to the situation in Oregon, this net loss was second only to the one reported one quarter later in March 2009.

Rising out of the trough of late 2008 and early 2009, the job flow picture improved somewhat in December 2009. Both Oregon and the nation registered a net job loss in this quarter, but in both cases the loss was much smaller than that of the previous December. Oregon's net loss of 7,577 jobs corresponded to 0.6 percent of total employment. Gross job gains amounting to 6.8 percent of employment were offset by gross job losses of 7.4 percent. The national picture was slightly better, with a gross job loss of 6.5 percent of employment just slightly surpassing a gross job gain of 6.4 percent. The U.S. net job loss of 0.1 percent of employment represented 228,000 jobs, the smallest net loss since the recession began.

From this perspective, the fourth quarter of 2010 was the best in four years for both Oregon and the U.S. In both cases, seasonally adjusted gross job gains outpaced seasonally adjusted gross job losses, resulting in net job gains. Although Oregon's net job gain of 0.3 percent of employment in December 2010 was smaller than the 0.8 percent net gain observed in September 2010, it is a significant improvement compared to fourth quarter data from 2008 and 2009 and is in fact 0.1 percentage point higher than the net gain of December 2007. Similarly, the U.S. net job gain of 0.5 percent of total employment in December 2010 is a clear improvement over the three previous Decembers.

Examining trends in the average size of job gains and losses is another way BED data can be used to study labor market dynamics. One notable trend that emerges from gross job gain and gross job loss data is an apparent moderation in gross job flows over the span of the data series.

Gross job reallocation describes the total amount of "job churn" in the economy, and is equal to the sum of gross job gains and gross job losses in a given quarter. This can also be expressed as a job reallocation rate, which is the sum of gross job gains and losses as a percent of employment in a given quarter. In Oregon, the job reallocation rate has decreased markedly over the span of the BED data series (Graph 3). Even the 15 percent job reallocation rate at the height of the 2008 recession was well below levels that had been normal prior to the 2001 recession. By comparison, the average job reallocation rate between March 1993 and the onset of the March 2001 recession was 17.5 percent.

This decrease in gross job reallocation does not correspond to a decrease in the numbers of Oregon establishments gaining or losing jobs. BED data show that these numbers have generally trended upward during periods of growth, with the number of establishments gaining jobs decreasing during recessions.

If the number of establishments gaining and losing jobs is not decreasing, what is causing the downward trend in job reallocation? BED data suggest that this moderation is related to decreases in the average size of gain for establishments gaining jobs, and the average size of loss for establishments losing jobs. In Oregon prior to the 2001 recession, the average job gain was between 3.8 and 4.6 jobs per establishment, while the average job loss was between 3.8 and 4.4 jobs per establishment (Graph 4). Between the 2001 and 2008 recessions these numbers decreased along with the job reallocation rate. The largest average loss during the 2008 recession was 3.6 jobs per establishment, a smaller average loss than in any quarter prior to the 2001 recession. In each quarter of 2010, average job gains and losses have been between 3.0 and 3.4 jobs per establishment - the lowest in the BED data series.

Observing similar trends in BED data at the national level, some researchers have suggested a link between this apparent shift in job flow behavior and structural changes in the labor market. For example, the use of temporary layoffs and temporary help in response to structural shocks may at least partially explain this shift, as noted in the Federal Reserve Bank of Philadelphia's August 2011 working paper titled Job Flows, Jobless Recoveries, and the Great Moderation.

Data from Oregon's unemployment insurance administrative records appear to support the suggestion that a change has taken place in the way temporary help services are used. In 1988, jobs in the help supply services industry including temporary help services, accounted for about 1.3 percent of total private employment. By 2000 this share had more than doubled, rising to 3.2 percent. Beginning in 2001, temporary help services are classified as a separate industry; data for 2001 forward are not directly comparable with prior years. However, it can be noted that between 2001 and 2010 the temporary help services industry accounted for an average of 2.4 percent of total private employment, reaching a maximum of 2.7 percent in 2006 and a minimum of 1.8 percent in 2009, during the trough of the last recession (Table 2).

Using the evidence from these two data series, it is at least possible to say that the help supply services industry (including temporary help) grew significantly in the decade prior to the 2001 recession, and that the temporary help services industry accounted for a relatively stable share of total private employment between 2001 and the onset of the 2008 recession. This indicates that Oregon's temporary help industry could be a factor in moderating gross job flows.

Data on the quarterly gross job gains and losses come from the Business Employment Dynamics program at the U.S.  Bureau of Labor Statistics. A more detailed Business Employment Dynamics Report is available at www.QualityInfo.org on the Publications page in the News box.