A Tale of Two Recessions:
Part 2 Hires
Quits and Hires during the 2001 and 2007-2009 Recessions
One of the more useful and interesting data series published by the US Department of Labor, Bureau of Labor Statistics (BLS) is the Job Openings and Labor Turnover data series or JOLTS. Data from JOLTS produces some very useful data for gauging the dynamics and economic wellbeing of the national and regional economies. These data are: job openings, hires and total separations. Separations include quits, layoffs/discharges and other separations. The BLS also publishes rates for all these JOLTS measures (for example, the number of quits per 100 jobs). This two part study looks only at the rates for some of these measures (quits and hires). These measures will be compared to the annual percent changes (measure each month) in the total number of jobs from the Current Employment Statistics program. Using this measure will provide an idea of how the jobs numbers change through time and through two recessions and one full recovery cycle. A very detailed study of JOLTS data was published by Klemmer and Lazaneo in the BLS Monthly Labor Review, August 2010. Part 2 is an analysis of hiring since 2001 as measured by the hires rate (the number of hires per 100 jobs).1
Hires Rates and the Recessions
The last two recessions (March 2001-November 2001, and December 2007-June 2009) showed somewhat different worker/employer behaviors for the rate of quits and the rate of hires from JOLTS data. Hiring occurs when employers perceive that they have enough additional business to require more workers and they can add employees to their payrolls and remain profitable. Typically employers will increase hours for incumbent workers during the earlier stages of recoveries and then add more workers to their payrolls later. This reduces the risks of adding more workers when the economy may weaken. Hiring rates are very sensitive to changes in employer hiring practices and relative small changes in the rate can indicate big changes in the labor market. Hiring rates in this entire study vary from slightly over 4.0 per 100 workers to slightly below 3.0 hires per 100 workers with an average of 3.54 over the 2001-early 2012 timeframe. Apparently, a certain amount of hiring must occur just to keep the economy running.
The big wrinkle in the current hiring situation is that employers are not optimistic about the future and reluctant to add more workers. Another factor contributing to slow hiring is that current workers are so productive that hiring is not needed. The BLS Index of Labor Productivity for business (output per hour) has increased from 88.2 in 2001 to 110.13 in 2011 (2005=100). Productivity varies greatly by industry but overall the nation’s workers are much more productive now than in 2001.
As mentioned, adding hours is another employer response to early recovery scenarios. The Index of Hours from the same series follows the typical recovery scenario, the index increased from 93.3 in 2009 to 96.6 in first Qtr. 2012 (2005=100).2 More productive workers are working more hours, requiring less hiring by employers.
Returning to hires data, Chart 2 shows the hires rate 2001-current on the left axis and the annual percent change (by month) in the number of jobs from the Current Employment Statistics series on the right axis. Of note are the timing and level of the changes in these two measures.
Chart 2 Hire Rate and Annual Percent Change in Jobs
Data Source: Bureau of Labor Statistics, prepared by Brandt Information Services, Inc.
Note: Grey areas are recessions
The 2001 Recession
In January 2001, the hires rate was 4.4 per 100 workers, the highest figure in this analysis. Changes in hiring proceeds changes in employment and the hire rate dropped inconsistently throughout 2001 and finished the year at 3.6. The rate of annual job growth was 1.3 in January 2001, and had declined to 0.9 percent by March 2001, the official start month of the recession of that year. The hires rate reached bottom in December 2001 at 3.6 per 100, a little ahead of the low point in job growth (February 2002, -1.6 percent decline). Following the official end of the recession (November 2001), the hires rate kept in the 3.4-3.9 range with some low points in early 2003 while jobs declined fairly consistently starting in July 2001. Annual percent change in the number of jobs remained negative for almost 30 months, a prelude to the jobless recovery in the next, and much worse recession.
The hires rate mirrored the doldrums in job growth and did not started increasing until mid-2003, and being a good leading indicator signaled the start of actual job growth in December 2003.
Recovery from the 2001 Recession
Throughout the recovery from the 2001 recession, both series performed as expected–the hires rate always leading the way to indicate better economic times. The hires rate peaked at 4.1 in August-September 2005, and seven months later the annual rate of job expansion reached its recovery peak of 2.2 percent in April 2006. Over the next eight months, the annual rate of job expansion eroded to 1.5 percent while the hires rate bounced around happily between 3.8 and 4.1 hires per 100 jobs. The average hires rate during the entire recovery was around 3.8 hires per 100 jobs. The annual change in the number of jobs continued to decline as recessionary forces built up and by August 2007 had slipped to 0.9 percent. The hires rated had already slipped slightly to 3.7 hires per 100, which is below the average rate during the recovery (3.8). Small changes in JOLTS data are harbingers of worse times too, and the annual rate of job growth plummeted rapidly and by April 2008 was only +0.1 percent, dropping to -0.1 percent in May. The recession had already started in December of the prior year according to the National Bureau of Economic Research.
Hires in the Great Recession
Chart 2 illustrates why JOLTS data are so useful for economic analysis. Well before the official start of the recession and several months before the bottom fell from under job growth, JOLTS hires rates had started to deteriorate. The first downward movement was from May-July 2007 (3.9-3.7) followed by a jagged but inexorably downward trend to 2.8 per 100 jobs by June 2009. Hires rates tend to change slowly and plateau and stick to one value during the different phases of business cycles. The slide in the hires rate (peak to trough) during the prior downturn took almost 40 months; this decline happened in 32 months and was steeper. Following the JOLTS data, the annual change in the number of jobs declined precipitously to -5.0 percent in both July and August of 2009. The slippery slopes of this recovery have not allowed the hires rate to gain traction and move up, at least during the early stages of this very weak economic rebound.
Hires and the Recovery
While the annual rate of post- Great Recession job expansion peaked at 1.6 percent in February 2012 and recently declined, the hires rate is stuck at 3.3 per 100 jobs as of May 2012. Of note about the hires rate is how it compares to the historical data and the data during the depths of the Great Recession. Recall that the hires rate peaked at 4.1 several times during the prior recovery, but averaged 3.8 per 100 jobs during the entire recovery, spanning over 6 years. The current hires rate of 3.3 per 100 jobs matches October 2008 levels of hiring, during the depths of the recession. Changes in the hires rate have stalled, and the rate has been either 3.2 or 3.3 for several months. Since the number of jobs changes upward only after the hires rate increases, we can expect more disappointing news about growth in the number of jobs in the coming months. If the hires rate upticks even slightly to 3.4 or 3.5 we can expect increasing job counts and some impetus to the recovery.
1 Typical BLS practice is to refer to hires and quits rates as percent, this would be confusing in a narrative with so many percent changes for the CES (jobs) series, so this narrative will use hires (or quits) per 100 jobs. In BLS language: The hires (or quits) rate is computed by dividing the number of hires (or quits) by employment and multiplying that quotient by 100.
2 Source: BLS Website, series index PRS84006033, Major Sector Productivity and Costs, Business Sector
Notes: The hires rate is used in this analysis since it concentrates more on the dynamics of hiring rather than the levels. The hires rate published in April of 2012 (3.1) translates into 4,175,000 hires. http://www.bls.gov/news.release/jolts.nr0.htm has the BLS news release on Job Openings and Labor Turnover Summary.
All recessions and their start and end dates are determined by the National Bureau of Economic Research (NBER).