Unfortunately, analysis is a slow process as you need to test your observations, and many initial observations are not historically relevant. A relatively insignificant data release – Job Openings and Labor Turnover Survey (JOLTS) for April 2011 – from the Bureau of Labor Statistics is a case in point.
Econintersect spends significant effort to understand employment dynamics, and produces an employment index as part of its monthly economic forecast. Although JOLTS itself is not used to develop this forecast as the data is released over a month after period close – it does provide a rear review reality check to our forecast.
Facts we know are true is that employment went bad (very much less good) in May 2011 (analysis here). Was this less good employment number an aberration, or a new trend? Is there anything in the April JOLTS data which offers an indication or warning?
The green shoot in the JOLTS data has been – and continues to be Jobs Openings. Since the end of the Great Recession of 2007, jobs openings have been gently improving.
The number of job openings in April was 3.0 million, little changed from 3.1 million in March. After increasing in February, job openings have been flat. Job openings have been around 3.0 million for three consecutive months; the last three-month period with levels this high was September—November 2008. The number of job openings was 549,000 higher than at the end of the recession in June 2009 but remains well below the 4.4 million openings when the recession began in December 2007.
Separations (quits, fires, and retirements) have been trending down.
The total separations, or turnover, rate was unchanged at 2.9 percent for total nonfarm in April on a seasonally adjusted basis. Over the year, the total separations rate (not seasonally adjusted) was essentially unchanged for total nonfarm, total private, and government.
Yet hires have been in the same range since mid 2010.
In April, the hires rate was essentially unchanged at 3.0 percent for total nonfarm. The hires rate also was essentially unchanged for all industries and regions. At 4.0 million in April, the number of hires has increased from 3.6 million in October 2009 (the series trough) but remains below the 5.0 million hires in December 2007 when the recession began.
It should be obvious even to the most casual observer that hires are not equal to jobs openings less separations. There are many more elements to jobs creation.
Job Openings may be the canary in the coal mine.
It appears year-over-year jobs opening growth trends lead the trends in employment growth trends by approximately a half a year. About six months ago the trend lines for job openings started a less good trend line.
This is a simple observation. Unfortunately, the JOLTS database is relatively new – and testing any theory requires more than one economic cycle. So, for now, this remains an observation.
Econintersect has identified specific economic pressures that generally track employment gains. These pressures are forecasting less good employment data. However, there is a concern that recent price inflation is also effecting employment – and inflation is not included in Econintersect’s employment model.
ADP Employment Growth Crashes: Our Assessment by Steven Hansen
Is the Stimulus Effecting Unemployment Claims? by Steven Hansen
BLS: Jobs Up Strongly in April 2011 by Steven Hansen
College Training for Unemployment by Mike Konczal
Economic Damage Storm Track of The Great Recession by Ted Kavadas