Although the US is not officially in recovery, data from the Bureau of Labor Statistics shows that the US labor market is poised for a turnaround.
After 18 months of consistent downward trending, the number of people being hired in the US economy increased by 140,000 in July, according the Bureau of Labor Statistics' Job Openings and Labor Turnover Survey (JOLTS). This corresponds to the July increase in Hiring Demand we reported on this blog.
Source: Bureau of Labor Statistics - JOLTS
Most economists focus their gaze on the monthly releases of 'Nonfarm Payroll Employment' numbers. Month-over-month changes in the level of US employment drives the economic outlook. The Department of Labor's release of Unemployment Insurance Claims is also useful to economists – that data is released on a weekly basis and does not suffer from any 'reporting lag', which means it's as close to a 'real-time' monitor of labor market movements as we'll get.
The Bureau of Labor Statistics' Job Opening and Labor Turnover Survey (JOLTS), using a different methodology, details the underlying dynamics of the US labor market and can provide a layer of detail not available in either Nonfarm Employment counts or Unemployment Insurance Claims reports.
Even though the month-over-month changes in the level of US employment are the primary metric employed by economists, those results are in fact driven by two underlying series: the number of Hires and Separations.
Remember our "Job Accounting Identity":
Change in Employment = Hires – Separations
If more people are being hired than fired, employment rises. If more people are being fired than hired, employment shrinks.
The time series graph above shows the evolution of both the number of Hires and Separations in the US economy, measured in thousands and on a seasonally adjusted basis. In 2006 and 2007, Hires outnumbered Separations and employment grew. In January 2008, at the beginning of the current recession – which we've marked with a vertical line – the two series crossed and more people were exiting their jobs than being hired. Notice how the distance between the two lines is the greatest in January 2009, when the BLS reported a loss of 741,000 workers, the worst hit taken during the current recession.
Note that changes in employment are the distance between the two lines. The number of Separations has been trending "downwards" on an absolute basis because Separations are in fact usually measured as a percent of the total labor force. The economy has lost 7 million jobs since the start of the recession, and so fewer people need to be fired on an absolute basis to keep firing at a fixed rate relative to the size of the overall labor pool.
When the number of Hires "crosses" the number of Separations on the way up, the US economy will be able to generate gains in employment. We'll be able to identify an "unofficial start" to the recovery when the number of Hires maintains its upwards trend. When Hires is able to maintain its upward trend, employment gains will be a matter of time, and we'll be able to go from "slowing losses" to "gains".
How soon is this going to happen? The number of online job ads has been stable and rising slightly since March. That trend continued in August. Reporting lags in the JOLTS data series (two months) means that we can continue to expect improving results there. Month-over-month drops in US employment levels have been subsiding, and UI claims reports have been positive for the past two months. If the current "weak recovery trend" continues, the recession may come to an end as soon as January.
All Entries