Labor Force Participation and the Unemployment Threshold

On Friday, my colleague Julie Hotchkiss shared in this space the results of her new research (with Fernando Rios-Avila, a Georgia State University colleague) on the recent and prospective behavior of the labor force participation rate (LFPR). The punch line, from my point of view, is this:

Our results suggest that relative to the average LFPR over the years 2010–12, the average LFPR over the years 2015–17 will rise by about a third of a percentage point—again, if the labor market returns to prerecession conditions. (Italics original)

As Julie notes:

[T]he Federal Open Market Committee has substantially raised the stakes on disentangling…movements in labor force participation…by introducing into its policy deliberations concepts like unemployment thresholds and qualitative assessments on “substantial” labor market improvement.

Though the meaning of “substantial labor market improvement”—a condition for adjusting the FOMC’s current large-scale asset purchase program—is somewhat ambiguous, the unemployment threshold for considering moving the federal funds rate off the near-zero mark is less so. As the Committee indicated in its May press release:

[T]he Committee decided to keep the target range for the federal funds rate at 0 to ¼ percent and currently anticipates that this exceptionally low range for the federal funds rate will be appropriate at least as long as the unemployment rate remains above 6½ percent, inflation between one and two years ahead is projected to be no more than a half percentage point above the Committee’s 2 percent longer-run goal, and longer-term inflation expectations continue to be well anchored.

It is widely understood (a sign of the times, no doubt) that changes in the unemployment rate are not entirely independent of what is happening with the participation rate. We have discussed this issue before here in macroblog. But in light of the new research coming from our own shop (and other research cited in Julie’s post), it seems like a good time for a refresher.

First, a step back. Multiple upward revisions to the employment situation since the December jobs report—you can follow the trail courtesy of Calculated Risk herehereherehere, and here—have led to a more robust picture of the labor market than certainly I was thinking. Here is what the record looks like for most of the recovery:

With an assist from the Atlanta Fed Jobs Calculator, we can provide further perspective on these numbers. In particular, under the assumption that the labor force participation rate will remain at its current level of 63.3 percent (among other things held constant), we can map the recent job growth numbers to a rough date when the unemployment rate will reach 6½ percent.

That looks interesting, but then taking the Hotchkiss and Rios-Avila research onboard means the assumption of a constant labor force participation rate may not be justified. So, turning again to the Jobs Calculator, the following table answers this question: If we continue on the 208,000-per-month pace of job creation of the last six months, and the labor force participation rate is X, what would the unemployment rate be by June of next year? For reference, the first row of the table replicates the earlier result under the assumption that the participation rate will maintain its current level; the second row takes into account the Hotchkiss and Rios-Avila research; and the third assumes an even larger bounce back in participation:

It is probably worth noting that the full increase in the Hotchkiss and Rios-Avila estimates happens in the 2015–17 timeframe, raising the interesting possibility that the threshold for considering interest rate increases could occur sometime before the unemployment rate moves back above the threshold.

Also, it is not at all obvious that rising labor force participation would necessarily arrive along with a rising unemployment rate. From 1996 through 1999, for example, the participation rate rose by nearly by 0.7 percentage point (the difference between the rates in the first and third rows in the table above), even as the unemployment rate fell by just over 1½ percentage points. The key was the strong employment growth over that period—almost 260,000 payroll jobs per month on average.

All of that, as should be clear by now, embeds a whole bunch of assumptions, which may make this the most important part of the FOMC’s decision criteria:

In determining how long to maintain a highly accommodative stance of monetary policy, the Committee will also consider other information, including additional measures of labor market conditions…

This piece is cross-posted from Macroblog with permission.

One Response to "Labor Force Participation and the Unemployment Threshold"

  1. benleet   May 16, 2013 at 9:54 pm

    Heidi Shierholz at Economic Policy Institute Blog has posted short articles on this. The word "participation" is misleading, better is "employer hiring attraction". I think you are way off. The Labor Dept's Monthly Review also has had articles discribing the dynamics of falling "participation" due to generation "baby boomer" dropout rate.
    My blog says,
    "The labor participation rate is at a 34 year low; not since May 1979 has it registered at 63.3%. The U3 unemployment rate stands officially at 7.6%, but arguably it is 12.7%, one in eight workers — a truer indication of labor market distress. (I updated this essay to April 2013, since last month.) Heidi Shierholz at wrote that it could be 9.8% using a hypothetical participation rate estimated by the CBO. I think the CBO estimated rate is too low. The historical average labor participation rate over the 20 year period 1988 to 2008 looks like around 66.3%. If we use the 20 year average labor participation rate, then the U3 for today at that rate would be 11.7%, not 9.8%, not 7.6%. If today we had a participation rate of 67.1%, the rate of 1997, '98, '99, and 2000, then another 8.756 million workers would be participating, and all of them would be out of work, raising the unemployment rate from 7.6% to 12.7%, an increase of 5.1%.
    In addition, the employment to population ratio has not hit such a low since 1983, 30 years ago."