A lot has been written about the long-run decline in the labor force participation (LFP) rate among prime-age men (usually defined as men between 25 and 54 years of age). For example, see here, here, here, and here for some perspectives.
On a not seasonally adjusted basis, the Bureau of Labor Statistics estimates that the LFP rate among prime-age males is down from 90.9 percent in the second quarter of 2007 to 88.6 percent in the second quarter of 2016—a decline of 2.3 percentage points, or around 1.4 million potential workers.
Many explanations reflecting preexisting structural trends have been posited for this decline. But how much of the decline also reflects cyclical effects and, in particular, cyclical effects that take a while to play out? We don't really know for sure. But one potentially useful approach is to look at the Census Bureau's Current Population Survey and the reasons people give for not wanting a job. These reasons include enrollment in an educational program (especially prevalent among young individuals), family or household responsibilities (especially among prime-age women), retirement (especially among older individuals), and poor health or disability (widespread). In addition, there are people of all ages who say they want a job but are not counted as unemployed. For example, they aren't currently available to work or haven't looked for work recently because they are discouraged about their job prospects.
To get some idea of the relative importance of these factors, the following chart shows how much each nonparticipation reason accounted for the total change in the LFP rate among prime-age males between 2012 and 2014 and between 2014 and 2016. The black bars show each period's total change in the LFP rate. The green bars are changes that helped push participation higher than it otherwise would have been, and the orange bars are changes that helped hold participation lower than it otherwise would have been.
A note on the chart: To construct the contributions derived from changes in nonparticipation rates, I held constant the age-specific population shares in the base period (2012 and 2014, respectively) in order to separate the effect of changes in nonparticipation from shifts in the age distribution.
Notice that the decline in the prime-age male LFP rate between 2012 and 2014 has essentially fully reversed itself over the last two years (from a decline of 0.53 percentage points to an increase of 0.55 percentage points, respectively). The positive "want a job" contribution in both periods clearly reflects a cyclical recovery in labor market conditions. But the most striking change between 2012–14 and 2014–16 is the complete reversal of the large drag attributable to poor health and disability. Other things equal, if nonparticipation resulting from poor health and disability had stayed at its 2012 level, prime-age male participation in 2014 would have only declined 0.10 percentage points. If nonparticipation due to poor health and disability had stayed at its 2014 level, prime-age male participation in 2016 would have increased only 0.14 percentage points.
The incidence of self-reported nonparticipation among prime-age men because of poor health or disability has been declining recently. According to the Current Population Survey data, this reason represented 5.4 percent of the prime-age male population in the second quarter of 2016. Although this is still 0.7 percentage points higher than in 2007, it is 0.3 percentage points lower than in 2014. Some of this turnaround could be the result of changes in the composition of the prime-age population. But not much. Around 90 percent of the LFP rate change because of poor health and disability is due to age-specific nonparticipation rather than shifts in the age distribution, suggesting that some of the turnaround in the incidence of people saying they are "too sick" to work is a cyclical response to strengthening labor market conditions. We've yet to see how much longer this turnaround could continue, but it's an encouraging development.
For those interested in exploring the contributions to the changes in the LFP rate by gender and age over different time periods, we're currently developing an interactive tool for the Atlanta Fed's website—stay tuned!