U.S. Labor: A Not-So-Great Participation Rate

Measuring unemployment presents a policy and a forecasting problem of considerable concern. Americans appear to have become less inclined to work these days, especially since the Great Recession of 2008–09. The so-called participation rate—the portion of the population older than 16 who either are working or looking for work—after rising from 59% in the early 1960s to 67.5% by the turn of the century, had drifted back down to about 65.5% just before the 2008–09 recession, and since then has fallen precipitously, closing in on 63%. These changes may seem minor here on the page, but the difference from the peak participation rate at the turn of the century to the present rate amounts to some 6.7 million people.¹ Their behavior going forward could easily change perceptions of the economy’s strength, its reality, and, possibly, policy.

The degree of concern in this matter hinges on which explanation for the drop forecasters, analysts, and policymakers prefer. On one side are those who describe the fall-off in terms of discouraged job seekers. These analysts worry that the poor jobs market has convinced many to give up looking, but that they will flood back to the search as the labor market improves. Because these people will not find work immediately, the measured unemployment rate will rise, causing the Federal Reserve to regret any decision to raise interest rates in the interim. On the other side are those who consider the drop in participation as a reflection of demographic shifts, particularly the aging of the population. They see the situation as much more stable, consider the headline unemployment rate an accurate reflection of the economic fundamentals, and see less risk should the Fed go ahead with its plan to raise interest rates gradually and eventually shrink the central bank’s balance sheet.

In the thick of the debate over causes, and by implication how matters will work out going forward, is James Bullard, president of the Federal Reserve Bank of St. Louis. He has assembled considerable evidence to support the notion that the new, lower labor-force participation rate has less to do with frustrated job seekers and instead is a stable reflection of the changing age composition of the population. Through this line of analysis, Bullard argues that the growing proportion of older people, retirees, in the total population over the age of 16 quite naturally reduces measures of participation. He concludes from this that the situation is not likely to change anytime soon, at least not quickly, that today’s relatively low 5.5% rate of unemployment is an accurate reflection of the state of the labor market and is not at all liable to a sudden rise, and that, consequently, there is little risk to the Fed going ahead with its plan to raise interest rates gradually.²

But for all the academic and policy analysis Bullard has assembled, a significant measure of doubt remains. For one, Bullard has long stressed the need to raise rates—and for reasons well beyond questions of labor force participation. His policy preference leaves him with a strong desire to find relative stability in the participation situation. None of this is to suggest that such a fair-minded person would consciously tilt the evidence. Rather, it is a simple recognition that even the best of minds is vulnerable to biases. Second is the speed at which the participation rate has fallen during the great recession and since. Demographic influences trend to unfold gradually. A purely demographic influence might have allowed a gradual decline, such as occurred in the early years of this century. But the sudden drop since the Great Recession suggests that something else is also happening: that in addition to a growing proportion of retirees, a frustration with the job search has perhaps also induced people of prime working age to give up for the time being. Third, Bullard’s careful assembly of evidence also turned up studies that ascribe a considerable portion of the drop in participation since 2007 to cyclical instead of demographic considerations; some of these studies point to half of it, some to all of it.³

Since a truly balanced assessment of these positions leaves the field open to just about any conclusion, it is worth looking at matters through the lens of limiting assumptions. If reality is likely a combination of the demographic and cyclical explanations, then it is reasonable to work from the possibility that, say, half the drop in participation since 2007 is cyclical and so subject to a reversal should the jobs market improve further. That perspective would imply that a population of some 3.7 million frustrated job seekers could come back to the search. If all of them were suddenly to do that, the headline unemployment rate would climb some 2.5 percentage points, to 8.0%. But even if eventually this many people could plausibly return to the job search, they are unlikely to flood back to the market all at once. Practically, then, policymakers at worst would have to cope with perhaps a more manageable rise in the unemployment rate, to 6.5% of the workforce.⁴

This higher unemployment rate would certainly paint a less robust economic picture than many have now. It might even sway some on the Federal Open Market Committee, the body that actually sets monetary policy. The balance of influence might then slow the implementation of the planned interest-rate increases. But since the Fed considers more than just the unemployment rate when setting policy, and since such an unemployment rate is not much different from policymakers’ original target for a policy change, it is not likely to reverse the underlying decision to raise interest rates gradually or seek ways to trim down the Fed’s balance sheet. Still, it would create a lot of doubt among market participants and, with it, market volatility.