Over the last several years, the economy has continued on a slow-but-steady march to full employment. Along with improvements in nominal wage growth, we’ve seen evidence that more and more sidelined workers continue to pour into the labor market, seeking work and getting jobs. This growing labor force participation rate (LFPR), which has beaten many experts’ more pessimistic projections, is the subject of this jobs day preview post.

Projections of labor force participation changed dramatically once the Great Recession hit and many experts quickly decided that cyclical drop-offs in participation were actually structural trends. Think of cyclical changes as being short term, driven by the aggregate demand shortfall that caused the Great Recession and its aftermath. Structural changes are due to long-run trends, such as the aging of the workforce or the retirement of baby boomers. In and immediately following the Great Recession, there was a steady and deep decline in labor force participation. Even after the unemployment rate began to recover after a sharp spike, the participation rate continued to decline. That relationship is clearest when you look at the prime-age population, as I’ve pointed out before, but is true when you look at overall labor force participation and unemployment as well.

The figure below shows the relationship between the unemployment rate and the labor force participation rate between 1989 and 2019. It’s clear that the labor force participation rate continued to decline even as the unemployment rate started to recover in the aftermath of the Great Recession. Remember that to be counted as unemployed, you must be actively looking for work in the four weeks prior. With so many would-be workers falling off the official count of the unemployed, because the weak economy meant they did not believe there were job opportunities for them, many analysts began to question whether they would ever return.

Back in 2007, before LFPR trends could be contaminated by the cyclical drag of the Great Recession, the Bureau of Labor Statistics (BLS) projected that the participation rate would be much higher than was later projected once the effects of the Great Recession began being absorbed into these forecasts. For instance, in 2007, BLS projected LFPR to be 65.5 percent in 2016, but that projection fell to 63.0 percent in the forecast they made in 2013. What’s relevant here is that the 2013 BLS projections implied that the decline was permanent, and would not expand with an improving labor market—and would continue to steadily decline from that point on, falling to 61.6 percent by 2022. So-called structural changes, which were due to foreseeable demographic factors back in 2007, now took on even more meaning. BLS doesn’t try to parse out cyclical from structural factors in its forecasts, instead they just push forward recent trends. This means their forecasts mechanically absorb cyclical effects and allow these effects to depress LFPR projections even a decade in the future.

The figure below shows three different sets of projections that make the point that cyclical factors were being deemed as structural. The first three bars in each year show labor force participation projections from a much-cited Brookings Paper on Economic Activity in 2014, the Congressional Budget Office (CBO), and BLS from 2014 through 2019. The 2019 year presented here represents a linear extrapolation between each of their projections between 2018 and 2019, essentially getting us one-quarter of the way to the full year 2019, thereby representing an estimate of the first quarter of 2019 to compare to actual data. The fourth bar in each year shows the labor force participation as estimated by the Current Population Survey in each month.

What’s clear from the figure is that far more workers were temporarily, not permanently as many experts claimed, sidelined by the Great Recession. At EPI, we’ve long thought that a significant portion of the drop in the labor force participation rate since the start of the recession was due to weak job prospects, not structural factors. Further, counts of these so-called missing workers might have been understated because labor force participation was also weak in the decade leading up to the Great Recession.

Employers have long since known that those missing workers were a captive audience for jobs, which let them keep wage growth low for the current workforce with the threat of replacing incumbent workers with those on the bench. Every month, workers are getting jobs, and a historically large share of them are returning to work after having been out of the labor force the month prior, as opposed to officially counted among the unemployed. I’ve thought the unemployment rate has been overstating the strength of the labor market for some time and can explain why it has taken so long for the seemingly low unemployment rate to translate into stronger wage growth. What I hope to see on Friday when the latest jobs numbers come out is a continued rise in participation as promising job opportunities present themselves to the latest job seekers and temporarily sidelined workers.