Thank you Baby Boomers: The new norm is 60,000 jobs

Mike Smitka, Economics, Washington and Lee University

A simple projection from pre-Great-Recession levels suggests we need 150,000 new jobs each month to keep up with population growth. (See this WSJ blog post.) This is wrong. Lower fertility 20 years ago means that we are not seeing a steady climb in the number of young people are entering the labor market, while we are seeing the baby boomers gradually exit the labor force. My bottom line: 58,390 jobs a month. Now such precision is illusory, labor market data aren't that precise, and these are projections. So let's use 60,000 as a rough guide.

The implications are significant. If we use the sort of simple projection that lies behind the WSJ projection, then (using 2007 as a base), then in a normal economy we would today have a labor force of 160 million million. With an actual labor force in the latest survey of 151 million, this would indicate that we are 9 million jobs short of where we should be, and thus still deep in recession. That's clearly not the case. Across a variety of metrics we're approaching "full employment" (which ought not be pinned to a specific number as measured using one particular data set). Jobs are easier to find, even if wages are not yet rising. To give one qualitative indicator, during the depths of the recession in 2009-2010 "quits" in the JOLTS survey (Job Openings and Labor Turnover Survey) were at 1.3%, the lowest in the history of the data. The level is now 2.0%, near the pre-recession average of 2.2%.

Why 60,000? Prior to 2007, the labor force participation rate of prime-aged workers was nearly constant. So the core of my projection was to use the Census Bureau population projections by age, and multiply by this "normal" level of participation to get the number of jobs. I did the same for younger ages (16-24, where in the Great Recession participation fell by far more than other groups), and older ages (where participation actually rose – the "Boomers" have been slower to retire than their elders). I've not redone this estimate to use the most recent population projection, but today's the last day of the W&L Winter Term, so I won't do that now. When I do, I'll provide an upper and lower bound, but I don't anticipate much change.

So how do things stack up? I provide below my own projections of "normal" employment versus where's we're at in the March 2016 data. To that I add a graph of age-specific participation levels. (Data are available to do that by gender; I don't report that here.) The bottom line is that at the current rate of job creation, we'll be back to normal in about 2 years.

One apology: at the bottom of the WSJ post they do quote FRB Chair Janet Yellen that the number is now under 100,000 jobs a month. You have to read to the bottom, however, to see that. For regular commentary on labor force development, see the Atlanta Fed's blog.

Indeed, the Atlanta Fed's latest post notes these various factors, including the one I've not built into my formal projections: the increased participation of older, post-normal-retirement, Americans. Their estimate is that over 2007-2011 increased participation fully offset the onset of "boomer" retirement. (Click HERE to see their graph.) Now that's not the case the past 5 years, but I still ought to revisit my baseline and put in adjustments for this, which shows up so vividly in the age-specific participation rate graphs I generate.