Real people looking for jobs.

Photographer: Mark Kauzlarich/Bloomberg

So You Say the Unemployment Rate Is 'Fictitious' ...

Justin Fox is a Bloomberg View columnist. He was the editorial director of Harvard Business Review and wrote for Time, Fortune and American Banker. He is the author of “The Myth of the Rational Market.”
Read More.
a | A

Last year, Gary Cohn, the president and chief operating officer of Goldman Sachs Group Inc. whom President-elect Donald Trump has chosen to be his top economic policy adviser, declared that the official U.S. unemployment rate -- 5.5 percent at the time he spoke -- was "very, very fictitious."

Cohn was speaking at the Jack Welch College of Business at Sacred Heart University in Connecticut. Welch, the former General Electric Co. chief executive officer, is the fellow who made the infamous -- and unsubstantiated -- accusation in 2012 that "the Chicago guys" were messing with the unemployment numbers to help Barack Obama get re-elected. Cohn was suggesting something less sinister -- that the unemployment rate was "only that low because the participation rate has gone downward." Only those who are actively looking for jobs count as unemployed; labor-force dropouts don't. Explained Cohn:

The participation rate really measures people out in the U.S. population that are looking for jobs. There are so many people who are frustrated looking for jobs that they’ve just stopped. If the participation rate normalized -- this is a fun fact -- if it normalized to Day 1 of the Obama administration, we’d still be at an 11 percent plus unemployment rate.

I tried replicating that calculation using the most recent employment numbers that would have been available when Cohn spoke, those of March 2015, and came up with a 9.7 percent unemployment rate (it would be 9 percent now). The adjusted rate had last been above 11 percent in October 2013. So Cohn's number was out of date, although not wildly far off.

Is this really a better way of calculating the unemployment rate, though? Not exactly! The denominator used in calculating the labor-force participation rate is the entire civilian, non-institutional population age 16 and older. So as the giant baby boom generation drifts into retirement -- the first baby boomers began turning 65 in 2011 -- that puts serious downward pressure on labor-force participation regardless of how the economy is doing. You can see evidence of this effect if you compare the trajectory of the overall labor-force participation rate with the prime-age rate, which covers those age 25 through 54.

Labor-Force Participation
As a percentage of civilian, non-institutional population
 
Source: Bureau of Labor Statistics
ATTENTION! Y-axis does not go to zero!

The overall labor-force participation rate has dropped more, and made less of a rebound since 2015, than the prime-age version. Prime-age participation is now only 1.4 percentage points lower than it was in January 2009, compared with a three percentage point drop in the overall rate. Not all of this difference can be attributed to retiring boomers -- young people have been staying in school longer, too, in part because they haven't been too excited about their job prospects. But the overall picture this gives is of a markedly healthier labor market than the one Cohn described.

If you're not just out to bash the Obama administration and you're willing to look further back than 2009, though, you can find evidence of lots of non-retirement-age Americans giving up on the labor market. Here's prime-age labor-force participation, broken down by gender, going back to 1950:

Men and Women in the Labor Force
Prime-age (25 through 54) labor-force participation rate
 
Source: Bureau of Labor Statistics

Prime-age men have been dropping out of the labor force since the 1950s. Women, after making big increases in labor-force participation from the 1950s through the 1990s, have done some dropping out of their own since 2000. As a result, the fact that the headline unemployment rate was the same this November (4.6 percent) as in February 1998, June 1965 and March 1955 does not mean that today's labor market conditions are really comparable to conditions back in 1998, 1965 or 1955. Something has been happening to push (or pull) men in particular out of the labor market. The unemployment rate is designed, and has been since its very beginnings, to ignore those dropouts. But their numbers have grown so large that ignoring them paints a really misleading -- maybe even fictitious -- picture of the employment situation

But the attempts to correct for this misleadingness are often at least as misleading as the headline unemployment rate. Cohn's "11 percent plus unemployment rate" is an example. A more dramatic one is former Office of Management and Budget Director David Stockman's estimate of a "real" unemployment rate of more than 40 percent -- a statistic that Trump frequently cited on the campaign trail. Here's how Stockman got there last year:

At the present time, there are 210 million adult Americans between the ages of 16 and 68 -- to take a plausible measure of the potential work force. That amounts to 420 billion potential labor hours, if we accept the convention that all adults are at least theoretically capable of holding a full-time job (2,000 hours/year) and pulling their share of society's need for production and work effort.

By contrast, during 2014 only 240 billion hours were actually supplied to the US economy, according to the BLS estimates. Technically, therefore, there were 180 billion unemployed labor hours, meaning that the real unemployment rate was 42.9%, not 5.5%!

So if you count every last student, parent staying home to take care of kids, adult staying home to take care of elderly parents, disabled person and person lucky enough to be able to retire before age 68 as unemployed, then yes, the unemployment rate is -- or at least was in June 2015 -- something like 42.9 percent.  That's pretty silly, as even Stockman points out in the essay from which the above passage is taken. It also provides no information on how this compares with past years and decades.

There actually is a perfectly good measure, readily available from the Bureau of Labor Statistics, that does allow for long-run comparability -- for men at least. This is the employment-to-population ratio, and for women the huge gains in labor-force participation tend to swamp everything else before the 2000s. Here's the prime-age employment-to-population ratio, broken down by gender, since 1980.

Men and Women at Work
Prime-age (25 through 54) employment-to-population ratio
 
Source: Bureau of Labor Statistics
ATTENTION! Y-axis does not go to zero!

This paints a picture of a labor market that's gotten a lot healthier since the last recession (which the headline unemployment rate also shows), but is still quite weak by historical standards (which the headline unemployment rate doesn't show). And it's a chart we really ought to keep watching.

  1. How did I get that? I multiplied the January 2009 labor-force-participation rate of 65.7 percent against the March 2015 estimated 16-and-over population of 250.1 million to get an adjusted labor force of 164.3 million. Using the actual March 2015 employment number of 148.3 million, that gave me an unemployment rate of 9.7 percent.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

To contact the author of this story:
Justin Fox at justinfox@bloomberg.net

To contact the editor responsible for this story:
Stacey Shick at sshick@bloomberg.net