What’s Happening With the Job Numbers?

I wrote here about the fact that the federal government seems to have cranked out absurdly optimistic job numbers just in time to give the Democrats a boost in the midterms. But that isn’t the only puzzle in current job data. It turns out that America’s employment situation is a bit of a mystery. I asked my colleague John Phelan to explain what is going on. This is his response.

In the first two quarters of 2022 the economy of the United States shrank. Under previous administrations a recession would have been declared instantly, but not this time. How could there be a recession, some argued, when employment was booming?

These estimates of the number of new jobs in the United States come from the Bureau of Labor Statistics’ (BLS) Establishment survey. As Business Insider explained in 2014:

…data for the Establishment is collected by the U.S. Bureau of Labor Statistics as part of its Current Employment Statistics (CES) survey, which incorporates the payroll records of some 144,000 non-farm establishments and government agencies, covering workers at some 554,000 individual worksites. In addition to determining the number of people employed at the surveyed locations as of the payroll period including the 12th of each month, the BLS collects data on the number of hours worked, earnings and the industries in which individuals are employed at the surveyed organizations.

These data do, indeed, show sustained increases in the number of new jobs in the economy: increases in every month since December 2020, 263,000 in November 2022 alone, or 3.8 million new jobs since January.

But when the BLS puts out its Employment Situation report each month, it provides another estimate from the Household survey of the number of people actually employed. This, as Business Insider explained:

…is conducted by the U.S. Census Bureau, which surveys some 60,000 American households during the week of the 12th of each month as part of its Current Population Survey (CPS). In addition to determining the employment status of the individuals in each surveyed household, which it classifies as employed, unemployed or not in the civilian labor force, the Census collects data on their demographic profiles, including race, Hispanic origin, age, sex, et cetera.

The BLS explains:

The numerous conceptual and methodological differences between the household and establishment surveys result in important distinctions in the employment estimates derived from the surveys. Among these are:

– The household survey includes agricultural workers, self-employed workers whose businesses are unincorporated, unpaid family workers, and private household workers among the employed. These groups are excluded from the establishment survey.

– The household survey includes people on unpaid leave among the employed. The establishment survey does not.

– The household survey is limited to workers 16 years of age and older. The establishment survey is not limited by age.

– The household survey has no duplication of individuals, because individuals are counted only once, even if they hold more than one job. In the establishment survey, employees working at more than one job and thus appearing on more than one payroll are counted separately for each appearance.

Which is ‘better’? On points 1 and 4 the Household survey would seem to be a better measure. On points 2 and 3 the Establishment survey would seem to have the edge. Either way, to put it in simple terms, the Establishment survey measures the number of jobs while the Household survey measures the number of people employed.

And when we look at the data on people employed from the Household survey a very different picture emerges. The number of Americans employed has actually fallen in four of the last ten months (and both of the last two) and there were 1.3 million more Americans employed in November than there were in January (you can see the same pattern in the state level data, as in my home state, Minnesota, for example).

These are both surveys with different methodologies measuring different things so we shouldn’t be surprised that the two don’t match exactly. But when they diverge so substantially, we must ask why.

But what if there is no discrepancy, or, at least, it is much smaller? Yesterday, John wrote about a report from the Philadelphia Federal Reserve which found that:

…the employment changes from March through June 2022 were significantly different in 33 states and the District of Columbia compared with current state estimates from the Bureau of Labor Statistics’ (BLS) Current Employment Statistics (CES).

In the aggregate, 10,500 net new jobs were added during the period rather than the 1,121,500 jobs estimated by the sum of the states; the U.S. CES estimated net growth of 1,047,000 jobs for the period.

The Establishment survey – the job numbers – are just wrong, in other words.

We should note that, over the period the CES (jobs) was reporting a gain of 1.0 million new jobs, the CPS (people employed) was reporting a fall in the number of Americans employed of 347,000. If the Philadelphia Fed is correct, there is still a discrepancy, but a much smaller one, which could be explained by more people holding multiple jobs to make ends meet – hardly a ringing economic endorsement.

So why might the jobs numbers be so wrong? One can never rule out sheer incompetence where government agencies are involved. But Zerohedge – who deserves a hat tip for his work on this subject – hints at something darker:

As an aside, it appears this is not the first time the “apolitical” Bureau of Labor Statistics has pulled such a bizarre divergence off: it happened right before Obama’s reelection:

And then again: right before Hillary’s “100% guaranteed election (because one wouldn’t want a soft economy to adversely impact her re-election odds).

I do not like to impugn people’s motives. Suffice it to say that when you hear blockbuster jobs numbers, ask yourself how many more people are actually employed.

Notice: All comments are subject to moderation. Our comments are intended to be a forum for civil discourse bearing on the subject under discussion. Commenters who stray beyond the bounds of civility or employ what we deem gratuitous vulgarity in a comment — including, but not limited to, “s***,” “f***,” “a*******,” or one of their many variants — will be banned without further notice in the sole discretion of the site moderator.