A good report, but not good enough to stave off fears that the recovery is losing steam. The US added 661,000 jobs in September according to today’s jobs report from the Bureau of Labor Statistics, and cut the unemployment rate by a half-point to 7.9%. However, the latter appears to be an artifact of a shift of nearly a million people out of the labor force rather than an organic cut to the number of unemployed people.

The topline number is certainly strong, but it’s also the smallest gain in the recovery so far. It’s a disappointing number in part because the ADP report on Wednesday appeared to indicate more robustness in the labor market:

Total nonfarm payroll employment rose by 661,000 in September, and the unemployment rate declined to 7.9 percent, the U.S. Bureau of Labor Statistics reported today. These improvements in the labor market reflect the continued resumption of economic activity that had been curtailed due to the coronavirus (COVID-19) pandemic and efforts to contain it. In September, notable job gains occurred in leisure and hospitality, in retail trade, in health care and social assistance, and in professional and business services. Employment in government declined over the month, mainly in state and local government education. …

In September, the unemployment rate declined by 0.5 percentage point to 7.9 percent, and the number of unemployed persons fell by 1.0 million to 12.6 million. Both measures have declined for 5 consecutive months but are higher than in February, by 4.4 percentage points and 6.8 million, respectively.

The U-3 unemployment number comes from a formula which uses the labor force as its denominator. If the labor force increases, it tends to push up the U-3 number (unless more jobs get added than people to the workforce). If people exit the workforce, the U-3 number tends to decline. In September, 879,000 people left the workforce, which accounts for almost all of the decline in unemployed (970K) in the same Household survey.

In another tell, the number of overall employed people rose only 275K in the Household survey, while the civilian labor force declined by 695,000 — even though the civilian population grew by 184,000. This is most likely a survey issue, to which the Household survey is particularly prone. It could mean one of two things. First, if these numbers are real, then there was a lot of job destruction in September and it has discouraged a lot of workers to get out of the market. If they’re not real, then the next survey will likely correct those numbers and we will see a jump in the U-3 unemployment rate in the October report. (I’d bet on the latter.)

CNBC reports this as a miss on expectations, which were already on the low side. They also provide a chart showing the slowing momentum:

Nonfarm payroll rose by a lower than expected 661,000 in September and the unemployment rate was 7.9%, the Labor Department said Friday in the final jobs report before the November election.

Economists surveyed by Dow Jones had been expecting a payrolls gain of 800,000 and the unemployment rate to fall to 8.2% from 8.4% in August.

However, CNBC also looks at the long-neglected U-6 rate to see a more positive outlook on unemployment:

The decline in the unemployment rate came along with a 0.3 percentage point drop in the labor force participation rate to 61.4%. However, a separate, more encompassing measure that counts discouraged workers and those working part-time for economic reasons also saw a notable decline, falling from 14.2% to 12.8%.

The U-6 rate came in handy during the recovery from the Great Recession, as the usual topline numbers hid a massive overhang of sidelined workers. Its utility evaporated over the past three years as the Trump boom coaxed almost all of them back into the workforce. If this job-creation momentum keeps slowing, we might end up back in the same position, where the topline numbers shield the reality of the job markets, and where we have to guess at the causes of wage stagnation — as we did between 2010 and 2017.

The big question will be how this plays into two political issues. We are still on the positive side for job creation now, so the White House can keep pointing to economic growth. The job additions so far this year exceeded expectations at the beginning of the pandemic, even if they’re slowing down a little too soon now. Team Biden will argue over the momentum, but with their candidate arguing for slower reopenings and potential new shutdowns, it’s tough to see where they’re going to resonate on job creation.

This might light a new fire under some people to reach a compromise on a new stimulus plan, too. Both sides have constituencies that are demanding action, and any sense that recovery might be flagging is a problem for both parties. Too many people are on the cusp of layoffs now to let this slide all the way to an election … or so one would think.