October Jobs Report Preview: Will Revisions Trick or Treat?

The foundations of the jobs market are shaky, but not yet shaking. Repeatedly over 2023 – through rising interest rates, a banking crisis, a “tech-cession,” strikes and strike threats, and war and conflict – the labor market has managed to defy expectations of an imminent implosion. It hasn’t always looked that way in real time: At several moments, payroll growth has appeared on the cusp of souring, only for a clean narrative to be washed away with subsequent data revisions.

Backward revisions have been atypically volatile in recent months, but the vulnerability of the macro labor market and economic narrative to revisions that are usually a footnote betrays the frail convictions underlying our collective sense of where the economy is headed. Momentum would suggest that the post-revision trend of gradual cooling – if not modest reacceleration – is likely to continue. Still, near-horizon risks continue to skew toward the downside, and momentum is never infinite.

Here are three trends we'll be watching for in the October jobs report:

Revisionist History On the Rise Post-Pandemic

The headline payroll jobs numbers published by the Bureau of Labor Statistics undergo two revisions during the months immediately after their release.

Data are first reported at the start of the following month referenced in the data – approximately two weeks after the data are collected. This report is called the “Advance” estimate. As more data are collected, the Advance estimates are revised one month later, known as the “Revised” estimate. Finally, two months later, the data are revised once again in the “Final” estimate. (There are also annual benchmarking revisions and seasonal adjustment updates.)

Revisions have been, on average, larger post pandemic than they were during the decade-long economic expansion that preceded March 2020, making it particularly difficult to discern a definitive trend for the jobs market.

Average revision between the Advance and Final estimates of the headline nonfarm employment growth has been nearly three times larger in absolute terms than it was from 2009 to 2019. Revisions between the Advance and Revised readings have jumped even more dramatically.

These dramatic swings between the initially reported and final data can be particularly confusing for economy watchers. For example, one month it can appear that the labor market is doing ok, only for revisions the next month to suggest a more dire situation, and then further revisions a month later to swing back toward a more optimistic view.

At the industry level, the biggest drivers of total revisions (from the Advance to the Final estimate) post-pandemic have been Leisure & Hospitality, Professional & Business Services, and Financial Activities.

While some economists have speculated that data revisions become particularly large leading into economic turning points – such as when the economy is on the cusp of a recession, or when a recession is about to flip to expansion – there is, perhaps, a more natural underlying explanation. When the economy is in flux – as it is at turning points – there are more firm formations and bankruptcies. Further, company payrolls change more quickly due to layoffs and rapid hiring and companies in turmoil may simply be less likely to respond to government surveys.

All of these factors introduce volatility into the data collected by the BLS – volatility that is unlikely to abate as uncertainty and structural change continue to rattle the foundations of the economy. 

Self-Employment Buffers Likely to Cloud the Payroll Outlook on the Near Horizon

After increasing steadily since 2011 – with a brief pause during the pandemic – the number of workers who are self-employed steadily increased through 2021 by around 120,000 per year. The share of all workers who are self-employed, however, trended lower from about 11 percent in 2011 to less than 10 percent in 2019. It surged during the first year and a half of the pandemic, but again began to retreat as the labor market accelerated and is now just over 10 percent – only slightly higher than where it stood at the end of 2019.

As a share of total employment, self-employment historically declines during tight labor markets and increases during soft labor markets. This is particularly true during times of turmoil in knowledge-driven sectors – such as the dotcom bust in 2000 and the present – when companies seeking to control costs may opt to hire employees as contractors instead of as full-time employees.

Over the months ahead, this corporate balance sheet risk management could further distort our understanding of the state of the labor market: Even if payroll growth slows, there is likely to be some labor demand buffer from self-employed and contract workers moderating the adverse implications of slowing payroll gains. A pivot toward contract work would also distort conventional metrics of wage, salary and compensation growth since these workers typically do not benefit from standard employment benefits.