a man walks past a ‘now hiring’ sign posted outside of a restaurant in Arlington, Virginia
A key measure of labour market tightness has been the relationship between job openings and the number of unemployed people; with the former outnumbering the latter by a ratio of 2.0 to one. © AFP via Getty Images

The writer is chief investment strategist at Charles Schwab

It’s often been said that a key risk in a monetary policy tightening cycle is that the Federal Reserve hikes interest rates until something “breaks”. That raises questions of how far the Fed will now go to tackle surging inflation.

Part of the reason that is cited for the central bank’s current aggressiveness is the strength of the US labour market and the potential for that to add to inflation.

But a look under the hood highlights that there may already be some breakage in the labour market, not picked up by traditional headline indicators — including payroll growth and the unemployment rate.

The “establishment survey” is what generates the headline payrolls number each month when the Bureau of Labor Statistics releases its US employment data. According to that survey, 315,000 jobs were added in August, which was strong, but well down from the prior month’s 526,000. Of course, counting payrolls only results in an estimate of the number of jobs created; it doesn’t measure unemployment.

That’s where the US household survey comes in, from which the unemployment rate is calculated. It’s a survey of households’ members, so it counts people, and whether they’re employed or not.

A recent trend picked up by the household survey is the increase in multiple job holders. If one person picks up a second or (God forbid) a third job for economic reasons, that is still counted as one employed person per the household survey. However, it’s possible those additional jobs get picked up as individual payroll jobs within the establishment survey.

An additional sign of underlying cracks in the labour market is the falling number of full-time jobs and the very sharp inflection higher in part-time employment. The gain of 442,000 jobs in the household survey in August appeared on the surface to be strong. But that was more than all accounted for by part-time workers, with full-time jobs actually shrinking by 242,000. It was the third month in a row of declines, totalling 465,000 over that period.

Another fly in the ointment of labour market statistics is associated with job openings — the most common tracker coming from the Job Openings and Labor Turnover Survey (Jolts). A key measure of labour market tightness has been the relationship between job openings and the number of unemployed people; with the former outnumbering the latter by a ratio of 2.0 to one.

The problem is that the Jolts statistics arguably overstate the number of actual individual job openings. One of the criteria for a job opening is that there is “active recruiting” for workers by an establishment. That may include advertising, internet notices, signs, word-of-mouth “announcements”, contact with employment agencies, or setting up at a job fair or similar source of possible applicants.

In addition, the pool of labour available for those jobs spans beyond just individuals who are unemployed. Potential job switchers, included in the number of people employed, should also be considered as potentially competing for those job openings. This suggests that the labour market may be less tight than conventionally believed, confirmed by recent research by the St Louis Fed.

The Fed has explicitly stated that its goal is to weaken job openings, without a significant rise in the unemployment rate — a narrow opening in the needle it’s trying to thread. But the Fed also cites the need for more restrained wage growth — which is elevated by historical standards, but remains below the rate of inflation. This means real wage growth is still in negative territory.

There is another reflection of weakening demand for labour and that is the number of hours each week companies are asking of their workers.

Notwithstanding the healthy reading on August payroll growth, there was yet another reduction in the workweek, which has been flat or down in five of the six months through August. At 34.5 hours, it is tied for the lowest reading since April 2020, when the pandemic lockdown was in full force. The decline in hours worked was so significant that it resulted in the first decline this year in the index of aggregate hours worked.

With labour the highest input cost for many companies, and economic growth and demand weak, the hints of weakness in the labour market are likely to foreshadow further deterioration to come. As the Fed has been pointing out, it may be a necessary ingredient in the quest to quell the surge in inflation.

Copyright The Financial Times Limited 2022. All rights reserved.
Reuse this content (opens in new window) CommentsJump to comments section

Follow the topics in this article