top of page
Alex Haseldine

Not as Bad as it Feels

The July US payrolls report was the catalyst for a rash of recession fears, prompting the mini crash in risk assets and a major shift in expectations about the pace of Fed rate cuts.


Much of the angst was because the rise in unemployment triggered the Sahm rule, (whose inventor doesn't actually believe that it is so relevant this time around). And once you leave that particular statistical regularity aside there is no real sign of imminent recession in the data itself: it certainly isn't as bad as the market reaction makes it feel.


In the first place the number of workers not working or working part-time because of the weather jumped to 1.54M for July vs vs 280,00 in June. That suggests that August payrolls may well reverse some of apparent weakness in the July data, and potentially "un-trigger" the Sahm rule!


Before that we will have the preliminary estimates of the annual benchmark revisions to the payroll data, scheduled for the 21st August at 10am. That will feature new information from state unemployment insurance and tax records as well as data for Q1 from the Quarterly Census of Employment and Wages. Most FOMC members will want to see those two reports before drawing firm conclusions about policy.


In the meantime, running through the July data itself doesn't exactly scream recession.


The household survey has been diverging from the non-farm payroll survey for some time.

Non-farm payrolls suggest that employment has certainly slowed, but it's not flashing red. It's currently growing at 1.3% (6 month annualised rate) vs the non-recession average of 1.66% (1990-2019).

Once you factor in average hourly earnings (still holding up at 3.8% yoy), falling inflation and a farily standard average work week, consumers seem to be doing fine.

Average hourly earnings have decelerated to 3.8% year on year, and in realterms are up around 1%. as the July ADP commentary notes: With wage growth abating, the labor market is playing along with the Federal Reserve's effort to slow inflation. If inflation goes back up, it won't be because of labor." (In the second, quarter unit labour costs were up by less than 1% yoy).

Job openings are still slightly above their pre-pandemic peak: 8184k vs 7594k in November 2018.

However, while labour demand has been drifting lower at the margins (household employment marginally lower and job openings declining), the labour force has been growing fairly quickly (up 1.2M since January). So the Sahm rule has been triggered more because of increased labour supply than outright falls in employment. That's the opposite of what happened prior to previous recessions.

There is currently approximately 1 job per available worker (if all job openings were filled), down 3.6% from the highs of March 2022. This is nevertheless as tight a labour market as any before the pandemic by this particular metric.


Powell's summed up the state of play before the July data as follows: “Strong job creation over the past couple of years has been accompanied by an increase in the supply of workers, reflecting increases in participation among individuals aged 25-54 years and a strong pace of immigration. Nominal wage growth has eased over the past year and the jobs-to-workers gap has narrowed. Overall, a broad set of indicators suggests that conditions in the labor market have returned to about where they stood on the eve of the pandemic - strong but not overheated.”


We doubt whether the Fed will read the July data as the the starting gun for the next recession, but they have already fired the starting gun for a rate cut in September: how aggressively they want to cut will depend on just how much confirmation they get from the benchmark revisions and August payrolls about the state of the labour market.


But 25bp cuts this year would be the most likely view when you take the emotion out of the picture.

Comments


bottom of page