What Do You Do With Jobs Data You Just Don’t Trust?
Let's make sense of a revision-heavy January non-farm payrolls report that we don’t completely believe. Meanwhile, there's something that will influence markets more.
You're reading 0 of 1 free page.
Register to read more or Unlock Pro — 50% Off Ends Soon
The January non-farm payrolls report arrived Friday morning. What do we make of it?
Let’s start with the annual revisions.
"The seasonally adjusted total nonfarm employment level for March 2024 was revised downward by 589,000. On a not seasonally adjusted basis, the total nonfarm employment level for March 2024 was revised downward by 598,000, or -0.4 percent. Not seasonally adjusted, the absolute average benchmark revision over the past 10 years is 0.1 percent."
That's not something that normally attracts a lot of attention. So in addition to that adjustment for the total employed as of March 2024, they took the monthly numbers down, on average by 20,000. From a monthly average of 186,000 to 166,000. That brings total overstatement (as I figure it) to a bit over 800,000, but since the March 2024 revision bore the brunt of it (the annual part), the monthly revisions don’t seem too bad. It's all a bit weird and reinforces the difficulty we face making investment and policy decisions on data that is less accurate and timely than is ideal.
The algos latched on to the Establishment Headline of only 143,000 jobs created. That immediately caused a rally in Treasuries. That was, pretty much the only weak part of the report:
- Two-month revisions were 100,000. Not just strong, but strong enough that maybe we can be less suspicious that every month is overstated, only to be revised down later?
- Earnings were much better than expected, with last month’s annual level ratcheted up to 4.1% from 3.9% and this month’s annual level clocking in at 4.1% versus expectations of 3.8%.
- The Unemployment Rate dropped back down to 4%. Maybe we will hear less about the Sahm rule in the coming weeks. The number is even more impressive since the Labor Participation Rate actually increased. That would all be great if it wasn’t for the Population Adjustments in the Household Survey. The adjustment added 2 MILLION employed and “only” 2.1 MILLION to the labor force. That mostly begs the question of whether the so-called Sahm rule was ever triggered last year, given how far off we were on the Household Survey data.
So with all of this:
- The Fed has little justification to cut for employment, as the two adjustments to last year seem at least partially offsetting (the unemployment rate is one piece of data the mainstream media understands and can run with).
- Wage growth is yet another thing that will add to concerns about the potential for inflation to reverse the general downward trend.
- I would do the “shoulder shrug” emoji here, because quite honestly, we are all stuck trying to analyze a set of numbers that we don’t fully understand, don’t completely believe, and can move on to more important questions: What is next from the Trump administration?
- Policies, those enacted, those threatened, and the responses to those policies, domestically and internationally, are going to move markets far more than this report.
The NFP is losing its ability to drive markets as it really is only one piece of a complex puzzle when judging jobs, and the Trump administration’s policies are going to do a lot more for jobs and interest rates, than we can divine from any one number.
Yields should drift higher, but stocks will move on stories other than yields. They seem to be in “rally” mode and grinding higher, but just grinding, not spiking. Call it a neutral view here on risky assets, with an inclination to sell into strength, but only an inclination, not an overwhelming desire!
Good luck. It is incredibly difficult to believe that in less than two weeks we’ve absorbed and “solved” cheap AI, a tariff war, and so much more.
It's going to be an “interesting” year.
