July jobs report: How a big headline miss may be overstating weakness

[Music]

Lauren Brice: Welcome back to J.P Morgan's Making Sense. I'm Lauren Bryce from the North America rate sales team here with our chief US economist, Mike Froley. Mike, good to have you.

Michael Feroli: Thanks. Good to be back. 

Lauren Brice: On the first Friday of each month, we sit down to unpack the jobs report, and today we'll cover what the numbers tell us and how they influence your thinking on the path forward for the economy and for the Fed. Lots to chew on in today's data alone with a large miss on the headline number, but a notable drop in the unemployment rate to a low 4.1%. Can you walk us through your initial read on the report and what stood out to you? Yeah, sure.

Michael Feroli: So you mentioned we had a big miss on the headline down 23,000. A lot of that was due to a decline in local government education services of down 50,000. We're not too worried about that. We think some of that may be difficulty seasonally adjusting those numbers around the summer recess. Nonetheless, private hiring did cool to only 30,000 last month and it was revised down the prior two months. So I think the picture you're seeing is you had this pop in hiring in the late springtime. That seems to be kind of fading now into a more stable pattern of low but pretty steady private hiring. As you mentioned, the unemployment rate did tick down again, which was good. On the other hand, that occurred alongside yet another decline in the participation rate. That has been declining quite a bit. Recently, a full percentage point since the beginning of the year. So that was a little disappointing. And then finally, we had another surprise in the average hourly earnings number, which was up barely one 10th of a percent. And on a year ago basis, wage growth now cooled down to 3.2%.

Lauren Brice: Okay. So the signal versus Doyen's question feels particularly live this month. When we look at the claims data we got this week, which were remarkably low whilst ADP came in at its softest in months. You've consistently taken trend signal from ADP, even though the month to month number can be pretty noisy. How do you reconcile these two different messages in the context of the data we have in hand this morning as well?

Michael Feroli: Well, certainly this month, the ADP was on the right side of consensus in terms of flagging a softer number. And ADP relates to private hiring, right? So actually not too far from that 30,000 number I mentioned earlier. I don't think it's all that inconsistent though with what you're seeing in the jobless claims numbers because when you look at the details of the household survey, what you see is actually, even though hiring is not that boomy, you're not really seeing much indication of layoff activity. You didn't have permanent job losers among the unemployed went down. So I think all that is consistent with this environment in which hiring doesn't look all that boomy, but at the same time, you're not seeing a lot of layoff activity or other signs of businesses contracting. 

Lauren Brice: Okay. And then when we look at leisure and hospitality specifically, which has been distorted for the last couple of months running from the Memorial Day calendar shift, you flagged July as the month the sector should normalize back to trend, but we did not see that today. How does this inform your read on the broader picture for service sector hiring?

Michael Feroli: Yeah. So as you mentioned, we had a decline of about, or alluded to you, we had a decline of 40,000 in July. We don't think that was necessarily distorted by calendar issues, nor for that matter, do we think there was much of a big World Cup effect in terms of leisure and hospitality employment. So this just looks like a bit of a disappointment, a genuine disappointment. Now we'll get retail sales next week. So it will be interesting to see whether that decline in hiring corresponded with maybe a pullback in spending for things like restaurants and bars and so forth.

Lauren Brice: Okay. And then if we look at government employment, which has been a source of noise recently, a sharp jump in local government in May before fading in June, which was tied partly to poll workers. We saw a significant miss today with government hiring down roughly 50,000. When do we get a cleaner read on whether federal workforce reductions are producing a more durable drag?

Michael Feroli: Yeah. So I think in terms of federal employment, that's been a little steadier since, at least since late last year when you had all the deferred resignations show up. So you had a small decline in federal employment this month, but we're not looking for, on a go forward basis, as much noise in the federal numbers as we had over the last year when some of the bigger DOGE related effects made it through. And then local government, as I mentioned earlier, you're going to have some noise around the summertime when you have layoffs and then hiring again later in the summer. So that noise may not be totally done, but the federal side should be more, I think, quiet going forward. 

Lauren Brice: And then if we look at JOLTS data, which showed, I guess, tentative improvement, the vacancy to one employment ratio still edging up and the hiring rate nudging higher, though the quit rate stayed stuck near cycle lows. How do all these factors taken together help your thinking around whether we might be leaving the low, high, low fire environment we've been in for some time?

Michael Feroli: Right. Well, it doesn't look like that's ending anytime soon. We saw that again, as you mentioned in the JOLTS data. I think we saw hints at that in the July unemployment report data. I think the low quits may be reflective of some of the lack of confidence that people seem to have in the labor market, which you see in a number of surveys. That may also be why you're seeing wage growth slow even with a relatively low unemployment rate. But right now it seems like that sort of lack of dynamism in the labor market is something that's going to be here for a while.

Lauren Brice: And then I guess in a similar vein, last time we spoke, we discussed the AI narrative as it relates to shifts in labor market dynamics. Has anything changed for you there as we move through this hyperscaler earnings season?

Michael Feroli: Not really. I don't think we're seeing yet big effects in terms of the actual numbers. As I mentioned, it may be feeding into psychology and how confident people are in the labor market, but in terms of the actual numbers, no, no big effects. And I would say one of the areas that has been kind of the center of concerns about AI related job loss is professional and business services. Within that broad category, you have things like software engineers, accountants, all these professions that may be at risk. There you're actually seeing job growth this year pick up relative to last year. And you had another good increase in employment in that sector last month. So I think right now the AI story is still mostly perspective rather than anything we're seeing in the data. I might also add that within today's number, you did see a pretty decent increase in construction employment, particularly non-residential construction. So some of that big data center build out may be showing up in that sector. 

Lauren Brice: Okay. And then I guess if we step back a little bit, we've been discussing since the beginning of the year, the expectation for job growth to catch up to a strong underlying growth impulse. Taking in recent developments domestically as well as abroad, how are you thinking about the US growth narrative as we look through the second half of the year?

Michael Feroli: Yeah. So I don't think our view changed all that much because of the July jobs number. You had a lot of noise, some of it related to government. I think the picture on private hiring looks relatively steady. Now how that translates into overall GDP growth is a bit tricky. We do think we in the current quarter, third quarter, should get a pop in inventory building, which will temporarily support those numbers. But overall, I still think we're kind of in a sort of a 2% GDP economy and we don't see much sign that we're breaking out to the upside or downside yet of that.

Lauren Brice: Okay. And then if we turn lastly to the Fed, last week's press conference brought about a sharp move higher in backend yields and a steepening of the curve with markets questioning Morsh's credibility after two consecutive holds. The backend still hasn't really been able to find a bid. What do you think we need to see from Fed speakers for that to change?

Michael Feroli: Yeah. So besides the chair, you have had pretty consistent signaling from many other Fed officials who have spoken in the period since the last FMC meeting. And I think what you're hearing is consistent, which is most of them have what's called a bias to hike unless they see further moderation in inflation. Now, today's number was greeted a little dovishly, which makes some sense given the questions it may raise about the momentum of growth, the questions it raises about underlying wage inflation. But I still think the probably more decisive news we'll get between now and the next meeting will come in the two CPI reports that we will get one next week and then one in September. 

Lauren Brice: Last question from me. We have today's data in hand, but another jobs report and two additional CPI reports still ahead of the September FYMC meeting. What could lead you to change your call for an early move in the policy rate versus your base case of a first move in December?

Michael Feroli: I don't think what we saw in the July report gives us big reason to change our expectations. Probably at the margin, as I mentioned earlier, probably at the margin takes down some of the risk of September. But for September to increase us a risk, we would, I think, need to see those two CPIs come in, the core numbers come in on the firmer side. So I think it's easy to say that if we got two three tenths or an average of three tenths in those two reports, that they would hike in September. I think if you got two one tenths, they would pause and then there's somewhere in between that will probably put them on their indifference curve.

Lauren Brice: Mike, thank you. Always a great framework for a complicated week of data.

Voiceover: Thank you. Thanks for listening to Research Recap. If you've enjoyed this conversation, we hope you'll review, rate, and subscribe to JP Morgan's Making Sense to stay on top of the latest industry news and trends available on Apple Podcasts, Spotify, and YouTube. This communication is provided for information purposes only. Please visit www.jpmm.com/research/disclosures for important disclosures. Copyright 2026 JP Morgan Chase & Co. All rights reserved. Take 

[End of episode]