Moody's Talks - Inside Economics - Dazed and Confused

Episode Date: June 5, 2026

Dante joins the Inside Economics crew to dissect the May jobs report, which he describes as shocking. The team discusses whether there is enough evidence to declare that the labor market and underlying job growth have shifted into a higher gear, and debates the growing disconnect between the payroll and household surveys. The stats game delivers some interesting insights about the strength of job growth for women and the impact that remote work has had on young college graduates. Hosts: Mark Zandi – Chief Economist, Moody’s Analytics, Cris deRitis – Deputy Chief Economist, Moody’s Analytics, and Marisa DiNatale – Senior Director - Head of Global Forecasting, Moody’s Analytics Follow Mark Zandi on 'X' and BlueSky @MarkZandi, Cris deRitis on LinkedIn, and Marisa DiNatale on LinkedIn Questions or Comments, please email us at [e-mail]. We would love to hear from you.  To stay informed and follow the insights of Moody's Analytics economists, visit Economic View. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Welcome to Inside Economics. I'm Mark Sandy, the chief economist of Moody's Analytics, and I'm joined by my two trusty co-host, Chris DeReedies, Marissa Dina Talley. Hi, guys. Hey, Mark. Hi, Mark. Hi, Chris. And this is Jobs Friday, June the 5th. We just got today the numbers for May of June, excuse me, May of 2026. And as is typical, we have Dr. De Antonio, Dante De Antonio, join us. Hey, hey, doctor. Dr. How are you? Hi, Mark. I'm doing all right. How are you? I'm doing okay. You always ask me if I'm doing okay, and I always answer, I'm doing okay. We got a little bit of a technical thing going, so there's only, if you're listening to this, then no worries.

But if you're trying to view this on YouTube, all you're seeing is Chris and Dante. The rest of us are having a little bit of Internet difficulty, so we're keeping our cameras off. But, you know, you guys are good-looking guys. I mean, there's something to look at, sort of, I guess, you know? Appreciate the vote. Versus laughing. Yeah. Yeah.

Versus laughing. Well, we got a lot of talked about. Hey, I do want to say that I am deep in the bowels of the Congressional Budget Office, CBO. This is twice, I'm on the board of economic advisors for CBO. And twice a year for a day we get together, all the economists on the board. board, advisory board, along with guests from outside, convene and talk about issues that are relevant, pertinent to the CBO's forecasting process. Obviously, they're the folks that do the budgeting.

So really fascinating day, although I will say the job numbers didn't really get on the radar screen, at least to a large degree. But for this podcast, we're going to be all about the job numbers, aren't we? Is there any other issue to talk about other than the job number? They were they were strong, surprisingly strong. But I don't know, anything else we should be discussing other than jobs? I think that should do it. Well, Chris.

Well, what's the vibe in the room? What was the vibe in the room? The vibe? Among the economists. Yeah. You know, they're mostly acomeditions, so they're not like day to day. You know, there's some Wall Street folks. And they were, there was a lot of discussion about the room.

run up an interest rates as a result of the employment numbers. We're now, we'll talk about this, but the 10-year treasury yield, what is it, 36 or something? I mean, excuse me, 4.6? 4.6%? Something like that. Close. Yeah, 4.056, something like that. Yeah, yeah.

And then a couple of other economists talking about the economy, but no, it's more longer-term issues, you know, around labor force, productivity, AI, you know, those kinds of things. So less about the here and now. But anyway, let's dive in. Dante, you want to give us a rundown on the numbers? I don't know where you want to begin. Feel free.

Yeah, feel free. Oh, and by the way, just get this, you know, out there and pass so we can work. We can go past this. You know, your estimate was like way off, my friend. What was that all about? You know, I missed, I didn't hear your estimate yesterday, though. You asked me for mine, but then you didn't give me one back. Did you ask me for mine?

I didn't, but you didn't offer it, so now I don't know where you stood before this morning. I'm the only one that had to put myself out there here, right? Well, that's a basic principle of forecasting. Don't forecast unless you have to, right? Yeah. There you go. There you go. I think if I remember, I told you yesterday that I'd be shocked, right, if this number was above 100,000.

So I'll stick with that shocked and I'll add maybe a little bit confused at the same time to what we got today. Good way to describe it, that's for sure. Shocked and confused. The headline numbers 100. Yeah. There you go. Got 172,000 jobs added in May and strong positive revisions to the prior two months,

a total of a little over 90,000 in revision upward to March and April. So all that combined brings three-month average growth to 188,000, right? So we added almost 600,000. jobs in the last three months since a war started in the Middle East, which seems a little bit counterintuitive. If you look at the industry composition just in May, it's still fairly concentrated, just not... Dante, can I just stop you real quickly there? So, you know, if you look at the monthly job gains going back to, let's say, the start of 2025, you know, up until what, the February

of this year... It was basically flat, you know, payroll employment, the numbers you're now talking about based on the survey of businesses. You know, some months up a little bit, some months down, but net, net, we hadn't gone anywhere. And now all of a sudden March, April, and now May, you said an average of 188,000 per month. And that coincides, by the way, with the start of the war in Iran. That just feels weird, no? I agree, yeah. Okay, so you're going to get to the weirdness at some point here.

You know, what's going on? Yeah, part of that. So obviously there was a big decline in February, right? And some of the increase in March and April can certainly be attributed to volatility. Some seasonal adjustment issue there. We had a big strike in healthcare. So some of that decline in February, we got sort of payback for that in March and April. But now we're obviously two months beyond that decline in February.

I don't think you can argue that the gain in May is related to that prior loss in any way. But the concentration of job growth was sort of unusual from what we've seen of late. So health care was still a strong performer, right, added almost 50,000 jobs. But the majority of the rest of the top line gain came from leisure and hospitality and from government, right? So we got 70,000 jobs in leisure and hospitality, 52,000 jobs in governments. Almost all of that government increase was confined to local government excluding education, which, again, is not typically the thing that we think of as a big driver of monthly job gains. outside of those three industries, there wasn't a whole lot of action, right? So construction was up about 17,000, which is certainly a good sign.

Finance was down 22,000. That adds to a sort of string of losses in finance that we've seen of late. So I would say neither of those are particularly surprising. And then the rest of the industries were basically a wash. There wasn't a whole lot happening up or down elsewhere. And the, I mean, I almost like AI is playing a role. here. I mean, AI is juicing construction jobs, right, because of the data center boom. I think that's what that is. And then the financial services is kind of the flip. It feels like

that, you know, AI, because outside of the tech industry, financial services seems to be the most at risk from AI disruption, at least early on. And that feels like what's happening here. So AI feels like it's juicing jobs on the one hand and weighing on jobs on the other. Would you, would, does that feel right? Is that a fair characterization of what's going? I know you can't know for sure, but does that, does that feel right? Yeah, that feels right. I mean, if you look at finance, I think we talk to losses.

We're in insurance and in, you know, credit intermediation, basically banks. And so, you know, it feels like those are places that would be ripe for some of that early automation to have an impact. Okay. And so, extracting from the industries where you had a little bit up and a little bit down, you're saying there's three. sources of the job gain in the month of March, or excuse me, the month of May. One was leisure and hospitality. Two was health care, which has been doing this for all along, even going back to the sort of 2025. And the third, which is seemingly coming out of nowhere, is this big gain in local government employment. And it's not education. It's ex-education. Is that right?

That's right. Yep. Okay. All right. Okay. Okay. proceed. Yeah. So I mean, I think I, I, alluded to you earlier. I'm a little bit skeptical about the gains in leisure and hospitality and government, right? I think if you dive into the details a little bit, so if you look at that local government excluding education piece, it was up about 44,000 jobs in May, you know, that is among the largest monthly increases going back to the year 2000, right? There was a few times right after the pandemic in late 2020 where it increased by that much on a monthly basis, which is not surprising. The only other two times that it happened since 2000, those months were immediately followed by decline. in that industry's employment, right? So it was only ever gone up that strongly and then come right back down, presumably because of a seasonal adjustment quirk. So even though this is not education, there is actually a pretty strong seasonal pattern in this local government X education segment. It actually sort of mirrors education and that employment tends to increase in the summer and then fall off in the fall. So you typically get increases in May and June. My guess is this increase in May was a little bit bigger than usual.

It may just be pull forward from an increase that would typically happen a little bit later in June. And so my guess is you're going to see some payback, some weakness on the other side of this in that local government piece. For leisure hospitality, my interpretation of what you're saying is don't count on these kind of job gains, monthly job gains going forward. That this is more of an anomaly, seasonal adjustment issues in all likelihood. and that the one thing that's durable throughout is health care. So that, that's real. Like, you can count on that. You can get 30, 40, 50K per month from there.

But you're saying I wouldn't count on consistent job gains. Certainly not like what we saw this past month for leisure hospitality and for local government. That's right. Yeah, I mean, the math works out conveniently for me. If you remember what I did give you my forecast was it was 50K, right? And so government and leader in hospitality, if you take those two out of the top line, you get exactly 50K, right? So if you- And Chris, isn't he a great forecaster?

I mean, that's masterful. See how he does it. It's masterful. If you just remove the problems, right, then that's what you're left. Well, there's information there. You're saying you're surprised at how strong things were, and here's the surprise. Yeah. Plain as day.

And, you know, of course, we don't know for sure what's going on, but you're in, you're in, you're intuition, your sense of things is this is seasonal adjustment related in all likelihood, because the gains are just kind of out of bounds. That's my sense. Maybe Chris or Marissa will put me in my place and prove me wrong. But yeah, if you look at leisure hospitality, again, it looks kind of quirky, right? You always have this big run-up in payrolls in Leisure hospitality in May and June and throughout the summer.

So if you look at the unadjusted data, right, the gain this May was pretty close to the gain last May, right? So unadjusted, Lee's-Rost value is up 466,000 this year. Last year, it was up $458,000. So pretty similar increases. If you look at the seasonally adjusted values, last year, the seasonally adjusted increase was

$3,000. This year, it's $70,000. Right? So all of that difference, right, the unadjusted number basically rose by the same amount. All of that difference is coming from seasonal adjustment, right? The factor is different this year. Maybe that's more correct and more right, or maybe it's just,

sort of a quirk in the data and we're getting this big bop, it doesn't really exist. That's so interesting. Does that apply for both leisure and hospitality and local government for both those sectors? I didn't even dive that deep for local government just because it looks so unusual relative to history. Reasonable hospitality increases by big numbers pretty frequently, historically, but this one looks unusual because of that seasonal adjustment quirk. Got it, got it. So sticking to the payroll survey, anything else in there that kind of stands out to you. I mean, I noticed there was hours work that remained unchanged, so neither here nor there. Average hourly earnings, they seem to be, that's wage growth. That seems to be continuing

to moderate, no, on a year-over-year basis? That's right. Yeah, the year-over-year rate came down to 3.4%, which is, it was a little bit, I think it was 3.6 last month. That's actually now, you know, in line with what the ECI was, the employment cost index for the first quarter, right, which is the latest data we have. So I think all signs sort of point to wage growth, certainly moving closer to 3% than 4%, where it was at the beginning of last year. One of the thought I had, Dante, is the birth and death model.

You know, maybe you can explain the need for the birth and death model, and maybe what do you think? Is that playing a role in the stronger numbers? And maybe in also the upward revisions to the previous months? I could certainly explain it. I didn't look closely at the adjustments this month yet, so maybe Marissa did, and she can comment on whether that may have played a role. The reason why we have it, right, is that the, you know, over the month changes in employment are estimated using, you know, a matched sample approach, right? So for firms that responded to the survey last month and this month, right, the matched pairs, you look at the change, right, the growth in employment or, you know, not growth in employment across that matched part of the sample.

The problem there, obviously, is that you can't get new firms, right? So if a firm didn't exist last month and now it does, right, you can't possibly get that birth of employment showing up in that match sample. Similarly, right, if a firm doesn't report this month, we don't assume that that firm closed, we just assume we didn't get the data. So we don't actually, you know, sort of include that potential for firms closing in that match sample, right? We're only looking at firms that operated both in last month and this month. So the birth, death model is designed to provide an adjustment. to that match sample estimate to account for employment driven by new firm creation and employment that was destroyed by firms that were closing since we can't observe those in real time in the in the in the employment in the payroll survey data so it's a necessary adjustment to account for what we know is happening right the firms are opening and firms are closing that we don't directly observe in the sample and we do know that business formation is up if you look at the app business application data that the IRS collects. So if you're a new company, you need to file with the IRS and you apply for that,

and that data is collected. And that has picked up. And that does feel consistent also with AI, right? Because a lot of those companies feel like they're AI related. So it might be playing a role. But you're saying you didn't take a closer look at that birth, death model. Okay. Mercer, did you?

Well, I didn't, but I'm looking at it right now. Okay. It did contribute a lot to leisure hospitality. And it normally does, right? Because that is a industry that has very high churn and turnover in businesses, lots of businesses, constantly going out of business, new businesses open. So in May, the birth death adjustment to leisure hospitality was plus 96,000 on a not-seasonally adjusted basis. So Dante, I think you said it was like plus 400.

thousand NSA in this industry. So about a quarter of that gain is coming from the birth death model on a not-seasonally adjusted basis. So it is large for leisure hospitality. It's the industry with the biggest positive birth death adjustment. And that's, I thought it would be AI, but that's certainly not AI. That has nothing to do with AI. Yeah, interesting. Okay. Okay, Dante, I always ask. I'm going to ask again. What do you think the underlying rate of monthly job growth is? You know, abstracting from all the vagaries of the data that you've identified and then some, what do you think it is? What did you think it was last month and what do you think it is now this month, given the strong data that we just received? Yeah, so I think throughout the first couple months of the year, I think my answer to that question was typically 50K, right?

That had been sort of, it seemed like a pretty reasonable, you know, trend job growth. I think it's hard to stay at 50K given what we've seen in the last three months, but also I don't think, you don't want to overreact to a couple strong months either. So I'd say it's probably, you know, 75, 80,000 now, you know, sort of and wait and see what happens over the next few months. Okay. The other question I had more fundamentally is, do you think that the tax cuts could be playing a role here? You know, the tax cuts for both businesses, we've got the expensing. That's the most prominent aspect of the tax cuts for businesses. And then, of course, the individual tax cuts, they were in full swing in March, April, and May.

They've got to have been playing a role here, right? It may be that contributes to leisure and hospitality. It seems like it might. It could, yeah. And leisure hospitality is also strong in the prior two months, too, right? March and April, it got revised higher. You know, it added 74,000 total across March and April plus another 70 in May, right? So you're talking about, you know, almost 150,000 jobs just in leisure and hospitality.

So certainly that would fit with the, you know, sort of tax refund story and consumers may be feeling a little bit better about their position over the last couple months. Okay. Okay, so, you know, this is, that is more fundamental, but it's temporary, right? Because now we're on the other side of the April tax filing and those refunds are, checks are fading very quickly if not, they've not already ended. So if that is an explanation, that's not going to help the strong job growth. That's not going to help support job growth going forward, or at least not for very long. Yeah, that's right. Okay. All right. So the other question I ask is, what do you think break even monthly job growth is that's the rate? monthly job growth consistent with enough jobs to maintain stable unemployment? What do you think that is? And we'll get to the household survey in a second. And just to tease it, the unemployment rate did not change. It was 4.3 percent where it's been now for a few months. Yeah, I think that question gets to where this gets a lot more confusing and complicated, right? Because there's a pretty big disconnect between those two surveys. And that's sort of, you know,

break-even job growth would be reliant on what we think labor force growth is right now. And if you take the household survey its value labor force growth is negative, right? So we've got negative labor force growth. We've got seemingly stronger job growth, but the unemployment rate is still the same. So I still think break-even job growth is maybe 50,000 somewhere slightly below there. I don't know that my view on that has changed fundamentally here. So I think job growth maybe is a little stronger, but break-even probably is not. So what that all implies is if that can, if that's right and it continues, then unemployment should start to fall here. Participation should start to rise.

That's what you're saying. Because the break-even is less than the underlying job growth. Well, that mostly goes to does the disconnect between the two surveys close, right? Because payroll, employment growth isn't what matters for the unemployment rate. So those two, that could keep running strong and the unemployment rate could stay steady if you keep having the surveys tell you two different things. So I guess that's right, right, right. You're right. It's complicated.

Before we go to the household survey, though, maybe I should ask Chris and Marissa the same question. First, Chris, anything to add to Dante's description of the payroll employment report? No, not really. I guess what I can add is I looked at some hotel data earlier this week, RevPars, revenue per available room. And that's way up. So there is some corroborating evidence here that the hotel industry is. is doing quite well. Bookings are up.

They're filling the rooms. They're able to charge top dollar. So, you know, despite the energy price shock, you're seeing that consumers are still wanting to travel and traveling. And that would support some of the strength and the growth of the labor market there. I think the seasonal factors are still valid. But underlying that, I think there might be something fundamental. Yeah.

That might be the tax refunds. but it probably might even be beyond some more fundamental strength than that even. Yeah, I think wealth effects. Certainly playing a role here, wealthier households, you know, not really impacted all that much by energy prices and, you know, going ahead with the plans that they had made months ago. But the weird thing is on that one, and I don't, I mean, I don't, that's a very important observation, but more broadly,

consumer spending growth is pretty punkish, right? It's 2%, it hasn't really picked up. It's kind of going nowhere. Of course, that doesn't mean that you can't see more travel, more people going to hotels and that lifting employment. But broadly speaking, it doesn't feel like demand, consumer demand is picked up to any significant. It's not falling off to any significant degree,

but it's not accelerating either. Okay. It might be brought by African. You're saying an average. Yeah, so it could be that the leisure hospitality industry is relatively strong at the expense of some other parts of the spending economy. Good point. Well, if you remember the spending data that we got, we talked about it last week, leisure hospitality spending was one of the bright spots in the consumer spending report. Oh, that's right.

So that had popped higher and we were losing about why that might be. Oh, for once in the blue moon, all the data line up. Those data line up, right? That makes sense. Okay, that makes sense. That's funny. Okay, so Chris, what do you think, given all the data we got today, what underlying job growth is? What did you think it was before today?

And now what do you think it is? Yeah, I think I was fairly aligned with Dante. So I thought it was in the underlying job growth was in the 30,000, 50,000. Right. That's for month range. And now I have to push that up, maybe six. I think Dante said 75. Maybe I'll go a little less than that, but still substantially higher.

So, 65,000. Okay, so we're averaging 75. And what do you think break even is? Dante was at, what do you, Bonte, you were at 50. So what do you think break even is? So previously I was saying 35,000 based on my, your model, my high-tech modeling. Right, yeah. But now, again, it seems like if you believe the data, you know, factor in some revisions,

I'd have to bump that up. But I'd say probably 45,000. I'm not willing to go all the way. I think there will be some kind of convergence of the two series that Dante mentioned. Okay. All right. Mercia, anything else to add? Did you take a broader look at the birth, death model?

Any other things stand out there across industry or anything else? No. It is quite positive for the month of May, as it typically is. and most of that is leisure hospitality. I don't really see anything else to point out there. I will say I did look at the seasonality in the local government ex-education category. This was, so as Dante said, with leisure hospitality,

maybe there's something weird going on with the actual seasonal adjustment factor. In the local government ex-education, I think there's a seasonal shift going on because the gain in the month of May looks a lot like the gains that you usually get in the month of June. So whatever kind of hiring ramps up for the summer in this category, it seems like it just happened earlier this year than it normally happens.

The magnitude is very similar, but it just looks like it was earlier. And maybe that could be, I don't know, better weather, hotter, hotter weather, you know, drier weather throughout a lot of the country and a lot of the seasonal hiring might have happened earlier. Got it, got it. On the question of underlying job growth and break-even, where were you and where are you now? I am trying to remember where I was. I mean, I think I was pretty low on the underlying.

I think I was around 30K. Now I'm probably up around 60K, I'd say. Break-even is hard, as Dante said, just because these two surveys are so divergent, that, you know, the unemployment rate is 4.3 this month. It was 4.3% in May of last year, too, right? So it really, it's gone a little bit up and down all within the bounds of statistical significance, but it really hasn't changed in the last year. So, I mean, that tells me break-even might be somewhere near underlying. Like, it's probably a little bit under what the underlying, maybe. it's 40,000, maybe it's 50,000, 50 sounds high, but maybe it's 40. It's just hard to tell because the household survey has been so weak. If you look at employment from the household survey, I mean, it's pretty much, you know, it's, it was okay this month. It was positive, but I mean, it's been very divergent from the payroll survey. Right. Well, okay, so

in my mind's eye doing the arithmetic here, it feels like you guys, as a group, think the break-even is, excuse me, the underlying job growth, abstracting from the vagaries of the data is probably somewhere around 75K. I'm just picking on kind of a number. And that you're saying that the break-even is probably something around 50K. That's what it sounds like, feels like, right? Do I get the arithmetic wrong? Okay.

Dante's shaking his head. I got that right? Okay. All right. That sounds, that sounds, I mean, go ahead, Dante. Or was that Chris? Yeah, I was, I was going to ask you what yours was. And I think that's where you go. Oh, yeah, I was going to go. I was going to give to you, you didn't have to ask me, Chris. Of course, you don't have to ask me for my opinion. I should have figured. I was definitely going there. I mean, last month, I, I'm sure I said that the underlying job growth is 25 to 50K and the break even is. is 50 to 75K.

And I'm, I'm, I think, I think it's pretty much the same. I think I'm not changing my mind. I think it's 25 to 50K underlying 50, 75K, break even. And that that would imply that the unemployment rate will continue to slowly drift higher and the participation rate, Labor Force participation rate, will continue to drift lower. I mean, the reason, one reason, Marissa, that the unemployment rate has not changed over the past year at 4-3 is that the participation

rate has dropped, I think, by a point. You can correct me if I'm wrong, but it's declined, and that's helped to keep unemployment from rising. So I think we're still around 2550K on underlying and 50 to 75K on break-even. And that 50 to 70-5-K break-even is my payroll measure mapped to the household survey, because clearly the household survey, the numbers are just different. And let's go there next. Dante, give us a rundown on the household employment data. Sure. I mean, so if you look just at May in isolation, it actually sort of aligns pretty well with what the payroll survey was saying, right? So the labor force did actually increase a little bit. May. It was up 83,000. It wasn't enough to move the participation rate higher, but it held at 61.8%.

You had about 150,000 increase in household survey employment, a little bit of a decline in the number of unemployed people, and that netted out to a sort of no change in the unemployment rate. So if you look just at May in isolation, I think the story is pretty well aligned. It's really if you start looking since the beginning of the year that those stories diverge, where the household survey looks quite a bit weaker, January to May versus the payroll survey. So January to May, household survey were down, 300,000, you said?

In terms of jobs, employment is actually down 300K, roughly speaking. Yeah, so the straight household survey employment measures down 326,000. If you adjust it to the payroll survey concept, it's down 382,000. Oh, gosh. And in the payroll survey, employment is up 400K? They're like mirror images of each other. That's right, yeah. So, you know, on the household survey, the labor force and employment are basically down by, you know, between 3 and 400,000 since January.

where you're keeping the unemployment rate stable, but then you've got payroll survey employment that's up 400,000 over the same period. Right. And typically the law is economists put more weight on the payroll survey than the household survey because the payroll survey is a bigger survey business. Household surveys, what is it? It's 50, 60K in a household, so it's smaller. Do you think that's appropriate in the current context?

I mean, would it be more appropriate to take an average of the two? I mean, how do you think about this? Who's who's telling, I mean, these guys are, these two surveys, they're telling the exact opposite story. I mean, what do you do with that? I still lean towards over short periods of time. I mean, I tend to put more value on the payroll survey, and I would count, you know, January to May still as a pretty short period of time

in the grand scheme of thing. So, yeah, I would put my weight on, you know, the labor market's probably improving slightly in the beginning of the year, not, you know, sort of falling off a cliff. to me, though, the longer this, you know, if you go another three months and that divergence is as big or getting wider, then I think I start to think about it a little bit more closely. Over this time period, I sort of just discount the household survey and say, what's, you know, the payroll survey probably gives us the best sense. But I think that only works for so long before you have to start to question it. Right. Okay. And on the, oh, let me ask you, Maris, the same question before we finish the conversation around the household survey.

Which, who do you believe? The payroll survey or the household survey? How are you handling this schizophrenia? I mean, I think we've seen over the last couple years that the benchmark revisions have been downward on the payroll survey. And it's brought that more in line with the household survey, right? The household survey doesn't get benchmarked like the payroll survey does. So that kind of stays fixed, abstracting from the population control, of course. So I don't know.

I mean, I think maybe it, I agree with Dante. Over a short period of time, if we're looking at month-to-month changes, I would trust the payroll survey more. It's just a much bigger sample size with a lot less prone to error, sampling error. But over a long term, if this persists throughout this entire year, then I'd start to think, well, maybe we're going to get a downward benchmark revision that's going to bring the two back in line. Chris, do you have a view on this? Household versus payroll?

No, I think Dante Mercer explained. I would naturally tend towards an average. Just not sure which one to... That's what Chris does. He takes averages. I take averages. Right down the middle. So that means zero.

Zero job grosses. Yeah. Still not great. Still not great. Yeah. Okay. All right. Hey, Dante, so 4.3% on the unemployment,

what am I going to ask you? Oh, unrounded. It actually declined slightly unrounded. It was 4.34% last month, and it was exactly 4.30% this month. Oh, okay. That's what I was going to ask. Yeah, third significant issue. So we're down to 4.3.

In labor force participation. 4.296. Oh. Oh, did you go next year? Okay, all right. I did. I just had one digit. I just had one digit.

She's showing off now. Yeah. In labor force participation, as I said, that held constant unchanged. And it's still the, there's been a pretty substantive decline over the past year. And it's still older workers. Someone called them premium workers. Yeah. Premium workers. I thought that was pretty good. Who called them that? Was that you? I'm not going to tell you. It's inappropriate for me. That's fair. All right. You shouldn't ask this kind of questions. I just wanted know if it was you quoting yourself. You put me in a very awkward situation, Dante.

No, no. Okay. Anything else on the job numbers that you want to call out, Dante? Oh,

the revisions, the upward revisions. I mean, Marissa was intimating downward revisions, but I did read somewhere that we did get some

so-called Q-C-E-W data and it would imply an upward revision. Do you want to explain, did you see the same thing and do you want to explain it? I did see the same thing. I think we got data through the end of 2025. We got like preliminary QCW data through the end of 2025, which will also be revised here in the next quarter. And that, you know, does seem to suggest a slight upward revision

through the end of 2025, right, which again doesn't necessarily change the view of what's happening in the first five months of this year. But it would suggest that maybe, you know, things weren't quite as bleak as they looked in the second half of last year. year. Right. In QCEW, QCEW quarterly census of employment and wages, the unemployment is based on unemployment insurance records. It's a full count of jobs to which the payroll survey is ultimately so-called benchmark. And they release that data, the QCEW, the day being BLS, with a lag. And we just got the Q4, 2025 data. And if you look at it, it would imply some upward revision to the current BLS surveyed

the BLS data for that quarter. Although, as you point out, that revision is going to be revised. So who knows, you know, where that lands. Okay. All right. Okay. Did you mention the monthly revisions?

I did say they were strongly positive. Yeah. Okay. Yeah. Which is a big change, right? I mean, in all of 2025, we were getting big every month downward revision.

And it was confirmed by the big downward revision with the benchmark to the QCEW. But this is this is flipped. We're now getting upward revisions. Yeah. I mean, what to make of that? I'm just not sure. Yeah.

Okay. All right. So what did you say? Shocked and dismayed? No. Shocked and confused. Yeah. Shocked and confused.

I think that's a really great description. I'm shocked and confused. Brissa, what are you? Are you shocked and confused? A little bit. A little bit. I'd use dazed and confused, but yeah, that's trademarked. Dazed and confused.

Dazed. I like dazed even better. I like that even better. Marissa, are you dazed or shocked and confused? Um, I don't know what I am. I'm numb. Comfortably numb. I'm comfortably numb.

I'm comfortably numb. Okay. All right. Okay, well, let's play the game because I am. away traveling at CBO. We're going to keep this. We are going to definitely keep this shorter. But let's play the stats game.

Oh, before we play the stats game, maybe this won't be that short. Sorry. Chris, Chris, let me call on you. How did the markets react to all this? What are other people thinking about the employment report and what it means?

Yeah. So the stock market is down. The bond market, the tenure treasury, rose substantially, about 7,8 basis points. right? So the idea being that the Fed is more likely to hike than to cut at this point, right? So we'll see higher rates. I took a look at the implied probabilities of a Fed cut as well. And yeah, again, there too, we see pretty substantial movement towards a hike by the end of the year and even into 2027 at this point based on today's report. Yeah, I mean, how could the Fed possibly cut interest rates? I mean, like, what would, would, based on what? I mean, you're creating jobs now, a fair amount of jobs you believe the data. And he even discounted by half for creating jobs. You've got inflation that's well above target and the direction of travel continues to remain disconcerting. You've got high inflation expectations,

which are, I think, about as high as they generally get before the Fed goes on high alert. it just feels like impossible to cut. So, and yeah, I think it's not our baseline forecast. We've taken the rate cuts out and have not included a rate increase, but that's certainly a growing risk here. Yeah. What about the equity market? How did that respond? Negatively.

Negatively. Oh, interesting. And it does seem as though there's a bit of a rotation now outside of AI into the more traditional, more defensive. extensive sectors. So interesting dynamics. You got to all take it with a grain of salt. Sure, sure, sure. It's several weeks of, you know, straight up. And now there's some adjustment here. For you, I don't know, Ethereum, crypto, you know, really down a lot. So I hope you sold before. See how he's trying to turn this on me. This is like incredibly

clever. He's gaslighting you. What's that, Marissa? He's gaslighting you. He's gaslighting me now. My gosh, what, what happened to the respect? No respect. What about all that Ethereum? It's the wave of the future. I can know how sell Ethereum. What are you talking about? You're the Ethereum king. Yeah. One less plane ride is somewhere on the planet, the Sep Kianti or something, whatever. They make good Keontes in Italy, don't they? That's where they make. No? That's the only place they can make.

It's the only place they make. That's good. So yes, yes. Yes, yes. All right. Okay. Let's play the game. We all put forward to stat.

By the way, I'm not putting forward a stat just telling you right up top. But you three are. I want good stats. we try to figure that what the stat is through clues, questions, deductive reasoning. The best stat is one that's not so easy to get it right away, one that's not so hard we never get it. And as tradition has it, we always call on Marissa first.

Marissa, what is your stat? Okay, I have a pair of complementary stats, plus 389,000 and plus 114,000. Okay. Dante, do you want to lead the questioning? Do you want me to do it? You lead, you lead. Is it employment-related?

Yes. Payroll survey? Yes. Is it industry-based? No. No. Is it aggregated over some period of time? It's year-over-year.

I'll tell you that. Year-over-year. Both are year-over-year? Both numbers are year-over-year. Yeah, both numbers are year-over-year. year. I mean, is one of them the total change in headline employment over the last year? No.

No? Does it have to do with revision, Marissa? No. No. It's not industry. And it is jobs. We're talking about jobs. Yep.

Yep. And it's not industry-based. It doesn't have to do with the revisions. Can you give us a hand? So in the payroll survey, they obviously cut the data by industry, and then there's one other cut of the data that they do that we don't often talk about, and it has to do with that cut. Oh, okay. What is the, Dante, you should know this. You're from BLS. Come on, man. Really? You don't know? What's the other cut? That's a, I'm sure we're going to kick ourselves. They do a cut by women, right? Ah, there you go.

Male, female? Yep. Okay, so women are high. Yes, that's right. Yeah, that makes sense. So in the past year, May to May, total payroll employment is up about half a million. $389,000 of that is payroll employment of women. $114,000 is men.

And I think that just goes to the composition of the jobs that have been created, right? We're getting all this growth in health care. in education, leisure hospitality. These are industries, particularly the health care, that tend to be lean more female in terms of the composition of those jobs. So, and actually, if you look at men versus women over the past few years, the composition of job growth has generally been more favorable for women since, since the pandemic ended, coming out of 22-ish. But there's a particularly big gap right now in the past year. And that goes to health care, it goes to leisure hospitality, goes to probably local government too, right? Probably some of that too. Yeah. Right.

And also, and I should give a shout out here to Justin Wolfers because he did post about this maybe a week or so ago. He's a university Michigan. That's right. And we've had him on the podcast, great economist. And the composition, the BLS tracks the share of payroll growth going to men and women. That is breach now women. and make up 50% of payroll employment consistently now for the past year, which is pretty new.

It's been like 49% for the past like 5, 10 years or something, and now it's solidly at 50. That's a great stat. Can I meant to ask, I just slipped my mind what I was going to ask. Oh, manufacturing. Was that up employment, Dante? Was that up or down? It was up, I think it was 7,000. 7K?

Okay. Yeah. Okay. So we're starting to see some job growth there as well. Okay. Dante, you want to go next? What's your stat? I burned my first one talking about divergence between the surveys, but I got one more.

Let's go 0.3%. Mercy, you want to lead the way here? I think that was average hourly earnings growth over the month. Right. It was. It was that, but that is not that that, no. Yeah. Right. Okay.

That's too easy. Yeah. Does it have something to do with wages? It does not, no. Is it from the payroll survey? Nope. Houseful? Nope. No. Jolts. Oh. No? Is it labor market related?

It's labor market related, yeah. He said that like maybe it's tangential. Yeah. Is it, it's not wage related? No. It was released this. week. I'll give you that. Oh, it's productivity growth. Oh, yes. There you go. Right.

His favorite. I had to, you know, had to throw it to Chris. You know, got a low numbers. We might as well highlight that just like when we get high numbers, you know. Yes. Indeed. Indeed. You want to explain? Sure. So peripity growth was a 0.3% annualized in the first quarter of the year. That was revised down from the preliminary estimate, which was 0.8%.

I think Q4 was one one and a half, 1.6%, I think. So pretty dramatic slowdown from productivity growth. We saw very strong productivity growth in the middle of last year and even through the end of last year was still pretty strong. This is the weakest reading we've had since the beginning of 2025. And again, I think again, productivity growth had been strong largely because the labor market had been very weak, right? We were getting some output growth, but you had very weak employment growth. And so that netted out to strong productivity growth. Now that story might be changing a little bit, right, if the labor market is starting to pick up,

if you're adding a lot more jobs, particularly in Q2 and moving forward, and output growth is still fairly weak, that, you know, likely leads us to much weaker readings of productivity growth here, you know, in the next couple quarters, if those trends hold. It feels like it, right? Because our Q2 GDP tracker is coming in south of 2%. I mean, so real GDP, I think it's like one, one six or one seven percent, you know, very low. And then you throw in this much stronger job growth. That would imply unless something really changes in the month of June, that would imply a pretty weak, another weak productivity number, right?

Yeah, that's the way I'm thinking about it, yeah. Yeah, right. Although year over year, it's still pretty healthy. I mean, even if you smooth out some of the volatility, it feels like it's still around 2%ish, you know, productivity growth no that's right i mean you had some very strong readings in the middle of lat you know it was up to i think over five percent you know annualized in q3 last year so i mean it's still you know healthy on a year-over-year basis but the one thing you i think we can stay with confidence i mean ai is starting to have an impact on things but it it doesn't feel like it's really in any meaningful way

lifted productivity growth at least not yet doesn't feel like it would that be fair i would agree with that yeah you would okay yeah well the john We've seen as in low productivity industries, right? Regional hospitality, government, yeah, health care. Right, although we saw, we were seeing job loss and financial services, so that would lift. That would apply. Yeah, would lift, yeah, okay. All right, that was a good one.

Chris, do you want to give us a last stat we're going to do? What's your stat? Sure, it's a tough one, so I'm going to, okay? I'll guide you. It's from a research paper that came from New York Fed. 64% is the number. It is labor market related. I did circulate this paper.

Yes. Yeah. To share the youth unemployment that can be attributed to remote work. Ding, ding, ding. You got it. Well done. Whoa. Whoa.

Did you send this paper to me? I don't remember seeing this paper. Oh, maybe I left you off. Oh, geez. No respect. Do you see that? No respect. Wow.

She didn't know. Yeah. But it's in the Wall Street Journal. I'm sure you saw it. This is a plant. You sent it to everybody but me and say, oh, this is hard. Jeez. So what was the thing on that?

What was the stat? 64%. But what is it? This research found that 60, that the, um, increase in youth, well, no, young college graduate unemployment in the post-pandemic era. Remember that, you know, unemployment for recent college grads has fell a bit, but it did not fully recover to the 2019 levels, right? Young college graduates have been having a hard time finding work and their unemployment rate has been higher than it was previously. There's been a lot of chatter about, of course, AI being the main culprit.

But this research suggests that it's actually more remote work as a primary factor here. In part just because of the timing, right? This has been going on even before Chad GPT came on the scene. And the paper had some interesting points about just mentorship and the preferences that firms may have for hiring more established workers versus a young college graduate who, you know, may not be easily trained. in a remote work environment. That's fascinating. That's a fascinating result. Yeah.

I'll send you the paper. Yeah, yeah, please. Yes, please do. Yeah, I consume everything you sent. All you guys send me all the time. I consume it. Huge consumers of that information. You keep firing away.

Okay. I think we're going to call it a podcast, unless you guys have something else you want to chat about before we do. Dante, anything, any last parting words? No, I think we're shocked and confused. and confused, shocked and confused. Chris, dazed and confused, dazed and, and mrs.

And mermaids and numb, uncomfortably numb and confused. All those emotions describe me very well. All of them. I can't, I think I might, I think I like the dazed and confused more. But definitely that's going to be our title, one of those three, for sure. sure, unless Sarah vetoes, which she, you know, she often does. She's such a dictator. Have you noticed?

No, sorry, she on the line. Oh, she's here. Oh, okay. Only kidding. We're getting a little punchy. It's late on a Friday afternoon, a long week. Okay. With that, guys, have a nice weekend.

I hope the listener does as well, and please join us next week. Talk to you soon. Take care now.