Nick Bunker from Indeed joined the podcast to break down July’s surprisingly weak employment report. The team put forward their favorite interjections before breaking down the report into causes for concern, potential measurement issues and (a few) reasons for cautious optimism. The discussion turned to the “Sahm Rule” as the group pondered whether the recent rise in the unemployment rate is signaling a recession. Mark and Marisa both claimed victory in the Statistics Game before the podcast ended with the team putting forward their current recession probabilities.
Nick Bunker from Indeed joined the podcast to break down July’s surprisingly weak employment report. The team put forward their favorite interjections before breaking down the report into causes for concern, potential measurement issues and (a few) reasons for cautious optimism. The discussion turned to the “Sahm Rule” as the group pondered whether the recent rise in the unemployment rate is signaling a recession. Mark and Marisa both claimed victory in the Statistics Game before the podcast ended with the team putting forward their current recession probabilities.
(4:00) July Employment Report
(22:58) Labor Market Issues
(51:00) Stats Game
(1:02:50) Recession Risks
Today's Guest: Nick Bunker, Director of North American Economic Research at Indeed
Follow Nick Bunker: @Nick_Bunker
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' @MarkZandi, Cris deRitis on LinkedIn, and Marisa DiNatale on LinkedIn
Mark Zandi: Welcome to Inside Economics. I'm Mark Zandi, the chief economist of Moody's Analytics, and I'm joined by a bevy of folks. We've got my two trusty co-hosts, Cris deRitis and Marisa DiNatale. Hi, guys.
Cris deRitis: Hey, Mark.
Mark Zandi: Good to see you.
Maris DiNatale: Morning.
Mark Zandi: We typically start these podcasts with a little bit of chit-chat, but my son says he doesn't like the chit-chat. And the reason he gave me was, "You guys are just boring."
Maris DiNatale: No.
Mark Zandi: I'm not kidding. I'm not kidding.
Cris deRitis: That's harsh.
Maris DiNatale: Are we all boring? Or maybe just one of us?
Mark Zandi: Actually, he did single out one person. I'm just not going to tell you who that is. It could have been me. It could have been me. I'm just saying so, but anyway.
Cris deRitis: Ouch.
Mark Zandi: Alex is right. We're going to have to upgrade our chit-chat, but no chit-chat today because this is Job Friday. This is August the 2nd. And I'm just going to say one word before I bring in the other guests. The word is ugh, U-G-H exclamation point. Ugh. Anyway, we also have Dante. Dante, you come on every Jobs Friday. Good to see you.
Dante DeAntonio: Hi, Mark. I don't think we can be accused of being boring on Jobs Friday. I don't think that's possible.
Mark Zandi: I don't know, Dante, we live in a bubble, I think. But I love your shirt. I love your vacation shirt.
Dante DeAntonio: Thank you.
Mark Zandi: Thank you for joining us on your vacation.
Dante DeAntonio: No problem.
Mark Zandi: Yeah. How's the water at the New Jersey Shore? Oh, no chit-chat.
Cris deRitis: There you go. It slips right in.
Mark Zandi: That's right.
Dante DeAntonio: One word. Cold.
Mark Zandi: Cold. That's one word, cold.
Dante DeAntonio: The one word you need.
Mark Zandi: Very cold. Yeah. Very good. And, Nick, Nick Bunker, you're back. Good to see you, man.
Nick Bunker: Glad to be here. And glad that my sons are not old enough to give feedback on my conversation on the podcast because I think I might get similar.
Mark Zandi: Just you wait, new friend.
Nick Bunker: Oh, yeah. Oh, I'm sure. I'm sure I'll get the same feedback. So I'm just biding my time until then.
Mark Zandi: How old's your son?
Nick Bunker: I have two and they are turning two later this month.
Mark Zandi: Well, you got one more year grace period, and then the criticisms begin, about three and a half.
Nick Bunker: Yeah, well, I'm-
Mark Zandi: "Dad. Really, Dad?"
Cris deRitis: That's fair.
Mark Zandi: "What are you out of your mind?" They don't say that at three and a half, but that's what they're thinking.
Nick Bunker: Yeah, we'll get there. We'll get there.
Mark Zandi: Okay. All right. And, Nick, you're at Indeed. Oh, you've been on a couple of times. At least three, maybe two, three times.
Nick Bunker: I think this is my third time.
Mark Zandi: Third time. Excellent. Good to have. Do you want to just give the audience kind of a brief bio and anything you want to say about Indeed?
Nick Bunker: Yeah, sure. So I am the director for North American Research at the Indeed Hiring Lab. The Hiring Lab is Indeed's economic research arm. We're in the fortunate position where we get to use all the proprietary data, or at least a large majority of the proprietary data at Indeed, to understand what's happening in labor markets. So that's in the US, Canada. We also have economists in Europe, in Japan, Australia as well.
So we get to use data on job postings to come up with metrics like our job postings index, to tell folks what's happening to sort of state of demand. We also have measures of posted wages. And then we do bespoke longer research projects on a variety of topics. So we keep our eye on the labor market in a variety of ways, including the government data that we get across markets.
So excited about digging into not only the jobs report we got this week, but all the data that came out earlier. And then any of the data that Indeed has that can shed some additional light.
Mark Zandi: Yeah, very cool. We're very lucky to have you on Jobs Friday. This was a labor market data dump this week, wasn't it? We had lots of stuff come out. And I had this one word to say, ugh. U-G-H exclamation point. Ugh.
Okay. Let's dive right in. And I'm going to begin with Dante. Dante, you want to give us the nuts and bolts of the report? The jobs report for today?
Dante DeAntonio: Sure. I mean, I'll share your top line sentiment of ugh. I think it was hard to find much to in the report today. Headline job growth, 114,000 in July. Private sector job growth came in just under 100,000. Three month average job growth didn't really change all that much, mainly because we had a really weak reading in April as well, which rolled off the three-month average.
So we went from one ugh in April to another one now, which left average job growth, right about 170,000. Downward revisions, again, not nearly as big as they were last month, but still it seems to be that not only is headline growth weakening, but revisions to prior months are also causing things to look a little bit weaker than they had originally. The industry composition, again-
Mark Zandi: Can I ask a quick question?
Dante DeAntonio: Yeah.
Mark Zandi: I always ask this. Underlying monthly job growth. So abstracting from the vagaries of the data this month we had a hurricane that blew through and kind of maybe scrambled things. Seasonal adjustment. Abstracting from all of that, what do you think underlying monthly job growth is?
Dante DeAntonio: I think at best, 200,000. At worst, 150. So somewhere in that range.
Mark Zandi: Okay. 150 to 200K?
Dante DeAntonio: Yeah.
Mark Zandi: Okay. All right, go ahead.
Dante DeAntonio: Industry-wise, it was not a whole lot of new news. The composition was still largely the same, sort of everything throttled back for the most part. So you still had healthcare, legion hospitality, public sector, driving the majority of job gains. Construction also, getting a little bit stronger here as the year wears on.
Not a lot else to write home about industry-wise. Information was the one big loser in July, dropped 20,000 jobs. Other than that, there weren't any sort of major declines. And not a whole lot of action on the industry front.
On the house. Wages also came in weaker again in July. We had a week reading in June, another week reading in July. They were up 0.2%. Year-over-year growth is now down below 4%, it's at 3.6. So I think we can sort of firmly take concerns.
Mark Zandi: Average hourly earnings are 3.6% year-over-year through the month of July?
Dante DeAntonio: Yeah, that's right.
Mark Zandi: Okay, which feels like where you want them to be, right? Not too hot, not too cold.
Dante DeAntonio: Yeah. I certainly think any concern that wage growth was going to be a problem in terms of inflation should be pretty much gone at this point. It doesn't feel like... If anything, it feel like we're going to go the other direction here, not if the labor market's going to get hot again.
Mark Zandi: Right.
Dante DeAntonio: Average weekly hours ticked down a little bit. That likely the hurricane impact probably plays a role there. But we've seen that be weak here now throughout 2024. So I'm not overly surprised that it has stayed weak at this point.
The household survey didn't bring a whole lot else to be excited about. Maybe a couple of bright spots sort of mixed in with the bad news. The headline is the unemployment rate rose for the fourth straight month. It's up to 4.3%. We did get a big gain in the labor force. Labor force grew by over 400,000. Labor force participation ticked back up again.
Prime labor force participation, so for 25 to 54 year olds, actually jumped quite a bit, an increase for the fourth straight month. It's at its highest level since 2001. So if there is maybe a positive story there, it's that we're continuing to see a strong level of participation and employment among prime age workers.
Outside of that, employment was up a little bit. The number of unemployed workers did jump quite a bit, so there's not a whole lot on a positive front on the household survey either. It's sort of a sign of weakening across the report as far as I've seen.
Mark Zandi: Okay. So would you characterize the labor market, broadly speaking, how would you characterize the labor market? I mean if I asked you that question six months ago, you would've said strong, I assume. Now how would you characterize it?
Dante DeAntonio: Somewhere in between strong and weak. It's moderating for sure now.
Mark Zandi: That's a good one.
Dante DeAntonio: I definitely wouldn't call it strong anymore, but I don't think I'm ready to call it weak just yet. I think there's enough uncertainty around this month that I'm not believing that underlying job growth is 115,000 at this point either. So I'm not sort of all the way in the camp that thinks we're headed to zero and below immediately. But I think it's definitively weaker than it was six months ago, in my mind.
Mark Zandi: Got it. Okay. Hey, Nick, what do you think? First, there's a lot in that report, so if you want to flesh out anything that Dante said, or add anything on the numbers, that'd be great. But also I'm very interested in your just general impression. Are you in the ugh camp as well?
Nick Bunker: So I wouldn't say ugh. I will say yikes maybe to add a letter there, just for a little bit of variation.
Mark Zandi: I really like that. Yeah.
Nick Bunker: Yeah. I think literally the first word out of my mouth when I refreshed the PDF this morning was, "Yikes." And then maybe some words after that that you don't want to say on a family-friendly podcast.
Mark Zandi: Right. We're not that friendly, Nick. Just saying, so buckle in.
Nick Bunker: Noted. But I think like Dante, I see the labor market as materially weaker than it was a few months ago, but I wouldn't say it's weak right now. So I think if you look at this report at the current state of the labor market as it is in July of 2024, it's in a decent spot. You are seeing employment to population ratio for prime-age workers at levels we saw in the early 2000s, like 2001, when that was a strong labor market.
Wage growth has slowed down, but it's still in a vicinity we saw in 2019, was pretty strong and inflation's trending down. So real earnings are picking up. And a rise in unemployment, 4.3% unemployment, is higher than we've seen recently, but it's pretty low by historical standards.
That all being said, everything else in the report, it's the direction that's really concerning. That everything is not only weaker than it was six months ago, the outlook is it looks like it will continue to weaken. That payroll gains are still relatively healthy.
Mark Zandi: Good point.
Nick Bunker: But the diffusion index went below 50, and that's also been sort of slowly trending down over time.
Mark Zandi: Can you just explain that very quickly for the listeners? The diffusion index?
Nick Bunker: So the diffusion index is a metric that essentially tells you what percent of industries are adding jobs in a given month. So if it's 50 or above, that's a sign of broad base expansion. If it's below 50, that means most industries or sectors are subtract, not adding jobs. So that dipping down not only means that the magnitude of job gains that we're seeing is lower, but also the scale of the job gains is smaller as well. There's just fewer industries adding jobs.
In addition to that, the unemployment rate is just still creeping up. I think before this report, lots of people game-planning, does this thing rise above the Sahm Rule indicator? What does that mean? And you can have those conversations, but I think the underlying reality is it's going up. The trajectory is still in the bad numbers going up. That is-
Mark Zandi: Listener, we'll come back to the Sahm's Rule because that got triggered I think, right, Marisa?
Maris DiNatale: Yeah.
Mark Zandi: It got triggered. So we'll explain that and what it has meant historically.
Nick Bunker: So I think, look at the unemployment rate, you look at other metrics as well. It's clear the labor market is, it's weakening, whether or not it's weak. It's not weak right now, but it's trending. The direction is weakening.
I think the question right now is it's clearly still weakening. It's unclear whether it's that weakening is accelerating or just staying the same. But to my mind, regardless of that, the outcome of that debate, it's still not leveling off and it's not getting better. So trajectory is still towards weakness. And if that keeps continuing, eventually it's a weak labor market. And eventually whether or not it trips any specific statistical line, that's a recession for many people.
Mark Zandi: Noted. So you're yikes and all the trend lines look a little disturbing. Certainly not what you'd want to see. But let me ask you this, is there anything in the report that you came away with, "Oh, that's okay. That's pretty good." We mentioned wage growth as one of the indicators that my guess it's okay. It's not too hot, no cold. Anything else in the report that seems okay to you?
Nick Bunker: Yeah, so I will say, I mentioned it briefly, but the share of workers ages 25 to 54 with a job, that picked up. That's at levels consistent with the strong labor markets. That's good to see. I think that is one of the things that if that sort of merely... That has leveled off, essentially. It's up over the month, but over the last year it's kind of flat-lined.
And that is a sign that, hey, the labor market isn't deteriorating in a major way because the share of the people in that age group with a job is holding steady. So at least it's not trending down. So that is good to see.
And I think within the sector data, Dante referenced this, but construction employment's still going strong. That was good to see that. That's been an interesting sector over the past two years just because traditionally quite interest rate sensitive, but it's held on to it. Seeing it still adding jobs was a positive indicator for me.
Mark Zandi: And maybe I'll just ask you one more question before I move on to Marisa and get her views. You were talking about the trajectory here, and unemployment's rising, job growth is slowing. If those trends continue, then the job market goes from whatever it is to weak. It'll be weak here in not too distant future.
What do you think? I mean, do you think the trend lines will? Do you think we will stabilize here? Or in the next few months? Or do you think the trend lines will continue to move in the untoward direction?
Nick Bunker: I guess I'm hard-pressed to think of any signs of anything that would change the current trajectory. And in a positive direction, what would slow things down? Maybe it is there's expectations of interest rate cuts today, sort of brings the mortgage rate down a lot and then that props the housing market up and that there's some knock-on effects there.
Or maybe the consumption stays strong because inflation's come down enough that people feel good enough to keep up their consumption. But I've been thinking this morning about what's going to make me feel better about these trend lines? And I'm drawing a blank right now.
Mark Zandi: Okay. All right. Well, at the end of the conversation I'm going to ask everyone for their probability of recession starting at some point in the next year. Sounds like you're going to be on, you're on the negative side of that. But don't tell me. Don't tell me. We'll come back to that.
Nick Bunker: I will say the trend in my recession probability has been up over the last period of time.
Mark Zandi: Marisa, anything you want to flesh out in the numbers that we haven't already covered? And what's your kind of take on things?
Maris DiNatale: I think yikes and yeesh and wow are all-
Mark Zandi: No ugh?
Maris DiNatale: Good words. Ugh. Yeah, I forgot about ugh. All of these words, yeah. I'm worried because I too believe that this is not a weak labor market, but it is weakening and it appears to be weakening pretty rapidly. And it's not just the jobs report. We got the JOLTS Report this week. Hiring is now at its lowest rate in 10 years. This is the final step employers take before they start laying people off.
I mean, so if I'm looking for bright spots in the report, then I go to the household survey, which you guys have already said, labor force participation for prime age workers, basically back where it was at the peak, the all time peak of the labor market in the late 1990s, 2000s. That's great. We still don't see a lot of evidence of layoffs.
I mean I know unemployment insurance claims have trended up, but there might be seasonal things going on there. It looks a lot like it did last summer. It's not in this flashing red territory yet. And we don't see evidence of layoffs from other surveys, so we don't see it in this report. And we didn't see it in the JOLTS Report either.
We didn't see it in the Challenger Report, which we also got this week. Challenger layoffs, and this is the Challenger, Gray & Christmas. These are announced layoffs, they scrape it from the web. And those were actually down compared to where they were in the month prior. So we're not seeing employers laying off, but we're seeing them pull back on hiring very, very sharply.
Dante went through some of the industries. We also got losses, I think, Dante, you said the information industry had a job loss. So did professional business services, headline. And that wasn't just temp help, it was other stuff. Within financial services, banks lost jobs over the month in all kinds of banking. Not investment banks, Wall Street type stuff, but bank-banks, people that lend money.
A lot of our clients saw big job losses over the month. So I don't like what I'm seeing on the payroll survey side. And we didn't really say much about the hurricane. I'm not convinced there's much of an effect from that hurricane on this data, but we could talk about it.
Mark Zandi: The BLS actually in the-
Maris DiNatale: The BLS said there was not. If you look at the survey response rates, on the household survey side, the response rate was exactly what it was last July, 70.9%. But on the payroll survey, the response rate was 56.9%, which is the lowest it's been in any July since 2001. It was significantly lower than what it normally is in a July.
Now BLS may not be worried about that because that's still within their margin of error, and it's not affecting the statistical significance of these gains and losses, but there could be some effect there. And if I look across industries, the only industry where I could say maybe this is a hurricane effect, is there were some losses within transportation. Like ground transportation, passenger transportation, things like that, that could conceivably be affected by the hurricane.
But I don't think we can blame the jobs. Because I've seen some of this in the press this morning, blaming some of the weakness on the hurricane. But I don't think there's a ton of evidence to suggest that that's the case.
Mark Zandi: Well, I mean if the survey response rates lower, that maybe hurricane related.
Maris DiNatale: It could be, one survey.
Mark Zandi: So wouldn't that potentially bias the number to some degree? And then there's also a number in the report. I didn't look at it, but maybe someone else has. The number of people who said... You have, Nick? That's the number of people who say they aren't working because of weather.
Nick Bunker: Yeah. Who said that they have a job, employed, but they're absent due to weather. And also I think there was also usually work full-time but working part-time instead. So I think that's in the household, is what I've seen people point at.
Mark Zandi: Right.
Nick Bunker: Also some temporary layoffs as well.
Mark Zandi: And they were elevated? Those numbers were elevated?
Nick Bunker: Those numbers were elevated.
Mark Zandi: Okay. So there is some evidence that maybe this thing is overstated.
Maris DiNatale: But if it is, I'd say maybe it's like 10 to 20,000 jobs. So I don't think without this, that you'd get a plus 200K number on the payroll survey. I think-
Mark Zandi: Just for everyone out there, we're referring to that hurricane that hit Texas and-
Maris DiNatale: Houston, right.
Mark Zandi: Knocked things out for a while during the survey.
Maris DiNatale: And then the only other things I wanted to say is if you adjust the household survey to the payroll survey concept, you would've got a 19,000 K decline on the household survey. And that's mostly because if you look at the composition of job growth as reported by the household survey, a lot of it was among the self-employed.
So private wage and salary workers and government workers, I think they both fell, if I'm not mistaken, over the month. It was really the self-employed that boosted the top line number. So when you take out the self-employed, because they're not included in the payroll survey, you would've got a negative print on household employment.
Mark Zandi: Got it. Okay. So, Cris, I know three keen observers have given their take on the report and given their review, I'm not sure there's anything left to pick over on the report, but if there is, I'm all ears. But most interested in your perspective on this and maybe you have another description for how you feel. Yikes. Ugh. Wow. Yeesh. I think that was-
Maris DiNatale: Yeesh.
Mark Zandi: Yeesh. Yeesh. Got it. With an accent aigu. That accent aigu? I don't know what that means.
Maris DiNatale: I don't know what that means either.
Mark Zandi: It's French for something. I don't know. It sounds right. Anyway, I'm trying to lighten things up because my son says we're boring. How can that-
Cris deRitis: Yeah.
Mark Zandi: That wasn't boring, was it? That didn't feel boring to me.
Cris deRitis: How about a-
Mark Zandi: Rolling his eyes, but okay.
Cris deRitis: How about a hmm?
Mark Zandi: Hmm. I like that.
Cris deRitis: Yeah. Let me offer everyone listening to the podcast, a brief moment of silence. Take a deep breath and there's a lot of, a lot going on. I'd also advise not to look at your retirement or 401k statements this weekend. Go for a walk.
Yeah, it was a bad report, but we identified some of the positives here. And I guess my point, the main point I would make, is this is not terribly unexpected. We were expecting slowing, we were expecting to see things come in. We were actually surprised at the upside a few months ago.
So yeah, not a good report certainly, but I don't see this as the end of the world either. There's more script to be written here. So I think I'll offer that perspective check.
Mark Zandi: It's so funny, Cris is taking the-
Maris DiNatale: All of a sudden he's the optimist.
Mark Zandi: Yeah. All of a sudden he's-
Cris deRitis: Look at this.
Mark Zandi: How did that happen? What? What's going on here?
Cris deRitis: This is going right to my script.
Mark Zandi: Yeah. Oh, I see. Yeah, right. Exactly. Exactly. Okay. So you're saying you shouldn't have hair on fire here. I mean obviously this was not what you would want to see, but you get bum reports all the time. So this might be just a bum report.
But to Nick's point about the trend lines, I mean that resonates with me. I mean, I look at the unemployment rate, that's the one that really bugs me the most. I mean unemployment rate's 4.3, 4.3%. Objectively that's very low by historical standards. But that's up, what a percentage point from a year ago? Or pretty close to a percentage point.
Maris DiNatale: 0.9 I think.
Mark Zandi: And it is rising now pretty significantly. And now we're going to get into the Sahm's Rule. Historically, when you see increases in unemployment on this quickly over this kind of a period, it suggests that the economy is really already sucking wind. I think the intuition behind it is, at least my intuition behind it is, once unemployment starts rising, that makes people nervous. Consumers start to become more cautious.
And there's indications of that particularly now among lower income households. And that then causes businesses to pull back even more on their payrolls. Layoffs start to climb. And you get into this kind self-reinforcing, vicious, negative cycle that takes on a dynamic of its own, a life of its own. And that's a recession. In fact, what's the Sahm's Rule exactly? Marisa, what's the exact formula?
Maris DiNatale: It's the three-month, when the three-month moving average of the unemployment rate rises half a percentage point or more from its low point in the prior twelve-month period.
Mark Zandi: Okay. So we're 4.3, that's if you take a three-month moving average or probably 4.1 or something like that.
Maris DiNatale: Yeah.
Mark Zandi: We're up more than a half a point from the low point on three-month moving average over the past 12 months. So that rule has now been triggered. That regularity in history has been triggered.
Maris DiNatale: It went from the indicator was 0.43 in June and it jumped to 0.53 with this morning's report. So we're above that 0.5 threshold now.
Mark Zandi: We had been talking about this via email over the week, and the one thing I did not completely understand is that the Sahm's Rule is not a leading indicator of recession. It's actually a lagging indicator. Once it actually is triggered, historically, you're already in recession by a couple, three months, right?
Maris DiNatale: Yeah.
Mark Zandi: Okay.
Maris DiNatale: And I think that's been kind of widely misunderstood when, up until today, well, I haven't Googled it today, but there's so many articles about the Sahm's Rule. People on TV talking about it, talking about it as a forecaster, a leading indicator of recession. It's not.
And if you look at the original Sahm's Rule that Claudia Sahm herself says, and she's been doing a lot of interviews, telling everybody to calm down, this is not what it is. It's not meant to be a leading indicator. It's just that she observed over, and it's true, you can go back to business cycles since the 1950s and '60s. That when this triggers, you're typically about three months into a recession already.
Now you don't know you are because we don't know the official start and end dates of a recession until the National Bureau of Economic Research dates this a year after the fact. So if we go back and we look at these official start dates of a recession, the Sahm's Rule typically has triggered several months into a recession. So that would suggest, if this was true today, it would suggest that the economy went into a recession a couple months ago.
Mark Zandi: Right. And that's the value of it. That is when the economy's going into recession, it generally very difficult for people to actually know that at-
Maris DiNatale: Right, at the time.
Mark Zandi: In the fog of the data. And this so-called rule, this regularity historically, would cut through the fog and says, "Okay, you're in." So if this rule holds today, we're already in recession. Okay.
Maris DiNatale: That's right.
Mark Zandi: What do you think of this Sahm's Rule? And what do you think of it in the context, the current context, is that it's saying we're already in recession?
Nick Bunker: So I've been a person who has been watching this once the unemployment rate started slowly drifting up in 2023. I was like, "All right, I'm just going to, like every month this report comes out, check the indicator." And I think it's very useful in part because I think it gets at the underlying point of that the unemployment rate, when it starts rising, it doesn't just usually rise a little bit. It has some upward momentum to it.
So I think this gives you an indicator of not just that it's rising, but the speed at which the unemployment rate is rising. The fact that it's like the three-month average has to be at least a half point or so. It's both at a speed and a timeframe of how quickly and large those increases are.
So in terms of the current economic situation, whether it means whether or not we're in a recession, I am thinking a little bit of two years ago when we had two consecutive months of negative GDP growth, which it's not as back-tested as Claudia's research with the Sahm Rule, but that's usually a indicator that you have entered a recession.
And I don't think we entered a recession then. We were flirting with it. And I think if you take the Sahm Rule at its face value to say, "Oh, you're in a recession two months ago," I think about the other data that we have that you usually use to determine a recession, all those indicators are not indicating that we're in a recession. GDP growth, not only was above 2%, it was an upside surprise.
Personal disposable income is still growing. All of those indicators are pointing it. So I think the way I'm taking it right now is whether or not it's a recession, or if a year and a half from now, the fine folks in NBER when they meet in Cambridge, will say, "Oh, no, it was a recession." I think right now we can tell the labor market's clearly deteriorating. And maybe the overall economy's not in a recession right now, but the labor market's definitely feeling more pain than it did before and it's trending in that direction.
Mark Zandi: Yeah, my own sense is that the Sahm's Rule is not going to work this go-around, for two reasons. I don't think we're in a recession. First reason is it feels like the increase in the unemployment rate is mostly because of labor supply, not decline or weakening, significant weakening in labor demand. And the labor supply goes to the strong influence of immigrants into the country.
And we know there are a lot of immigrants that come in, they apply for work authorization, they get that within I think nine, 12 months, and then they're working. And we've seen that in a lot of different industries where immigrants tend to work. That labor demand is probably weakening as well.
You can see it in, I think you kind of feel it in hiring rates. You can see in hours worked, maybe temp help. So I'm not saying it's not, we haven't seen some weakening in demand. But if I said, "Okay, the unemployment rate's risen a percentage point over the past year," my intuition is that two thirds, three fourths of that's labor supply, one third, one fourth of that's labor demand.
The labor demand side of this is exactly what the Fed would want, right? I mean because unemployment was, from their perspective, too low. They suggested the economy was beyond full employment, too tight, generating wage and price pressures. We want to cool things off, we want unemployment rate to rise. And they pretty much got that. And what the rest of what we've gotten here is just labor supply. And I don't view that as being as serious a worry.
The other thing is, the best coincident indicator of where the economy is in the business cycle, by orders of magnitude, is jobs, it's payroll jobs. As Dante said, I think we're probably closer to 150K, after abstracting from everything in all the revisions. But 150K is still, that's good. I mean that's not a problem.
I mean that's exactly, they may be even on the high side of where you want to see it, if labor supply wasn't as strong as it is. So I don't know. I don't feel like the Sahm's Rule is going to work in a meaningful way this go-around. Nick, any reaction to that? And then I'll see if the others have a reaction as well.
Nick Bunker: On the supply side, I tend to agree with parts of that. Yeah, a lot of this has been supply side increase, like more people engaging with the labor force. I think in some ways, today's report's a little microcosm of that. Where you did see, yes, the unemployment rate ticked up, but the labor force participation rate did as well.
I think, and I hear you on the immigration side, I think one caveat I would add to that is that a lot of, or a decent chunk of the increase in the unemployment rate the last year plus has been in entrance and re-entrance to the labor force. So I think one line of thought there is, "Look, this is just more people entering the labor force." It's a more benign form of unemployment rate.
And I think my response to that would be, "Yes, it's good that they're entering, but the fact that they're not finding a job and also that we're seeing fewer people come from not into labor force, right into employment." That's a sign to me that, okay, yeah, people are still looking for a job.
So it's tight enough that people are still looking for a job, but maybe the tightness is loosening enough that when they do go to look for that job, they're not finding it as quickly. So I think it is, that measure I think is hard to untangle how much of it's supply and how much of it is demand.
Mark Zandi: What about my point on payroll employment? That is a coincident indicator.
Nick Bunker: Yeah. So I think part of the logic for liking the Sahm Rule is that the household survey doesn't get revised, the payroll survey does. And I think particularly in the period where there's been uncertainty in some corners, and straight up mistrust in others, about the revisions to the payroll number, that people have emphasized the household survey.
But I think I'm more in the camp of, I would lean towards the payroll survey more for underlying employment growth, which is why I don't think we're necessarily in recession right now because we're still adding jobs. Job openings and job postings are still elevated. So there's still some demand there if it's come down.
So I think that I'm taking, I'm drawing signal from the rise to unemployment rate. But I'm not saying, "Oh, the resident unemployment rate means the payroll survey's much lower than it actually is." So I would say I would agree that the current pace of payroll growth, 150-ish let's say, does suggest we're not in a recession.
But to go back to my other one of the day, which is trend, the trend is down. So it doesn't look like the payroll is going to pick up anytime soon. I don't see any signs of a clear leveling off there either.
Mark Zandi: Okay. Well, the other data point I wanted to throw out there to consistent with the idea that this might be more labor supply than demand, is the unemployment rate increase has been most significant among less educated workers. High school degree, less than high school degree, where you would expect the immigration to have its biggest impact, if that's in fact the factor.
What do you think, Cris? Do you have a view on this? On the Sahm's Rule? I mean the Sahm's Rule says we're in recession, so what do you think?
Cris deRitis: It certainly doesn't feel like recession right now. The other indicators, and I understand there's cloudy data here, but doesn't seem like that's been the case for two, three months at least, if that's the yardstick here. So I agree with you. I think this go-around it may not be right, or maybe take more to trigger. Maybe it's a little bit more delayed than typical.
Mark Zandi: Yeah. Dante, any other views on that?
Dante DeAntonio: No, I would agree. I think we talked about this earlier this week in the email thread that you mentioned. If you look back historically at business cycles where the Sahm Rule was triggered, in most cases when it triggers, three-month average job growth is already negative, which clearly is not the case today.
If you look at the couple cases where it wasn't already negative, in one case, there had been negative prints of job growth in the six months leading up to the Sahm Rule triggering. It just wasn't negative at that moment. And then in the other, I think two cases, within a month or two after the Sahm Rule triggering, job growth went negative. And it just doesn't feel like we're headed to a negative print on top line job growth in the next month or two.
I think again, if underlying job growth is 150K, it feels unlikely we're going to get to zero here in the next month or two. I mean, the bottom could fall out of everything, but with layoffs holding in check, it feels harder to imagine that threshold getting crossed that quickly.
If layoffs were on the rise already, then I'd be less sure that that couldn't happen here over the next couple of months. It just feels like the underlying fundamentals of the labor market are stronger than they usually are when the Sahm Rule gets triggered.
Mark Zandi: As part of that email conversation, we also had a bit of a back and forth on, well, okay, what is a good leading indicator? If you had to pick one leading indicator, meaning you want something that gives you a signal before you're into the economic abyss. Marisa, do you have a perspective on that? I know you have a perspective because you were on the email chain, but just for the listener.
Maris DiNatale: Right. Yeah. I mean, the other component of the Sahm Rule that we've always seen is that if you look at unemployment insurance claims, they are always trending up when the Sahm Rule triggers. Always, that's never not been the case. There have been cases where a recession started officially and UI claims were not yet trending up. So that happened in the 1980s, but that was maybe the only time that that was the case.
But every time the Sahm Rule has triggered, UI claims were up compared to where they were the year before. We're not there. I mean, on the four-week moving average of UI claims, I think we're up a little bit, Dante, we're up maybe 1%. But if you look at, I was looking at monthly UI claims just to put them next to the unemployment rate, and we're still down as of the last monthly reading on UI claims.
So unless we see a very large and unexpected spike in claims over the next couple weeks here, it doesn't fit with that either. And that goes to your point about payroll employment growth, right? I mean, if people aren't losing their jobs en masse, then we wouldn't expect to see a negative print on payroll employment. And there still is not much evidence that layoffs have picked up materially.
Mark Zandi: So just in case you didn't say it, I missed it. UI, for folks out there, unemployment-
Maris DiNatale: For me, it's always been... Yeah, because it's-
Mark Zandi: Unemployment insurance claims.
Maris DiNatale: Unemployment insurance claims. It's weekly data. It's administrative data, it's not coming from a sample or a survey. No one's estimating it. It's account of people filing claims for unemployment insurance.
It's one of the highest frequency labor market data points that we get. Probably the highest frequency one. So you can see every week what's going on. Now these get whipsawed. I don't know if we saw yet the impact of the hurricane and UI claims. I think we did. I think they were up in Texas. So things like that will affect UI claims.
Mark Zandi: Also, seasonal adjustment, it's hard to see because-
Maris DiNatale: Sure, and there's seasonal adjustment. But when you smooth it out and you look over several weeks, you can really see what this trajectory in layoffs is.
Mark Zandi: But they're up. They're up.
Maris DiNatale: It's not seeing that.
Mark Zandi: I mean-
Maris DiNatale: Yeah, they're up.
Mark Zandi: Meaningfully up. I mean, pardon me if I'm wrong, but you go back, I don't know, a couple months ago, they were hovering around 200,000 per week. Last week on a four-week, removing average basis, we're at what, 240? Something like that?
Dante DeAntonio: 238, yeah.
Mark Zandi: 235, 240.
Maris DiNatale: Yeah. Now-
Mark Zandi: It's still low. Again, seasonal adjustment might be the issue here. And you said the hurricane, that would definitely have an impact, but nonetheless.
Maris DiNatale: They are up, no doubt about it. But the same thing happened last summer. So we saw this increase in UI claims over the summer months in July and into August. And they were up more actually last summer than they are now I think.
So that suggests there may be some seasonal adjustment issue going on. I'm not totally buying into that 100%. I mean this could very well be signaling that layoffs are happening and they're just not showing up in the survey data yet. Dante looks at this very closely.
Dante DeAntonio: There's been a couple one-off issues too, not issues but one-off events that have, I think probably boosted claims a little bit. So if you look at the state data back in the middle of June, there was a law change in Minnesota that allowed teachers that were hourly, I think hourly teachers to file for unemployment when they're out of school for the year. And so you saw a jump in state claims in Minnesota sort of coinciding with the end of the school year in early to mid-June.
And then when you look at late June into early July, you always see a spike in Michigan when factories close to retool in the summer. But that spike was bigger by a pretty significant... Normally you see claims jump to like 10, 11,000 in Michigan in a couple of weeks in July. It was up to like 17, 18,000.
So none of these things are... And then the hurricane obviously after that. None of these are huge by themselves, but you're only talking about claims moving from an average of 215 to 240. So five, 10, 15,000 of an impact from any one of those things can explain probably at least half of that increase.
So I think they certainly look like they're edging a little bit higher. But I do think there are some one-off events here over the last two months that have contributed to that.
Mark Zandi: Okay. Yeah, I just get worried. That's another worry. We're good economists, this is exactly what we do. We explain away the data. And these are all good explanations, but it makes me nervous.
Hey, Nick, do you have a favorite kind of leading indicator? You look at all kinds of things in the labor market. Where we've coalesced around and the unemployment insurance claims is the best leading indicator coming out of labor market. Would you concur with that? Or do you have another kind of smoking gun indicator you look at?
Nick Bunker: I mean, I think prior to a few months ago, I always said sharp and sudden rises in the unemployment rate, but here we are. So I do think UI claims are where I would go as well.
Mark Zandi: You would? Okay.
Nick Bunker: I think in part one, as Marisa said, the high frequency nature of it. I think also the fact that, I mean really when you, and I think this comes through in the data today in the last few months, is like deteriorations in unemployment. And labor market's starting have one when hiring goes down. But when it really hits the fan is when layoffs pick up. Or that the labor market is so weak that people who are out of work start filing for UI claims.
And I think that's another benefit that I like of this data is that yes, it could be people who are just recently laid off, but also might be people who have been out of work for a couple weeks and are like, "I don't think I could find a job and claim it." So I think it captures those two dimensions of deterioration.
So it's usually a data release that it's not prime-time viewing for me. I'm not at my computer trying to wait for it. It's more like I just check my phone at 8:40 on Thursdays. I'm like, "What happened?"
Mark Zandi: While you're fishing on Tampa Bay or something? You're out there, you throw the line, you say, "Okay, let me take a look at the UI claims."
Nick Bunker: Or most Thursday mornings trying to convince my kids to finish their breakfast.
Mark Zandi: Right.
Nick Bunker: But I think now it's going to be something that, given the state of the labor market, where I need to start watching it a whole lot more keenly.
Mark Zandi: The other indicator that maybe give you a little bit more solace, and I know you know this data really well, is open positions, unfilled positions. They've come down significantly from where they were when the labor market was very, very tight a couple years ago. But they're still high relative to historical norms. Did I characterize that right? And what do you take away from that?
Nick Bunker: Yeah, I think you did. So I think sort of two series you can watch there. One, there's the government BLS's job openings number, which we got for the end of June. And then there's Indeed's Job Postings Index, which we have data for last Friday and early next week. I'll have data as of today. So it's more high frequency.
And like you said, both of those measures have come down quite a bit from their peak in early 2022. Our series is down 25-ish percent from its peak, but it's still, both series are still above pre-pandemic levels. So our data is about 12 to 13% above its pre-pandemic level. And actually have, the last month or so, has leveled off a little bit. But I do think that that data is consistent with hiring's slowing down. So I think that's another series that I am always, the minute it's available, checking it.
And I think that if you do see a more pronounced reduction there, I think that is a very, another clear sign of, okay, employers are pulling back on demand. And now maybe that would just be a continuation of the hiring trends we've seen of late, of reduction there. But I think sharp movements there would be very concerning.
The one thing I will say is that I love our data, but there's really the series we produce and share publicly, there's really only one business cycle in it. And that's the pandemic spring of 2020, which was anomalous in a variety of ways. So I don't think the exact... In looking at the data, I mean the training set is one week it was fine and one week it started just falling off a cliff. So I think understanding the exact rate of deterioration that's going to be very concerning, is something that we'd be learning in real time.
Mark Zandi: I've asked you this before, but just to get it out there again, are ghost positions a big deal? I mean, are there more? Meaning that businesses will post for lots of different reasons, one of which is not hiring somebody, they're just posting them. And that's relatively new, at least post pandemic. Is that an issue at all in the data?
Nick Bunker: I'll say two things on that. One, the data that we use to produce the Job Postings Index, that's not as though we're just counting up in the platform. It's someone hiring here. There's an entire team at Indeed that's making sure that these are actual, real positions to help job seekers. And there's the duplication efforts, a variety of other means to make sure that anything that's on the platform is going to actually help a job seeker find a job.
And then the second point is that that sort of ghost postings concern would be more attenuated in a series like JOLTS Job Openings because that, they're asking the employer, "Do you intend to fill this in the next 30 days?" So assuming everyone's telling the truth there, that's unlikely to be a ghost position. And if you look at the trend in our data and the trend in the JOLTS Openings, they are very, very similar.
So it does suggest that our measure is broadly tracking a measure where someone's trying to verify that there's an actual intended hire there. Which is not to deny that individual people might have experience with the ghost postings. But just to say that I don't think it's a large enough issue in our data to distort the macro picture.
Mark Zandi: So you think the reality is there's just more open positions today than there was say in 2019 prior to the pandemic?
Nick Bunker: Yes. And I think that another way to normalize or reference job openings is not just to themselves in a prior time period, but to say the level of unemployment or the level of employment. Especially if you look at openings to unemployed workers, use the JOLTS data and then that's basically June, it's basically what you were at in 2019 prior to the pandemic. So that would suggest it sort of normalized in that regard.
And I will say that that ratio is on the high end of the readings of the tightness of the labor market. And the fact that it is saying, "Okay, we're roughly at 2019 levels," does suggest that that's sort of at best, where we're with the tightest labor market right now as 2019.
Mark Zandi: We're going to go to play the stats game in just a second because we're already getting a little long in the tooth, but I want to ask a question on one other issue. And that's hires and quits. The hiring rate is down a lot and the quit rate is down a lot, at least compared to where it was a couple years ago. Even on a historical basis, they feel like they're on the soft side of where they've been in a really good labor market.
And the question is, do you think the reason why quits are down is because there's less hiring? Or do you think there's less hiring because quits are down? You know what I'm saying? Because people aren't quitting their jobs, then there's just no less of a reason to hire. So is there a direction of causality here? Do you have a sense of that? Any view on that?
Nick Bunker: Yeah, so I guess my short answer to that would be yes, it's both, which is a cop out. But I do think my thinking for how we got to the spot where the quits rate was 3% at one point, which was very elevated, was that basically the US economy and the US labor market got turned off and then turned back on. And then there was a huge surge in hiring.
And a lot of employers found that some of the people that they wanted to hire who were out of work, it wasn't as easy as they thought. So they started hiring people who already had a job or people who already had a job who were thinking to themselves, "Hey, that place is hiring for far more." Like someone who maybe looked around and realized that warehouse is paying how much hourly? And jumped to that.
And that started what has been called a vacancy chain. Where someone left their job and then, okay, now I've got to backfill this person. And you basically got a lot of backfilling. And somewhere around the spring of 2022, a lot of employers started saying, "Okay, I've sort of gotten enough headcount staffed up. I don't feel the same intensity to hire people." More people were coming back to labor force. So hiring slowed down.
So then the back-fills were less needed as more people stayed in line and now, sort of it drifted back down. And on the quits rate, it's below what we saw pre-pandemic. So I think especially, I think before it was very much a mix of both. And I think moving forward, moves in the quits rate, I think will be more a function of how much our employer is looking to hire.
Mark Zandi: Got it. Got it. Well, let's play the stats game. We each put forward a statistic. The rest of the group tries to figure that out with clues, questions, deductive reasoning. The best stat is one that's not so easy. We get it immediately. One that's not so hard, we never get it. And if it's apropos to the topic at hand, all the better. And we always start with Marisa. Marisa?
Maris DiNatale: This was hard because I have so many numbers I'd like to talk about. So it's hard to pick. I'm going to go with 53%.
Mark Zandi: In the jobs report?
Maris DiNatale: Yes.
Mark Zandi: Payroll survey?
Maris DiNatale: Yes.
Mark Zandi: Is it a diffusion index?
Maris DiNatale: Yes.
Mark Zandi: It's not the one month's diffusion index, because Nick pointed out-
Maris DiNatale: Right, no.
Mark Zandi: That was below 50. So is it like the 12-month diffusion index?
Maris DiNatale: It's the three-month diffusion.
Mark Zandi: Three-month.
Maris DiNatale: Yeah.
Mark Zandi: Nick, Nick, Nick. Do you see how that's done? Okay.
Nick Bunker: That was expertly done.
Mark Zandi: Expertly done. Because the next I'm going to let Dante do the next.
Dante DeAntonio: It's weird, I didn't hear three-month diffusion index come out of your mouth. I feel like you missed the mark a little bit there. I'm just saying.
Maris DiNatale: That's actually technically accurate. He didn't get it.
Mark Zandi: That was the next thing I was going to say, obviously.
Dante DeAntonio: Obviously.
Mark Zandi: I book-ended the thing. I was wrestling the fish to the ground. I'm in Tampa Bay fishing. Took it right to the bottom of the boat. Marisa, are you embarrassed by how that went? Or you're okay with it?
Maris DiNatale: No, I'm not embarrassed because I get to make the point that I wanted to talk about.
Mark Zandi: All right, go ahead. Go ahead. Go ahead.
Maris DiNatale: When you say it's a percent from the payroll survey, there's only a few percents that it could be from the survey.
Nick Bunker: Oh, nice.
Mark Zandi: No. Oh, my gosh. He's like-
Maris DiNatale: If I said household survey, 53%, God, that could be a EPOP ratio, it could be a participation rate, it could be any number of things. All right, but well done, Mark.
Dante DeAntonio: That's all he really wanted. Just a pat on the back.
Mark Zandi: That all I cared about now.
Maris DiNatale: Well done. Okay, so I think Nick mentioned the one-month diffusion index, which fell below 50. That is kind of becoming on my list of a leading indicator or at least a coincident indicator. Because if you look back at all the recessions, you almost always have, in the month or two that the recession begins, or right before, the diffusion index falls below 50. That didn't in the '08 recession, but it happened in all the recessions beforehand.
But there's also false signals in there, right? You can have it drop below 50 and then it pops back up. But the three-month diffusion index falling below 50, almost always, maybe always, signals that you're in a recession. And in July, this fell to 53%, which is the lowest it's been since March of 2010. So the lowest it's been in over 14 years.
So it's still above 50, it's at 53. But I'm going to watch this because as this trends down, and especially as you see that one-month go below 50, if we get another one below 50, then I'm really worried about the prospects of recession.
Mark Zandi: What is it typically? What would be the average three-month diffusion index?
Maris DiNatale: The average?
Mark Zandi: Just for some context? I mean, is it typically 60%? Or 55%? Do you know?
Maris DiNatale: I mean, yeah, it's been up in the 60s. So if we go back to a year ago, it was 60.1%.
Mark Zandi: Okay, nice.
Maris DiNatale: So it's come down quite a bit, right? And this is again, the breadth of job growth. The percentage of industries that are either holding payroll steady or hiring. So we're right above that 50 threshold. Over the last month that fell to below 50. So that's kind of worrying to me.
Mark Zandi: That was a good one. Okay. So, Nick, are you up for this?
Nick Bunker: Yeah, I'm game.
Mark Zandi: Okay. And, Dante, you up for leading the questioning here? And go easy on him, not too hard.
Dante DeAntonio: I can, yeah, sure.
Mark Zandi: Okay. Nick, you're up?
Nick Bunker: Yeah. My number is 3.4.
Maris DiNatale: Did you say 3.4?
Nick Bunker: 3.4.
Dante DeAntonio: Are you leading the question? Or-
Mark Zandi: No, sorry, sorry, sorry.
Dante DeAntonio: Is it from the jobs report today?
Nick Bunker: It is not.
Dante DeAntonio: Okay. All right. So is it from this week? Something that came out this week?
Nick Bunker: It is.
Maris DiNatale: So only Dante can guess this?
Mark Zandi: No, no, no. We can all guess. Dante can just-
Dante DeAntonio: I'm just questioning the witness, apparently.
Nick Bunker: He's a grand inquisitor.
Cris deRitis: Oh, that's new. That's new.
Dante DeAntonio: Is it labor market related?
Nick Bunker: It is.
Dante DeAntonio: Fair enough.
Maris DiNatale: It's ECI from the ECI. Is it?
Nick Bunker: It is.
Maris DiNatale: Yeah.
Dante DeAntonio: Annualized growth, wage growth. Is that?
Mark Zandi: Employment cost index.
Nick Bunker: Of what series?
Dante DeAntonio: Private sector wages?
Nick Bunker: Yeah. Yep, that's it.
Mark Zandi: Okay, so it's-
Maris DiNatale: Oh, annualized?
Nick Bunker: Annualized, yeah.
Mark Zandi: Annualized quarter to quarter growth. So we got the Q2 2024 data, that was up 3.4% annualized over Q1 for private industry wages and salaries. Okay.
Nick Bunker: Correct.
Dante DeAntonio: Got it.
Maris DiNatale: So do I get credit for getting that?
Mark Zandi: Yeah.
Maris DiNatale: Okay. Thank you.
Mark Zandi: That was very good because you said ECI first. So you get credit.
Dante DeAntonio: Call it a team effort.
Mark Zandi: Well done, Marisa. Well done. Well done.
Maris DiNatale: Thanks, Mark.
Mark Zandi: I'm patting your head over here.
Maris DiNatale: Thank you.
Mark Zandi: Nick, you want to explain? Why'd you pick that?
Nick Bunker: Yeah, sure. So I think to my mind, I think several other, many other people, ECIs are like the gold standard wage growth measure. I think for a while, many people had been waiting for wage growth to return to its pre-pandemic pace. Sort of sign of like, "Hey, here's a rebalanced, cooled off labor market."
And I think we got that in this report. 3.4% annualized for the series is the lowest we've seen in several years, really until since September, the fall of 2020. So that, say you're a member of, you're at the Fed and you're like, "Hey, this is great." So wage growth, it's cooled down.
I think now, especially with hindsight from this morning, it's a question of, okay, you got wage growth back to pre-pandemic levels, but what did it cost to get there? Or if that's the indicator that you were watching before you felt like things were in a good spot. Maybe you waited a little bit too long just because wage growth sometimes can be a leading indicator where things are in the labor market.
Mark Zandi: One quick question. There's this kind of rule of thumb that the wage growth, compensation growth, that's consistent with the Fed's 2% target, is what we got, three and a half percent. On a year-over-year basis, we're still at four. And across most of the wage data we're a four.
Do you think three and a half is the right number? Or do you think four is the right number? Do you think that's splitting hairs? Are we there in terms of where we need to be in terms of wage growth and inflation?
Nick Bunker: I think so, at least at this point in time. In part because usually the rule of thumb is inflation plus trend productivity growth, and maybe that's for a long-term goal. But right now, if you take productivity numbers on their face value, it's growing 2% year over year.
Mark Zandi: Two and a half.
Nick Bunker: So two plus two is four. Yeah, so two and a half. So 4% seems more very sustainable. In addition to that, you also have the fact that, okay, take inflation in account for that nominal wage growth of what we've got. We're still seeing some real wage gains, but that seems less inflationary to me right now.
Because real wages are growing, but they're still below their trajectory we saw before the pandemic. Workers are still trying to make up for a lot of those that cut in real wages from that big burst inflation. So it seems less concerning to me right now to have relatively strong real wage gains, given what real wages did in '21 and '22 and into '23.
Mark Zandi: Yeah, I'm with you. Okay, let's do one more because we're running out of time. Dante, you want to go? And, Cris, you want to question, be the questioner here?
Dante DeAntonio: Sure. I'm going to use my stat game soapbox to push back on the hurricane didn't have an impact argument. So that's just to give you some-
Mark Zandi: That was Marisa's argument.
Dante DeAntonio: Well, and the BLS's argument too, to be fair.
Mark Zandi: Oh, you're right.
Dante DeAntonio: I'm going to use this to push back a little bit.
Mark Zandi: By the way, Marisa's formerly BLS. We keep promoting people though.
Dante DeAntonio: Me too. So I'm knocking down my own people.
Mark Zandi: That's right, I forgot [inaudible 00:59:36]. That's right, that's right.
Dante DeAntonio: 461,000.
Maris DiNatale: The number of people who have a job but weren't at work because of weather.
Dante DeAntonio: So we mentioned it earlier. I was going to not use it, but I wanted to use it anyway. That's exactly right, Marisa, congratulations.
Mark Zandi: That's bogus. That's bogus.
Maris DiNatale: What?
Mark Zandi: That's bogus.
Cris deRitis: Wow.
Dante DeAntonio: She was deep in the numbers before the podcast. She was ready to go. She's prepared.
Maris DiNatale: Number.
Mark Zandi: That's dialing it in, my friend.
Dante DeAntonio: I didn't give her a tip that it was coming, so it's legit.
Mark Zandi: All right, go ahead.
Dante DeAntonio: So it's the number of people who are employed, but they weren't at work in the reference week due to bad weather. This is from the household survey. It's one of a couple pieces of evidence in my mind, that there clearly was an impact from the hurricane. So that's the highest number of people that were in this category, not at work due to bad whether, in a non-winter month, since September of 2017.
And that was the only time in the last decade that it's been this high, not in the winter. So oftentimes in either January or February, you get a spike because of snowstorms across the country. But back in September 2017, you had Hurricane Harvey and Irma both hits and that caused a big spike in this number. And so this is the highest since then.
So that's one thing. So there's clearly was an uptick in the number of people who weren't at work because of bad weather. In the household survey, they still count those people as employed. The problem is in the payroll survey, they don't. So if you don't work at least one hour for pay during the reference period, you don't get counted as employed.
I think the timing here matters a lot, right? So if a hurricane hits and it's not in the reference week at all, the impact is pretty minimal. If the hurricane hits in the reference week, but it's like in the middle of the reference week, it's still often doesn't have a big impact because if people worked Monday and a hurricane hit Tuesday, they're still employed.
If you look at Beryl, it made landfall on Monday of the reference week in Houston. So if people were out starting on Monday, if they were out that whole week, that's a big hit. So the timing matters a lot even though it wasn't the most destructive hurricane we've ever seen. I think the timing of it making landfall makes it such that it's likely to have a sort of out-sized impact on job growth.
We'll get a better sense of that when we get the state and metro job data in a couple of weeks to see did Houston really take a big hit? It's also clear in the UI data. So you see the spike in Texas in UI claims. Just in the reference week and the week after, claims are up a combined probably about 30,000, 25 to 30,000.
And you would think not everyone who wasn't at work in that week is filing a UI claim. Some of them are expecting to go back to work pretty quickly. So to me, even if the impact on the top line number is 20, 30, 40,000, that changes the story pretty much. If the headline number was 150 or 160, that's pretty much in line with what we've seen over the last couple of months on average anyway.
So I'm not sure that people would be quite as concerned about things if the headline number was 160 versus 114. So to me, I think it probably does matter in terms of how people read the report and how they feel about it.
Mark Zandi: Yeah, I mean what, we were 114,000 jobs, payroll jobs created in the month?
Dante DeAntonio: Yeah.
Mark Zandi: Again, I think underlying is 150, 150. That would be sort of the hurricane effect.
Dante DeAntonio: Right.
Mark Zandi: Okay, let's end the conversation with going around the horn and each of us providing their estimate of the probability that the economy is going to go into recession, start a recession at some point in the coming year. And also, I'm also really curious about your views on what the Federal Reserve should do next.
I mean the widespread expectation before today's jobs numbers were that the Fed would start cutting interest rates at the September meeting. They didn't in the July meeting. Big mistake in my view, but September meeting. And that would be a quarter percentage point decline, 25 basis point decline.
So I'm curious, what do you think they will do now given today's numbers? Or anything? Would this change your thinking about that? Make sense? Everyone good with that?
Okay. Let me begin with you, Cris. And also what was your... Last time we did this was probably a month or two ago. What was your prior probability of recession? How's it changed?
Cris deRitis: I think my prior probability was around 30%. I am at 35%. So a very modest increase. As I kind of alluded to earlier, this kind of is meeting my expectation here. There's risk, but I don't see that the risk has changed appreciably. I expected that there was this higher risk of Fed making a mistake.
To that note, I don't think the Fed is actually going to change their stance. I think their quarter point September, I think still seems the most likely. Not that they shouldn't, there's a difference between what they should and what they will do. Probably should cut-
Mark Zandi: That's what they will do. What should they do?
Cris deRitis: They should have cut on Wednesday.
Mark Zandi: Okay. Yeah.
Cris deRitis: They could cut. They have full rein, they could cut whenever they want outside the meetings. Or they could go to 50 basis points, which is what the market is currently pricing in. The odds of a 50 basis point rate cut by September is now up to 75%.
But I think that would be really, unless the data really support it in the next jobs report, I think that would be shocking to the market. It could actually be counterproductive. Then we actually go into recession because of that confidence.
Mark Zandi: And just for context, the kind of, so-called unconditional probability of recession is about 15%. So if you look, typically we have a recession every six, seven years. So about 15% and you're saying 35% probably.
Maris DiNatale: Right.
Mark Zandi: So elevated. And you're saying the Fed should have already cut, but they'll probably still cut one at the September meeting, a quarter percentage point.
Cris deRitis: Yep.
Mark Zandi: Okay. All right. Marisa?
Maris DiNatale: My probability of recession is 40%.
Mark Zandi: 40?
Cris deRitis: Whoa.
Mark Zandi: That's a big move.
Cris deRitis: That's a big jump.
Maris DiNatale: Yeah, it's a big jump.
Mark Zandi: Because you were at 20 I believe, weren't you?
Maris DiNatale: Yeah, I don't know if I was a 20 or 25, somewhere in there. So yes, it's a big jump and maybe I'm being overly reactionary here to this jobs report, but it's not just the jobs report. I mean it's all the other labor market data that is going south rather rapidly is very worrying to me.
I think the Fed should have cut on Wednesday. They most certainly will, hopefully in September. I fear it may be too late by the time they do. I mean another, as we were looking at all this data over the past week, Mark, another really good indicator of a recession is the Fed hiking rates and keeping them being on a hiking cycle for a long period of time. Recession has typically happened within a year after that's ended.
So I just worry that they've already moved too slow. I agree with Cris that a 50 basis point cut, barring some really wild data point that we get between now and September, might show that they're panicking and that would be counterproductive to financial markets. But yeah, I'm worried, much more worried than I was a few months ago.
Mark Zandi: Interesting. Dante?
Dante DeAntonio: I think last time we talked about it, I was at 25%. And I would say that's up to a third at this point. I'll go there. So it's a little bit more concerned but not significantly more concerned.
I think the base case here is for a 25 basis point cut in September. I do think you've obviously got quite a bit of data that's going to come in before then. So you've got another employment report, you've got two inflation prints that they'll get. If those all look like what we've seen over the last two months, I think that could give them enough fuel to go with a 50 basis point cut.
If they look fine, and if the employment report bounces back next month and job growth is back to 175 or 200K, and it looks like the hurricane was the big thing here. And maybe the unemployment rate stops rising or creeps back down even a little bit. I think any one of those things happening, I think they'd still cut 25, but I think any of that would ensure that they're not going to make a big splash with a 50 basis point cut.
Mark Zandi: Nick?
Nick Bunker: I'll come in I think a little bit below Marisa, like three eights, like 37 and a half I think.
Mark Zandi: Oh, jeez, I got to do that in my head.
Maris DiNatale: Can we round up to 38?
Mark Zandi: Three eights, eight, three divided by eight.
Nick Bunker: Full disclosure. I was like, "Three eights sounds right." And then I did, used the calculator on my phone real quick to double check my math. So I'm not bragging here. So I do think-
Mark Zandi: I think fractions are dead, by the way. I'm just saying.
Dante DeAntonio: Not a thing anymore. Nobody uses them?
Mark Zandi: Not a thing.
Cris deRitis: They're still teaching them.
Nick Bunker: Still teaching them.
Mark Zandi: Like a analog clock.
Maris DiNatale: Or a landline phone.
Mark Zandi: Like a landline phone. Inside joke, inside joke. Okay, Nick, go ahead. Sorry.
Nick Bunker: Yeah, so I'll say I don't know if, I think maybe I'm just trying to temper myself a little bit of overreacting to one month of data. But I think, like Marisa said, it's not like it's just this report, it's other reports. And also the trend recently has been in a very negative direction.
I fully expect the Fed to cut in September and I think even if it's 50 basis points or whatever, and if it is 50 basis points, then, boy, there's been some bad data since then. But I think it's also how quickly does that medicine actually start to solve things? I think one of my lessons the last two years is it can take some time for monetary policy to take effect, quite some time.
So it's not as though, hey, we get a few cuts and then all of a sudden the housing market gets going wild or investment starts booming and consumers start spending again. So I think it might be that they do start loosening, but it's a question of how effective that's going to be on a what timeframe?
So yeah, I think maybe my number, it's my recession probability. It's like perhaps a particularly volatile number right now and maybe the next report, it will come back down. But my concern about recession is definitely a lot higher than it was this time last month.
Mark Zandi: Do you have view on the Fed?
Nick Bunker: Well, I think they'll probably go quarter point given where things are heading. And I think that makes sense to me, but it's more of a, "That makes sense given the tools they have." Maybe there's a whole lot of board guidance in that meeting. The dots really start moving down and there's sort of projections for other things.
So they can sort of get more bang, more power out of that meeting. And things that aren't just short-term rate cuts and that could help. But I wonder, they have those tools, but are those tools going to be effective on the timeline that's necessary?
Mark Zandi: Yeah, well I was at 25%. I'm with Dante. I'd say a third, 33%.
Dante DeAntonio: Welcome to the couch.
Mark Zandi: Very close to Cris. Very close to Cris. Yeah. I think that's the first time I've been close to Cris in a while. We're pretty close.
Yeah, I mean, all the trend lines look a little disconcerting to me. They're just headed in the wrong direction and I'm not quite sure what stabilizes that, except for the Fed cutting interest rates. And they've been very, in my mind, inappropriately managing policy here for quite some time. I mean, in my view, the Fed achieves their objectives six months ago.
Full employment, unemployment rate. By the way, a 4.3% unemployment rate is on the high side of full employment, in my mind. If it ticks up another notch or two, I'd say that's a weak labor market. That's a labor market that's operating below full employment.
So they had achieved that goal. And on inflation, it makes no sense to me that they're targeting, they're a slave to the 2% target on the consumer expenditure deflator. When the only reason why the consumer expenditure deflator is above target is owner's equivalent rent, which you can't measure. Can't measure it in the good times. You can't measure it in these times, given what's going on in the housing market.
And there's growing evidence, research even from the Fed system, saying, "You shouldn't be targeting it." It's counterproductive. And if you look at the consumer expenditure deflator without owner's equivalent rent, it's been one and a half percent year-over-year for more than a year. So you're there. And if you are, why five and a half percent funds rate target? That's above any estimate of the natural rate, of the equilibrium rate, the r-star where policies neither supporting or straining economic growth.
So this policy just seems out of step. It's a mistake. It's been a mistake for a while and it's now creating an issue and the probabilities are rising. Now having said that, I don't think we're going into recession. I mean I do think the rise in unemployment, what's going on here is the economy's potential growth is a lot higher than it is typically because of all the labor force growth related to immigration.
And as, Nick, you pointed out, productivity growth is elevated right now. It's two and a half percent. In fact, if you say labor force growth is one and a half percent and productivity growth is two and a half, that's 4% potential growth. The GDP through the Q2 was 3%. That explains the rise in unemployment. You're operating the economy at a rate that's below its potential and this is going to go on.
But that's a different thing than saying demand is weakening. You've just got a lot of supply. The supply side of this country say, "I want to grow and if you don't let me grow, unemployment is going to rise. And that's going to be a problem if you don't let me grow." So cut interest rates.
So what are they going to do? Well, they're not going to do what I just said they should do. They're going to cut it. I agree with Cris, they're probably going to cut it a quarter point and that is a mistake.
Maris DiNatale: You think they should cut?
Mark Zandi: They should cut. They should cut a half a point and they should get the funds rate down to 4% as fast as possible because they've achieved their targets. We're there, please. We're there. We're there. For goodness' sake, we are there.
And a five and a half percent, no one thinks that that's the equilibrium rate. 4% is on the high side of any expectations. Get us to 4% as fast as possible, please. Please. Otherwise, something's going to go off the rails and we are going to be in recession.
So as you can see, I'm getting very frustrated with this. These guys need to move, otherwise my probabilities are going to be higher than a third. And Cris is, this is the thing that really upsets me. Cris could end up being right. That's really at the heart of all of this.
Maris DiNatale: Economy be damned. Cris could end up being right.
Mark Zandi: He could end up being right. Sort of right. I'm sure I'm going to give him bull. Yes, he's right. But anyway, all right. I got on our bit of a rant. Sorry about that, Nick. This is what happens.
Nick Bunker: Lots to agree with in that rant. So happy to hear it.
Mark Zandi: Okay, very good. Well, Nick, it's always a pleasure. Thank you so much for spending a Friday. I know you'd much rather be out on that boat in Tampa Bay fishing for, what do you fish for out there? Pompano or something? I don't know. You're a big game guy. I can tell. You're out there going for the sailfish or something.
Nick Bunker: Full confession, I'm not a fisherman whatsoever.
Mark Zandi: How I knew that. Economists just aren't fishermen, that just doesn't-
Nick Bunker: See, I'm good at comparative advantage. My brother-in is a fisherman, so he catches the fish and I specialize in eating them.
Mark Zandi: That's great. That's great. Well, thank you so much for coming on Inside Economics.
Nick Bunker: Yeah, my pleasure. I really enjoyed the conversation.
Mark Zandi: Just one plug. We have a sister podcast. Sarah, I'm going to have to ask you to come on and tell us the name.
Sarah: Moody's Talks Asia's Economy Unwrapped.
Mark Zandi: Okay. And I apologize to my colleagues. But that is actually a very good title and I didn't think of it, so kudos to that. I think I came up with Inside Economics, didn't I? I'm just saying. No? Marisa, do you remember?
Maris DiNatale: I wasn't part of it back then.
Mark Zandi: You weren't part of it. Cris, do you remember?
Maris DiNatale: I don't know.
Cris deRitis: I'll give you credit. Sure.
Maris DiNatale: No one's here to deny it.
Dante DeAntonio: Not willing to fight it on Friday afternoon.
Mark Zandi: I'm taking it. I'm taking the credit.
Cris deRitis: Take it.
Dante DeAntonio: You can have it.
Mark Zandi: Probably Sarah came up with it. Because she always comes up with it.
Cris deRitis: Sarah blessed it for sure.
Mark Zandi: She definitely, oh, absolutely. She blessed it. Okay. All right, dear listener, I hope you enjoyed the podcast. We'll talk to you next week. Take care now.