Listen On:
Moody's Talks - Inside Economics
Jobs Survey Smackdown!
The Inside Economics team dissects yet another upside surprise in the February jobs report and ponders the mixed messages between the payroll and household surveys. Employment is coming in hot but the unemployment rate rose to its highest level in over a year and wage growth cooled. The team theorizes on why the two surveys are so at odds with each other lately. Finally, they each opine on whether the data are leaning more toward a higher risk of recession or “no landing”. Surprisingly, they’re all in agreement.
Follow Mark Zandi @MarkZandi, Cris deRitis @MiddleWayEcon, and Marisa DiNatale on LinkedIn for additional insight.
Mark Zandi: Welcome to Inside Economics. I'm Mark Zandi, the Chief Economist of Moody's Analytics, and I'm joined by my good friends and colleagues, Cris deRitis. Hey, Cris.
Cris deRitis: Hey, Mark.
Mark Zandi: Marisa DiNatale. Hey, Marisa.
Marisa DiNatale: Hey, Mark.
Mark Zandi: And Dr. DeAntonio. Dante!
Dante DeAntonio: Hi, Mark. How are you?
Mark Zandi: Now that I see you, I know it's Jobs Friday. Not that anyone can tell me, but if you're on Inside Economics Podcast, it's got to be Jobs Friday. So, good to have you.
Marisa DiNatale: It's also International Women's Day. Happy International Women's Day.
Mark Zandi: Oh, is that today?
Marisa DiNatale: Yeah.
Mark Zandi: Yes.
Cris deRitis: There you go.
Mark Zandi: Okay. I think we did some really cool research, I saw, on female labor force participation around the world.
Marisa DiNatale: Yeah. Some of our colleagues did, and I think we'll have them on the podcast to talk about it at some point.
Cris deRitis: In a couple of weeks, I believe.
Mark Zandi: Oh, is that right?
Cris deRitis: That's right.
Mark Zandi: I took a quick look. It looked pretty interesting. You guys read the study yet, or that's to be done?
Cris deRitis: That's on my to-dos.
Marisa DiNatale: Yeah, I looked at the high-level stuff, but yeah, they did participation rates across different countries and representation of women in the labor force and in the workforce. It looks pretty cool.
Mark Zandi: Yeah, it looks pretty cool. I'll have to read that carefully. I know it's on EV, right, Dante, or it was on EV?
Dante DeAntonio: It is. Yup.
Mark Zandi: Okay.
Dante DeAntonio: That's Economic View.
Mark Zandi: Yup. There you go. Thank you. Yeah, Economic View. Okay. Well, this is March 8th, Friday, March 8th, 2024, Jobs Friday. We got the report for the month of February. So, Dante, you want to start us off here? Give us a sense of the jobs numbers. What'd they say? What do you think?
Dante DeAntonio: Sure. It was another impressive month for the labor market, another upside surprise. Job growth came in stronger than expectations for at least the third month in a row. Total job growth was 275,000. We've got some downward revisions to previous months. So, the net effect wasn't all that big in terms of trend job growth over the last three months, did cool a little bit given how big the revisions were to December and January. Job growth is still pretty broad-based, most major industries still adding to payrolls in February. It was a return to the concentration of job growth that we had seen towards the end of 2023 with healthcare, leisure, and hospitality and the public sector, adding the overwhelming majority of jobs in February.
Some other positive news under the surface of the strong headline number and some reversals from last month, average hourly earnings came in much weaker than they did in January. So, we got a big increase in January and a much, much smaller increase in February, only up a 10th of a percent, the smallest monthly gain in quite a while. So, the net effect of the last two months is really nothing to write home about. Year over year wage growth is now back close to 4%, which is where it's been and where it's been headed for a while. I think we'd also talked last month about some concern around average weekly hours. There's been this long-running downward trend in average weekly hours that would signal some softness in labor demand that did reverse course a little bit in February.
Now, if you look over the last 12 months or so, it's relatively stable. It doesn't look like it's headed south to any meaningful degree. I think probably still something obviously to keep an eye on, but doesn't seem to be as big of a concern as it might've been last month. On the household side of things, the household survey side of things where we get the unemployment rate, there was obviously a little bit of a surprise there with the unemployment rates edging up to 3.9% from 3.7% last month. Other than that, there wasn't a whole lot going on there. Labor force participation was basically flat, a little bit of increase in, if you look at prime age participation or prime age employment population ratio up a little bit, so not a huge amount of concern there.
The unemployment rate, I would read a little bit as noise. The employment series on the household survey side of things has been weak over the last couple months. Not sure whether there's a big story there in terms of the divergence, but certainly a little bit of a softer report on the household survey side of things and the payroll side.
Mark Zandi: The survey of households, is it 60K or 80K? I can't remember.
Dante DeAntonio: I think it's in the 60,000 to 70,000 household range.
Mark Zandi: As opposed to the, so-called, establishment survey, the survey business. I haven't looked, maybe you have, this feels like to me the household survey is coming in soft. I mean the job creation growth in the household survey has been coming in soft compared to the job growth in the establishment survey. Do I have that right? Do you know?
Dante DeAntonio: Yeah, it has been remarkably soft. Actually, I'll give away one of my stats I was going to use, but over the last three months, the household survey employment is down just under 900,000 and the payroll side's up almost 800,000 over the last three months. That is the biggest divergence over a three-month period ever except for April 2020, the immediate aftermath of the pandemic. Yeah.
Mark Zandi: Wow. That's interesting. That's with the revisions, because the establishment survey increases for the month of January and December were revised meaningfully lower, right?
Dante DeAntonio: They were, but yeah, job growth is still quite strong on the payroll side of things.
Mark Zandi: Okay. So, if you look at the establishment survey, which we tend to do and which is what the media tends to look at, it says a very strong job market. It may be throttling back a little bit, but underlying job growth feels like it's well over 200,000 per month. That I'd consider to be a strong job market. We can come back and talk about whether it's too strong, but if you look at the household survey, it is weak. It's a weak job market. I mean, obviously, negative number is really weak. So, what's the reality do you think? Well, maybe what's some theories about why the difference, what's going on, and what does it mean in terms of how we think about the labor market? Is it strong or is it weak or something in between?
Dante DeAntonio: I think the reality is almost certainly somewhere in between those two readings. The household survey does tend to be a little bit more volatile month to month. So, you can get a three, four, or five-month period where employment looks really weak and that it might swing to be stronger than the payroll side of things over the next couple of months. It's not unheard of. I mean the gap is bigger than it's been in a three-month period, but we've seen some pretty good divergences before. The other thing that was a little bit interesting is if you look at the age composition of employment on the household survey side of things, a lot of that decline in the last three months has been amongst 16 to 24-year-olds.
If you actually look at prime age employment, it's actually up over the last three months and the decline is all in younger and older workers. Whether that's something meaningful or just noise in the data, I'd lean more probably towards noise, especially for that 16 to 25-year-old group. Around the end of the year, you tend to get big seasonal movements because of people going back to school or being on break from school and working. So, you do tend to get large unadjusted movements in employment for that younger age group. So, that seasonal adjustment process could be a little bit funky this year.
Mark Zandi: Okay. So, it sounds like you're discounting the weakness in the household survey to some degree. Would you also discount the strength in the establishment survey to some degree? I mean, I mentioned the revisions. It feels like we're getting consistently downward revisions, and it feels like it goes back to low response rates. The initial set of responses are low. As the Bureau of Labor Statistics gets more survey responses in, they tend to be coming in on the soft side and we're getting these down revisions. So, does that sound right to you? Would you downgrade the strength and the establishment survey?
Dante DeAntonio: Yeah, I think the strength there is probably a little bit overstated. I mean, we saw huge downward revision to January. I mean, I think it's one of the largest, if not the largest, outside of the pandemic that we've had in the last decade.
Mark Zandi: Yeah. Have you noticed everything's the biggest ever? What's going on?
Dante DeAntonio: Yeah, it just seems like everything is more volatile and more subject to revision than it's been in recent history.
Mark Zandi: Wow. Okay. I know I ask you this every month and it's very difficult because these numbers are swinging all over the place because of the revisions and the noise and everything else, but what do you think underlying monthly job growth is abstracting from the vagaries of the data?
Dante DeAntonio: I still think it's probably around 200K, where we were towards the end of 2023 before we got a couple of big prints in December and January. We were at, I think, 180, 190, somewhere close to 200. We had been on a pretty steady downtrend, and I still think that's probably reality.
Mark Zandi: Okay, 200K. Do you think underlying labor force growth then is 200K?
Dante DeAntonio: That's actually the thing I'm a little bit maybe more concerned about. Labor force growth has been pretty weak over the last three, four, or five months. The 12-month average is now only about a 100K, where not all that long ago, the average was 200K pretty steadily if not a little bit higher. So, again, it's a question of, is there a little bit of volatility over the last few months and it's really 150 or 200 or is that 100K that we've now seen over the last 12 months? Is that reality? If it is, then I think the job market almost certainly has to slow down here pretty soon.
Mark Zandi: Although then none of that squares with the unemployment rate. The unemployment rate, if anything is pushed up a little bit here, it's not declined. So, if all else being equal, that would suggest that labor force is stronger than employment growth.
Dante DeAntonio: You would think that should be the mechanics. So, that's not what some of the data is showing at least over the last couple months.
Mark Zandi: Okay. I want to test one of the theory out on you, and I don't know if we talked about this in the pre-podcast last week or the week before, immigration. So, we do know we're getting a lot of immigrants into the country, given what's going on at the southern border. The CBO, Congressional Budget Office, nonpartisan folks that do the budget, they have to forecast the economy and therefore they have to forecast population growth and demographics. They came out with a study that said, "Look, we estimate 3.3 million immigrants came into the country in 2023 on top of 2.6 million in 2022 and some other big number in 2021." Just for context, typically, before the last few years, we were getting a million, let's say, immigrants. That's both legal and undocumented.
So, that's a lot of immigrants. Many of those immigrants or at least some of those immigrants are finding their way into jobs that will show up in the establishment survey, because you're just surveying businesses. You're not surveying individuals. The businesses are going to say, "I got X number of people on my payroll," but it may not show up in the household survey because those immigrants might be nervous about responding to any survey that they get. First of all, that's a small sample, 60, 70K. You may not be picking things up, but if you're an immigrant and you have any question about your status in this country, even if you feel comfortable with your status in the country, I'm not sure you'd want to respond. You may not.
Therefore, you get this gap that we're now observing. I don't think you explain the last three months. I'm sure there's a lot of noise there, but this seemingly growing gap that exists between the household survey and the payroll survey, does that make any sense at all what I just said? I guess it's logically consistent, right?
Dante DeAntonio: Yeah, I think it's definitely logically consistent. I think it's a question of, what's the magnitude of that effect and does it explain all of the gap that we've seen? That I'm less sure about.
Mark Zandi: Okay. Marisa, Cris, any comments on that theory? Any views on that? Any perspectives or anything you want to add?
Marisa DiNatale: I used to work on the current population survey before coming to Moody's many, many years ago. It was one of the common.
Mark Zandi: That's the basis of the household survey.
Marisa DiNatale: That's right, where the unemployment rate comes from. It's one of the most common questions we got at BLS is. are undocumented people being picked up in these surveys? BLS really doesn't have any good way to measure that, right? Because none of these surveys ask about a person's legal status or citizenship status. They just ask if you were born in a foreign country or not. That's all, and when you were born, when you came to the United States. So, it's certainly an acknowledged issue, but there's really no good way of quantifying it. Like you said, BLS assumes a lot of people won't respond to these surveys for that reason.
Mark Zandi: Okay. Okay.
Cris deRitis: I have a question then. Quick question. Does BLS conduct the survey in Spanish as well?
Marisa DiNatale: Yeah.
Cris deRitis: They do. Okay. So, language barrier, potentially less of an issue.
Marisa DiNatale: Right. Actually, I was just reading the other day, I was looking into it a little bit since the pandemic, BLS has modernized the way a lot of these survey collections happen. So, it used to be that people from the Census Bureau would call you. Well, first, they would do an in-person visit to your home for the first and the eighth months that you're in the survey and then they would call you for follow-ups in the in-between months. Now, they let people do electronic surveying, so they'll let you submit your answers via a web app or something like that. So, they're trying to raise response rates on these surveys by making it easier for people to respond. Yeah, they will do it in different languages as well.
Cris deRitis: But if you're a new immigrant, getting selected into the survey potentially could take some time just from a random sampling. I don't know how often that occurs.
Marisa DiNatale: Yeah, it's true of any sample-based survey. It's going to take time for that to show up in the sample.
Cris deRitis: So potentially that could be an issue here as well.
Marisa DiNatale: Yeah.
Mark Zandi: Right. Okay. So, bottom line, Dante, good report, bad report, what do you think?
Dante DeAntonio: Yeah, I mean I think it's a good report. If anything, job growth is still strong and the parts that we were focusing on that's potentially bad over the last couple months seemed to have gone away. So, I would call it a good report. Yeah.
Mark Zandi: Good report. Okay. Marisa, what do you think? Anything you want to call out in the report that Dante did not focus on?
Marisa DiNatale: I'll just call out a few things that Dante mentioned and just maybe delve in a little more. I mean, one thing to keep in mind that you mentioned, Mark, is the smaller sample size of the household survey. So, those estimates are inherently less reliable than the payroll survey. Just for context on household employment, you need a movement of over 600,000 in a single month in either direction for it to be statistically significant at the 90% confidence level.
Mark Zandi: Can I just say? I always found that weird. If I got a number that said 700 today, 1,000, I'd say that's just bogus.
Marisa DiNatale: [inaudible 00:16:21], but they'd be like, "Oh, that's actually real."
Mark Zandi: There's nothing statistically significant about that. That's just a noise. Yeah.
Marisa DiNatale: So it's rare to see a movement that big in the household survey. I think there was one in December or November, and then I think you'd have to go back to the beginning of January to get a plus or minus that big in the household survey. So, the margin of error is big in the household survey.
Mark Zandi: Yeah, good point.
Marisa DiNatale: So I just want to call that out. Then you mentioned the survey response rates. The survey response rate on the household survey this past month in February was 70% and it was even lower on the establishment survey. It was 67%. So, if you look at charts of these response rates, I mean they've just been coming down across the board for all of these government surveys, these in particular ones to watch. The JOLTS survey, which I think we'll talk about later, is even more abysmal response rate.
Mark Zandi: Right, 70% on the household survey, put that into historical context. Is that the lowest it's been since something?
Marisa DiNatale: No, it's not. I mean, if you look back prior to the pandemic though, you were up around 85%. So, really these response rates have really come down bad since the pandemic.
Mark Zandi: Yeah. Okay.
Marisa DiNatale: If you go back 20 years ago, they were like 90%. So, it's pretty dramatic.
Mark Zandi: Which again would be consistent with the immigration story to some degree. I mean, there's a lot of other things going on in terms of low response rates.
Marisa DiNatale: Just one more thing I wanted to call out, Dante mentioned the unemployment rate rose to 3.9%. That's the highest it's been in quite a while. It's still low, it's still below 4%. He mentioned a lot of that was teenagers. The teenage unemployment rate, if you look across all the demographic groups, that rose quite a bit. Men, the unemployment rate, it was either down slightly or unchanged. It was mostly women and it was mostly teenagers.
Mark Zandi: Okay. What's your assessment? Good, bad, somewhere in between?
Marisa DiNatale: Good, I think, yeah. I mean it's still getting solid job growth. It's nice to see that wage growth wasn't commensurate with the increase in employment like it was last month. These downward revisions to the payroll survey over the last couple months, we thought that would happen when we saw the December and January numbers. So, it's reasonable to think that the truth lies somewhere between these two surveys, that the payroll survey is probably overstating the strength in the job market and the household survey may be understating it.
Mark Zandi: In this discussion around, is the economy weakening and headed towards something, if not recession, something closer to recession, or is it strengthening and therefore going to put pressure on the Federal Reserve to not cut rates or even potentially raise rates? Our baseline is we soft land. Things throttle back. We come in. Unemployment stabilizes. It hangs around 4%. Life is good a year from now. The recession scenario is all these things that are starting to show some weakness in the numbers, the weakness in the household survey, the increase in unemployment. You're up now three-tenths, four-tenths of a percentage point on unemployment.
What's the rule of thumb? If you go up more than half a point in a certain amount of time, that's consistent with recession, or you have the no landing scenario where things are strengthening. I mean, if you look at average monthly job growth over the last three months, it's like 250, 275, something like that. That feels strong and we're going to overheat here. So, in thinking about this, I know the baseline, that's your most likely scenario, but of the no landing versus the recession, which do you think feels like it has a greater probability, Marisa?
Marisa DiNatale: Recession.
Mark Zandi: Recession. Okay. So, you look at this report and you say there's more weakness in here than strength.
Marisa DiNatale: Yeah. Even with the unemployment numbers, most of the increase in unemployment was people that lost jobs. It's still small, relatively speaking, and we're not really seeing layoffs pop, but you can look around at different UI claims, JOLTS, reasons for unemployment. It is all pointing to an increase in layoffs. The Challenger Report again showed a lot of layoffs. So, I would lean more on the recession side than the taking off again.
Mark Zandi: Okay. Dante, I asked you that question, good or bad, but now with this new frame, we got the baseline. I think we're all on board with the soft landing, some variant of the soft landing, but the two alternatives are we go into recession straight away here. We're just on a path towards recession or the no landing where we're going to reaccelerate. Inflation's going to pick up again. The Fed's not going to cut and may tighten and push us into a recession down the road. Where's the balance of risk in your mind, recession or no landing?
Dante DeAntonio: Yeah, I think I would agree with Marisa, just barely. I think it's toward recession.
Mark Zandi: Yeah. Okay. Cris, what do you think? Anything that these guys missed that you want to call out or you want to reemphasize? What's your assessment of the report?
Cris deRitis: I think they got it. They got everything. I might underscore the average hourly earnings because of the inflation implication, the Fed implications of that. That 0.1% growth is moderating. So, that seems to be less of a concern and I guess less of an ammunition for that reacceleration story, at least for now. The wages seem to be coming in.
Mark Zandi: Okay. So, in that balance of risks, you would say more likely recession than no landing?
Cris deRitis: That's right. That's right.
Mark Zandi: No landing scenario. Okay.
Cris deRitis: Yeah, there's some weakness. I point the continuing claims as one of those factors. People are taking longer when they do lose their jobs to find work. So, clearly some signs of some weakening going on.
Mark Zandi: Right. Okay. Yeah. I think, on balance, the report shows the economy is still chugging along, resilient, doing its thing, clearly no recession dead ahead, and a lot of cross currents. This wasn't a clean report, hard to come to strong conclusions on anything, but just feels like we're in a world where we're getting enough jobs to keep the economy moving forward without going into recession, but not so many jobs that we're going to overheat here and having no landing scenario. So, it feels pretty good to me. I don't want to say right down the strike zone. It was in the strike zone, but it wasn't like a fastball down the strike zone. It's not a curve. Who was that guy who threw the ball that moved around in different places? You know what I'm talking about?
Cris deRitis: A knuckleball?
Mark Zandi: A knuckleball. A knuckleball. Yeah. You can see I never played baseball. It was a knuckleball, right? I think you're here. No, I'm over here. No, I'm over here. I'm over here, but strike! Strike, right? Doesn't that feel like a pretty good metaphor for that report? Yeah, Dante's laughing.
Dante DeAntonio: Yeah. I mean, if somebody would've told you that you get above consensus job growth, but at the same time the unemployment rate goes up and wage growth cools, that would've seemed unlikely, right?
Mark Zandi: Yeah, I'm with you guys. I believe in the baseline. I believe in a soft landing, the economy just settling in here without a recession, but I do think the balance of risks, if we're wrong, we're headed towards recession more quickly. We're not going to have this re-acceleration, higher inflation, higher interest rates, and ultimately recession down the road. It feels like the near-term recession risks are higher than the no landing scenario. Before we move on, play the game. I promised these guys that this is going to be a relatively short podcast.
Cris doesn't believe it, but we're going to make this a short podcast. But before we move on to consider what this all means for monetary policy and for markets and for where we're headed, let's play the game, a statistics game. The game is we each come forward with a stat. The rest of the group tries to figure that out with questions, clues, 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. We always begin with Marisa. Marisa, what's your stat?
Marisa DiNatale: My stat is -1.474 million.
Mark Zandi: Goodness.
Marisa DiNatale: Dante knows what it is.
Mark Zandi: He's smiling.
Dante DeAntonio: It's the three-month change in the adjusted household survey employment.
Marisa DiNatale: It is.
Mark Zandi: Oh, really? Wow.
Dante DeAntonio: Only because that was my backup stat.
Marisa DiNatale: Dante and I pretty much had the same stat, right? But his was going to be the unadjusted three-month change, but I looked at the adjusted. So, if you convert the household survey to a payroll survey concept, so you take out ag workers, you add in multiple job holders, you take out the self-employed, then actually, the household survey has fallen by about 1.5 million employees in the last three months, which is pretty amazing considering what Dante said. You've never seen this aside from a recession. You see it in the pandemic obviously. Then you see it in 2009, coming out of that recession. This is even more job loss than in the 2001 recession. You just don't see that loss in the household survey except for a very severe recession.
Mark Zandi: So let me get this right. In the last three months, December, January, February-
Marisa DiNatale: December through February combined, the adjusted household survey employment has fallen about 1.5 million.
Mark Zandi: Okay. The unadjusted has fallen 900K.
Marisa DiNatale: Yeah.
Mark Zandi: The establishment survey's up 800.
Marisa DiNatale: Yeah.
Mark Zandi: I mean that's just pick your economy.
Marisa DiNatale: Exactly.
Dante DeAntonio: Rorschach, right?
Mark Zandi: Can you send me the mnemonics for that household adjusted?
Marisa DiNatale: Yup.
Mark Zandi: That's amazing. Someone said, just to reiterate, that that's the widest gap ever except for March of 2020, the start of the pandemic. Is that right?
Dante DeAntonio: Yeah. Either the unadjusted or adjusted concepts. The gap between either of those and the payroll survey over the last three months is the biggest that it's ever been, aside from April of 2020.
Mark Zandi: Geez. Actually, consistent with the idea that recession is a bigger risk than a no landing scenario, I know there's research out there that might be a little bit dated. I'm not seeing anything recent that shows that, yeah, probably should put more weight on the establishment survey than the household survey. It's probably a better survey because it's a bigger sample of businesses and households, but the household survey tends to lead the establishment survey and turning points in the economy. Marisa?
Marisa DiNatale: I don't know. Dante, you tell me if I'm wrong, but I've looked for that research and I didn't really find anything-
Mark Zandi: You didn't find anything.
Marisa DiNatale: ... definitive or updated?
Mark Zandi: It's a Brookings Institution's paper.
Marisa DiNatale: Do you know how old it is?
Mark Zandi: It's a good 10 years old maybe. Something 10, 12, something like that. I think it's a Brookings paper.
Dante DeAntonio: Yeah, I mean the logic behind it is certainly there that you've got a big factor in the payroll survey side of employment is this birth-death adjustment that's factoring in how much employment's being added or subtracted from new or closing businesses. That's based on historical data. So, at a turning point, that birth-death adjustment tends to lead you in the wrong direction. It's going to prop up employment when it should be falling, and you don't have that same issue on the household survey.
So, I think that's always been the logic behind that is that the household survey by design can react a little bit faster to a turning point. It doesn't have that historical basis that's propping it up a little bit in real time. But yeah, I don't remember the research offhand and how strong the evidence is for that.
Marisa DiNatale: I think that research too focused a lot on the last financial crisis. Part of the issue there, as Dante said, is this so-called birth-death model that's employed by the payroll survey and the construction industry is one of the hardest industries for the payroll survey to accurately measure, because so much of that employment is contract, not on the books. People move in and out of it. They move from manufacturing to construction and other industries.
So, during the financial crisis when much of the job growth was coming from the housing industry, the payroll survey missed a lot of the peaks and troughs in that industry, which was one of the big reasons for enormous revisions back then and why the household survey may have been a little bit more reliable. This time around, the housing industry isn't leading or lagging this. So, I don't know that that applies so much to where we are right now.
Mark Zandi: Where we are today. Okay. Okay. Fair enough.
Cris deRitis: Have either of you looked at some of the private data sources, the ADPs, paychecks? I've really been following them recently. Any signals there that would either confirm or diverge?
Dante DeAntonio: I have not been following them either. It didn't seem like either were giving a whole lot of important signal there for a while.
Cris deRitis: Yeah, that was my sense.
Mark Zandi: I've been following cursorily, meaning not carefully.
Cris deRitis: Oh, I thought you were referring to something-
Marisa DiNatale: I thought that was the name of a survey or something.
Cris deRitis: Yeah, exactly, cursorily.com.
Mark Zandi: They're coming in softer. Actually, they're coming in somewhere between the household survey and the payroll survey.
Cris deRitis: Perfect, perfect.
Mark Zandi: There's no negative numbers. The positive numbers aren't quite as positive as these have, which would be consistent with the idea that the establishment survey is certainly not the household survey. It's something in between. Okay. Dante, you want to go next?
Dante DeAntonio: Sure. I'm digging deep over here. Let's go with five-
Mark Zandi: There's a list.
Dante DeAntonio: The first two are gone, 5.687 million.
Cris deRitis: I'd say JOLTS number.
Dante DeAntonio: You're in the right ballpark. Yeah.
Marisa DiNatale: Right ballpark, or is it the JOLTS?
Dante DeAntonio: It is the JOLTS. It's JOLTS. Yeah.
Mark Zandi: So it's something in JOLTS. That's hires, isn't it?
Dante DeAntonio: It is hires. I know it's been a source of great consternation for you lately. Mark, I feel like you're really worried about the hires number.
Mark Zandi: Well, I still am. I still am. Explain the JOLTS numbers and what you're pointing to here.
Dante DeAntonio: Yeah. So, I mean, hiring has been on a pretty steady downward trend, which I mean made sense for a long time. Hiring was obviously inflated from the initial recovery from the pandemic. So, it was never sustainable to stay where we were in say, 2021 or even early 2022, but that rate has continued to edge lower. It's now slightly below where it was pre-pandemic, certainly indicative of an environment where firms are less willing to continue to staff up. They're certainly not laying people off in any large numbers yet, but certainly looks to be some weakness on the labor demand side of things. I think the big question is, does that plateau here?
We can live with the level of hiring that we have right now, but if that downtrend continues, then we could be in a pretty darker place by the end of 2024 if you continue to see that hiring come in. For now, we've been propped up by the fact that separations are falling at the same time, mainly from the quits rate falling. So, both of those are declining mostly in sync, and that leaves you with net job growth still. That hires are still greater than total separations, but again, there's likely some bottom in that quits rate. There's some level where that's not going to fall below. So, at some point, you need hiring to stabilize and plateau, or you're going end up in a world where job losses aren't out of the question.
Mark Zandi: Yeah, yeah. I mean, you got hires at a very low... They tick down again, so very low level. Quits are at a low level and layoffs are at a low level. The net of all of that is we're still creating a fair amount of jobs. This may be an unfair simulation, but if you assume the hires and quits stay where they are, let's say they don't go any further lower and layoffs just simply normalize, meaning go back to consistent with historical norms, well below that, I think we'll still get job growth, but it'll be much slower than what we're experiencing today. In fact, that is our forecast. That's our baseline forecast going forward. Does that sound right?
Dante DeAntonio: That sounds right. Yeah. I think you'd be at very low net positive job growth if you had layoffs normalized. Yeah.
Mark Zandi: Yeah. Okay. Okay. That was a good one. Cris, you're up.
Cris deRitis: Okay, 3.9 and 3.8%.
Mark Zandi: Labor market related?
Cris deRitis: Yes.
Marisa DiNatale: Well, 3.9 is the unemployment rate.
Cris deRitis: It is one unemployment rate. This is another 3.9.
Mark Zandi: But it's another unemployment rate?
Cris deRitis: Yeah, 3.9% is the total. There's another 3.9.
Mark Zandi: Some demographic group?
Cris deRitis: It's a companion to the 3.8. Yes, demographics.
Mark Zandi: Okay.
Cris deRitis: It's in honor of the day.
Marisa DiNatale: Men and women.
Cris deRitis: Yeah, there you go. There you go.
Marisa DiNatale: I don't know which is which.
Cris deRitis: Take a guess.
Marisa DiNatale: 3.9 is women.
Cris deRitis: Yes, 3.8 is men. Significance is the women's unemployment rate is now higher or at least this month was higher than men's. It had been significantly lower over the last year or so, about half a percentage point lower. So, suddenly, there was a big jump in female unemployment rate this month.
Mark Zandi: Now, there had been this more than a narrative, I think it's a reality that throughout the pandemic, women got hit harder than men. I think that went to the nature of the pandemic in which sectors got hit by the pandemic, but it feels like things have normalized in terms of men, women. We're pretty much back to pre-pandemic. Does anyone know? Marisa, do you know?
Marisa DiNatale: Women got hit harder because of the industrial composition and children being out of school and women still bearing more of that brunt, right? But the rebound for women was then stronger and women had actually been doing better than men comparatively in terms of both job growth and participation gains, but it looks like that might be normalizing. Yeah, women certainly had a bigger increase in unemployment over the month. I don't know how much of that is layoffs versus coming back into the labor force, because part of the reasons for unemployment, there was a big bump in re-entrance to the labor force too. But I don't know what that looks like by gender demographics. I don't know if you know, Cris.
Cris deRitis: Yeah. Well, I did note that all the labor force growth was women, right? Women were up.
Marisa DiNatale: It was women. Okay.
Cris deRitis: Yes. Women's labor force was up, 267,000, men actually fell, 116,000. So, all the growth this month. Again, these are demographic variables. You have to be a little careful talking about the survey having Y confidence intervals, but nonetheless, that's what the raw data suggests.
Mark Zandi: Right. Okay. I got two numbers, 2.6% and 1.6%, 2.6 and 1.6.
Marisa DiNatale: Is it from the jobs report?
Mark Zandi: No, but it's labor market related.
Marisa DiNatale: JOLTS?
Mark Zandi: Not JOLTS.
Marisa DiNatale: Oh, is it productivity related?
Mark Zandi: It is indeed. So, what's the 2.6?
Marisa DiNatale: I know what the year-over-year productivity number was.
Mark Zandi: Wasn't that 2.6?
Marisa DiNatale: I thought it was three point something.
Dante DeAntonio: 3.2, right? 3.2?
Mark Zandi: Yeah. Year over year?
Dante DeAntonio: 3.2 is annualized, the annualized.
Mark Zandi: 3.2 is annualized.
Dante DeAntonio: Okay. All right. So, 2.6 is year over year.
Marisa DiNatale: Year over year.
Mark Zandi: Yeah, 2.6 is year over year.
Dante DeAntonio: 1.6 is over the last three years, four years, one of those things.
Mark Zandi: Four years, pre-pandemic.
Dante DeAntonio: Pre-pandemic. Okay.
Mark Zandi: Average annual productivity growth, non-farm business productivity growth. So, we're still enjoying, at least through the end of last year, some pretty heady productivity gains. Of course, if you go back a year ago, productivity was falling. So, you got to abstract from the ups and downs and all arounds, but it feels like underlying productivity growth might be ever so slightly edging higher, 1.6 is a little bit higher than what we had been getting before the pandemic, in the years ahead of the pandemic. So, that's a good sign.
Unit labor costs, so that is labor compensation growth, less productivity growth that goes to business' profit margins and then ultimately to their desire to raise prices and inflation. So, that's been very weak and barely growing. It's positive but barely positive. So, that suggests that the margins are wide, they're not rising, but it doesn't feel like they're falling to any significant degree. So, that suggests that it's not putting pressure on businesses to raise prices. These productivity gains are consistent with the labor compensation growth and therefore keeping pressure off of businesses and inflation. So, very positive development.
Anything you guys want to say about the productivity? I've been at a conference here the last couple of days and it's all about AI. I think the general consensus is that this is going to be a big deal for productivity growth, but not quite yet. It's going to take a little bit of time for that to kick in. So, okay. Dante, anything you want to say? You're the skeptic.
Dante DeAntonio: I'm still skeptical and it hasn't changed my mind. I feel like you have to really squint to see an upward trend yet, but yeah, we'll see how it plays out.
Mark Zandi: See how it plays out. Okay, fair enough. Okay, I did say it was going to be a short podcast and it is going to be, but let's end with the market reaction and what it means for the Fed. Maybe, Cris, I'll turn to you, because I know you watch the market carefully. Did today's numbers change market expectations for the first Fed rate cut?
Cris deRitis: Not really. March is still off the table looking at the CME futures, about 3% chance of a March rate cut. So, no one is expecting that. May is right around 20%. At least the market expectation is the first cut's going to come in June, which is up around 75% probability. That seems to be the case. This report today didn't really change those numbers significantly. They bounce around anyway, so there was no abnormal volatility that I saw. So, that seems to be the case. Then through the year, I looked at December, it looks like the market is pricing in four cuts this year.
Mark Zandi: Four cuts. Okay.
Cris deRitis: It's a distribution, of course, but the midpoint of the distribution looks about four.
Mark Zandi: Or quarter point rate cuts.
Cris deRitis: Oh, correct. Yeah. So, it could be quarter points, it could be half point cuts. It doesn't really matter.
Mark Zandi: Typically, we have a meeting in June, July, September-
Cris deRitis: November and December.
Mark Zandi: When at the very end, December, I think. So, they're expecting out of the five meetings, there will be cuts at four of those meetings.
Cris deRitis: Correct.
Mark Zandi: The market is.
Cris deRitis: Correct. That's the implied.
Mark Zandi: That seems aggressive to me, doesn't it?
Cris deRitis: It does, it does. Still does.
Mark Zandi: Yeah. I mean, I think the one thing I took out of Fed Chair Powell's testimony this week, and he testified both in the House and the Senate, was that the bar for actually cutting rates is not that high. I think he said it and he means it, that he just needs to see another couple inflation reports that are consistent with the idea that we're going back to the Fed's target. We're there, but just confirmation that we're actually headed in that direction. That makes sense. I mean, if I were king for the day, I had enough confirmation, I'd go. Okay, fair enough. But it seems to me once the Fed starts cutting rates, at least initially, it's going to take its time, wouldn't it? If they're not in any rush to start cutting, why would they be in a rush to get them back much lower quickly? I just don't see what would be the catalyst for that.
Cris deRitis: Yeah, I'd agree with you, unless they're panicking, unless there's some weakness, right? They're really concerned about-
Marisa DiNatale: That's not what the market's saying.
Cris deRitis: No.
Marisa DiNatale: Stock prices are up, credit spreads are in. That's not consistent with the idea. Oh, we think there's going to be a weak economy debt ahead, therefore the Fed's got to lower rates more aggressively. That's not what you see.
Cris deRitis: That's right. That's right.
Dante DeAntonio: I guess I'd be a little cautious in looking at the CME futures because that's not just the baseline path. It's also risk management. There's some tale that investors are also hedging, discounting in there. So, they may be putting more weight because they're on that end of year activity, because they have some recession probability that they're trying to hedge. Again, at least that's how I frame it.
Mark Zandi: Okay.
Dante DeAntonio: It's not asking what is your baseline path.
Mark Zandi: It's an average of all these paths.
Dante DeAntonio: Correct.
Mark Zandi: If you have that on the tail, a really dark scenario that you attach meaningful probability to, then you get this result.
Dante DeAntonio: Right.
Mark Zandi: Yeah. Yeah, actually, that makes a lot of sense to me. That makes perfect sense.
Marisa DiNatale: Our baseline mark is unchanged from last month in terms of the Fed.
Mark Zandi: No, you of course convinced me. Last month, we had the Fed lowering rates in May. Then I think we had three rate cuts in the year, and I think we're now going to push it to June and still maintain three rate cuts. So, there'll be June, November, and just point of interest, that November meetings after the election, after the election. So, they're not going to cut rates right before the election. Then we have in very end of December. So, basically, it's one quarter point rate cut each quarter going forward. June is second quarter, September's Q3, December's Q4.
Then in 2025, one quarter point cut in each of those quarters. That would get us down to equilibrium r-star, that rate that's consistent with monetary policy and either restraining or supporting growth, which we put in about 3% sometime in early... Probably it'd be more like early 2026 when we get there, everything sticks to script. Interestingly enough, I mentioned this in conference, there are some really smart folks that are now thinking that equilibrium rate is 4 to 4.5%, which I found interesting. Yeah, I found interesting. Very interesting. Anyway, markets, how did they take it? I mean, this morning, I saw a lot of green in the equity market, but here we are towards the end of the day, reddish to flat, something like that.
Dante DeAntonio: It was pretty flat last time I checked.
Mark Zandi: I guess the real tell is the 10-year treasury yield, and I think that's still down a little bit, right?
Dante DeAntonio: I think that was pretty flat.
Mark Zandi: Flat too. Okay, so no reaction. So, the investors are interpreting it the way we're interpreting it. Good report, everything's sticking to the script here. Okay. All right. Well, very good. Anything else that anybody wants to bring up before we call it a podcast? I told you, Cris, this was going to be 45, 50 minutes. I think I held to that.
Cris deRitis: So what'd you think of the State of the Union?
Mark Zandi: I knew he was going to mess me up. Darn it.
Marisa DiNatale: Now that we have 30 seconds left, what did you think of the State of the Union?
Mark Zandi: Yeah, no, I'm not answering that question. Not answering. Actually, you can see my views on CNN. I wrote a very brief op-ed. I actually wrote it last night after the State of the Union. You can get my sense if you're interested of the speech from an economic policy perspective, but I'm not biting. No, Cris, I'm not doing it. I'm not doing it. Any final words, guys? Thank you. I appreciate that. Hearing none. We're going to call this a podcast. Thank you for listening in, dear listener. We'll catch you next week. Take care now.