The group dives deep into the jobs market report for May. They dissect the cross-currents in the numbers, and consider what the report means for financial markets, monetary policy and the macroeconomy. The economic fallout of the legislative deal to end the debt limit drama, including the end of the student loan payment moratorium, are also part of the discussion. And of course there is the statistics game.
The group dives deep into the jobs report for May. They dissect the cross-currents in the numbers, and consider what the report means for financial markets, monetary policy and the macroeconomy. The economic fallout of the legislative deal to end the debt limit drama, including the end of the student loan payment moratorium, are also part of the discussion. And of course there is the statistics game.
For the full transcript, click here
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 am joined by a group of colleagues, of course, my two co-hosts, Cris deRitis and Marissa DiNatale, hi guys.
Cris deRitis: Hey, Mark.
Marisa DiNatale: Hi, Mark.
Mark Zandi: Good to see you.
Marisa DiNatale: Hi, Cris.
Mark Zandi: Jobs Friday, this is now June, June 2nd 2023, so we've got a lot to talk about on jobs. And to help us with that, we've brought in Bernard Yaros, good to see you, Bernard.
Bernard Yaros: Hi Mark, good to see you.
Mark Zandi: Say something in Arabic, just anything quickly.
Bernard Yaros: [foreign language 00:00:47], that's hello.
Mark Zandi: I thought it meant, Mark, you know everything, I thought that [inaudible 00:00:55]. But that's good, that's pretty cool, that sounded exactly on point. And Matt Colyar, good to have you. Matt, have you been on Inside Economics before?
Matt Colyar: No, this is the inaugural.
Mark Zandi: Oh, really?
Matt Colyar: So this is nice, nice to be here.
Marisa DiNatale: So we have to [inaudible 00:01:12] him.
Mark Zandi: I was going to say watch out for Marissa, goes right for the jugular every single time.
Matt Colyar: It's good to know, I'm ready.
Mark Zandi: I'm gun shy now, I need backup. All right, good to have you on, Matt. And of course, Matt and Bernard along with Dante, Dante's AWOL today on Jobs Friday, he's generally with us, but the three of those guys manage economic view, so really, following the economic statistics real time. Good day to have you back on. I just got back from Toronto with clients, I had dinner with Canadian clients. I will say, surprisingly more optimistic about things. I don't know if you remember back, but we had a dinner in Toronto back last fall and they were darn right pessimistic.
And of course, I might have a compositional bias issue because people at the dinner may have changed a little bit, but I'm saying they felt a lot more optimistic than they were last time. But we're going to find out how optimistic you guys are too, because at the end of all this, we're going to talk about probabilities of recession and somewhere along the way we'll do the stats game as well. And I don't know if you guys feel like this, but I feel like there's so much to talk about. No?
Matt Colyar: Absolutely. Yeah.
Cris deRitis: Yeah, there is.
Mark Zandi: I don't know if we're going to get to it all. It feels like a two-hour podcast, but let's dive right in. Okay, Marissa, we're going to turn it to you. Can you give us a rundown on today's May, 2023 jobs numbers?
Marisa DiNatale: Yeah, it was a very interesting report. So non-farm payroll employment rose by 339,000. That blew all the forecasts out of the water. We were only expecting about 215,000 and we were on the high side of consensus forecast. So a significantly stronger report than we were expecting. Add to the that the previous two months of March and April both saw upward revisions to job growth that amounted to 93,000 combined. So almost another a hundred thousand on top of that 339. That significantly moves up the three-month moving average of job growth. It is quite a large increase in employment. It's the largest we saw, since the start of the year job growth. On the payroll side, it was pretty broad based. The diffusion index moved up above 60%, which is the percentage of industries that are either holding employment steady or adding to payrolls. We saw some big gains across many industries. Government had quite a large gain. They added 56,000 jobs over the month. In the private sector job base-
Mark Zandi: Can I ask, was that state, local, federal? What was that about, the government?
Marisa DiNatale: I think it was, let me see.
Mark Zandi: Bernard, do you know offhand? Matt?
Bernard Yaros: I think it is generally more of state, local. I would say.
Mark Zandi: State and local.
Bernard Yaros: Yeah. Yeah.
Mark Zandi: I meant to look.
Marisa DiNatale: Well, it was everything, but local government was the biggest addition. So it was both local education and non-education. That was two thirds of it, but federal and state added as well.
Mark Zandi: Got it.
Marisa DiNatale: Professional business services had a very large gain. Education and healthcare had a large gain and so did leisure hospitality. Although the gain there was I think 41,000. So that's a little bit smaller than we've seen in some of the previous months, but still quite large. The only major industries that, oh, and I should also mention the construction industry added 25,000 jobs, and that was more than double the gain in the previous month. So despite, we had seen a couple months ago that the industry lost a small amount of jobs and we thought, okay, maybe this is the higher interest rate environment and the slowdown in the housing market starting to kick in. But construction is really, really resilient. And that also was across the board, across residential, non-residential contractors, building heavy construction, kind of the whole gamut within the industry.
Manufacturing lost jobs, and that was in non-durable, good manufacturing. And then the other industry to lose jobs was information. This includes tech stuff, it includes media, entertainment, telecom, that kind of thing. And pretty much every component within information, lost jobs, everything else added at least on a broad-based industry basis. So if we're thinking about this from the perspective of monetary policy, this is obviously, really, really strong. The Fed doesn't want to see a strengthening job market, but some good news, at least on the payroll side that came out of it is that earnings growth moderated a bit compared to the prior month, So two things-
Mark Zandi: Wages, average hourly earnings.
Marisa DiNatale: That's right. Average hourly earnings growth was 0.33% up over the month. Over the year, it was up 4.3%. So those are both slower than we've seen in the prior month. The other thing is the work week fell, so the number of hours work ticked down by 0.1 hour. And this is pretty low. If you look at it historically, it's the lowest, it's the smallest, shortest work week, I guess, that we've seen since the pandemic. So that's, I think, what I want to say about the payroll survey, we can come back and fill in whatever I forgot that's important.
Mark Zandi: Okay. Did you want to go into the household survey?
Marisa DiNatale: Yeah. So the interesting thing to me about the report-
Mark Zandi: Adds the confusion.
Marisa DiNatale: ... is that the household survey shows almost the exact opposite as the payroll survey. So on the household survey side, and just so we know what we're talking about here, the household survey is a survey of households that the Census Bureau calls people up or visits them in their home. It's a much smaller sample size. It's about 60,000 households. So the estimates of it are less reliable. However, this is where the unemployment rate comes from, labor force data. So the unemployment rate rose three-tenths of a percentage point from 3.4% to 3.7. 3.7% is still a low unemployment rate, but that's a big movement in one month. Even just a movement in either direction, that is quite large. You have to go back over a year to see a movement in one month of 0.3%. Unemployment rose by over 400,000. The number of unemployed people rose by over 400,000.
And employment, the number of employed people fell, and it was really like the opposite of what the payroll survey showed. And at the same time, the labor force held steady. So the labor force increased modestly up by a little bit over a hundred thousand people. And the labor force participation rate held the same as it was over the month. So nothing strange was going on with the labor force. It really was an increase in the number of unemployed people and you can see that in some of the details. So if you look at how long people have been unemployed, there were a lot of newly unemployed people over the month, people unemployed less than five weeks. And also the reasons people gave for unemployment. Most of the increase in unemployment came from people who said they had permanently lost a job, not quit.
Actually, the number of quitters fell according to the household survey. So really, very difficult to reconcile this survey with the very, very strong payroll survey. It happens. It happens. It's not that uncommon for them to show different things, but I was looking at the magnitude of the difference between the two and it's enormous. You have to go back like to 2010 to see a difference this big between just the employment change in the two surveys.
Mark Zandi: So just a couple things before we move on. I thought, and I hadn't had a chance to look, but I thought if you look at the household survey reconciled to the payroll survey, so there are conceptually different, and BLS, Bureau of Labor statistics, keeper of the data does publish the household survey more apples to apples to the payroll survey. That actually showed a pretty sizable increase. Is that right? Do you know? Bernard, do you know?
Bernard Yaros: Yeah, that's correct. The adjusted household employment figure actually increased 394,000, which would be consistent with what we saw with non-farm payrolls.
Mark Zandi: And I guess the reason why household employment fell, and this goes to the measurement differences, is self-employed, the number of self-employed?
Bernard Yaros: Yes, there was a big ... Yeah. Yes.
Mark Zandi: Okay. Okay. So that helps reconcile it a little bit or somewhat.
Marisa DiNatale: Yeah, that's it.
Mark Zandi: Based on a gain in employment, they're both pretty strong apples to apples. Right? Would you say that?
Marisa DiNatale: Right, right. When you try to make the definitions the same. Yeah.
Mark Zandi: Okay. Second question goes around unemployment, that did increase and the fact that a lot of it was because of a sharp decline in self-employed might suggest maybe there's noise in there. Going back to your points, a small sample and not sure I'd read too much into it. But where do you think underlying unemployment is? Abstracting from the monthly vagaries of the data? Up 3.7, down, 3.4 Where do you think we are?
Marisa DiNatale: I think we're kind of in the middle of that. Some we're probably around three and a half percent, somewhere. And I know we'll get into this. The other thing is if you reconcile all the other employment data and all the other labor market data that came out this week, that's more in line with what we saw on the payroll survey side than on the household survey side.
Mark Zandi: It feels like to me, unemployment's a rock solid three and a half percent-ish. And that's where we've been for more than, at least a year. One month might be a little higher. One month, like today, was a little higher. One month, a little lower, but net about 3.5%. Okay. Okay. Let me turn to you next, Bernard, to fill in the gaps there. Anything you would call out that Marisa missed in her rundown?
Bernard Yaros: I think one thing that I've been looking at in recent months has been the composition of those unemployed. So the BLS does provide the composition of unemployed persons who are either on temporary leave, whether they're on permanent job loss, whether they're re-entrants into labor force or new entrants, others who are completing seasonal work. And one thing that was concerning towards the end of last year and early into this year was the rising share of unemployed that were on permanent job loss. That rose from about 21% towards the end of the third quarter and rose to nearly 27% earlier this year. But that has since started to tick down, and in this job report it was roughly flat. So I don't see any worsening in the composition of unemployed persons nationally. And even though we're looking at a low level of the aggregate number of unemployed is low, but it would be concerning if you start to see a growing and growing share of that being permanent job losses.
Mark Zandi: But you don't see it.
Bernard Yaros: You don't see it. Yeah. Yeah.
Mark Zandi: Okay. Matt, anything you want to call out in the bowels of the numbers? I know you dig deep too.
Matt Colyar: I think that 3.7 unemployment rate, like you guys, I think is maybe a little bit misleading. This isn't big uptick of people coming off the couch to now look for a job, which pulled us from 3.4 to 3.7. It's unincorporated self-employed people. That's a definition that could be unpacked a bit more, but they were working, they were doing something whether that's not viable for them anymore. It's not this 400,000 plus increase in a labor force, which is, that's the balancing that the Fed's looking for. So it kind of diminishes that increase for those who are really leaning on the unemployment rate to say, okay, the loosening has arrived. So maybe a little bit of a false loosening there.
Mark Zandi: Okay. So Cris, let me turn to you and if there's anything that you want to call out, that'd be great. But more significantly, just what's your broad takeaway here in terms of what's going on in the labor market, taking all these discordant kind of pieces of information coming from the report, all the cross currents, how are you reading this? And then I want to ask you how the markets are reading it, but that seems interesting as well.
Cris deRitis: Absolutely.
Mark Zandi: Feel free to take any of those questions in any order you want.
Cris deRitis: Sure. I'd say, first of all, I think this report underscores the number that's in every report, which is plus minus 130 K. You know what that is? That's the 90% confidence interval around the payroll. So for that reason, I don't think we want to read too much into any given month here, especially given the differences here.
I think the report does suggest strength still, certainly in terms of hiring, and we had the Jolts report earlier, I won't reveal any numbers, but for the statistics came. But it does suggest that the labor market remains quite resilient. We can, to be determined, how resilient it is and what the specific number is, but I don't see any question that things are falling apart here.
Mark Zandi: They're not falling apart, they're resilient.
Cris deRitis: And to Matt's point, I don't see 3.7 as really being truly underlying unemployment here, that there's a lot of noise in the data. So that's my take. In terms of what the Fed does with this, I think they take that into account. I don't think they'll overreact to this report in one direction or another. A little bit here for everyone to enjoy. Doesn't suggest that things are particularly overheated or that things are falling apart as well.
So I don't know that this report makes the difference in terms of their call. I think they'll wait for inflation, other measures before they decide what to do here.
Mark Zandi: Can I ask, you go to the market, just one quick question. So we came to the underlying unemployment rate is about three and a half percent. That's kind of where we've been for a year or so. And the today's numbers doesn't sway anyone's thinking around that. What about underlying monthly payroll employment growth? What do you think that is now? Is it 300K? Is it 250, 200? Do you have a sense of that?
Cris deRitis: It's hard. It's hard to really pin it down, but I think it's probably between 200, 250 would be my-
Mark Zandi: You do, okay.
Cris deRitis: My guess. When everything is said and done and the revisions come in, but that's my sense of things here.
Mark Zandi: And just for context right now, it feels like that's consistent with stable unemployment because labor force is growing 200 to 250K as well, but that's not sustainable though for long, given underlying demographics. Doesn't feel like we can sustain labor force and job growth of 200 to 250K. Meaning if labor demand remains that strong, if job growth remains that strong, at some point labor force isn't going to be able to keep up, unemployment's going to start declining again. Wage growth is going to start accelerating again. The Fed's going to start raising rates again. Right?
Cris deRitis: That's right. Although it's been at least-
Mark Zandi: Surprisingly resilient, as you say.
Cris deRitis: They could keep showing up. I'm not confident in terms of the timing of that. I agree with you, certainly, at some point, got to run out, but a month after, I don't know. I've been burned now a few months in a row here,
Mark Zandi: It's pretty incredible actually.
Cris deRitis: Yeah, it is incredible.
Mark Zandi: Yeah. Yeah. Let me ask you one more question, and I've asked it before, but I keep asking it. 3.5%. Does that feel like full employment or are we beyond full employment? Is the full employment unemployment rate, the NARU higher than three and half percent is, what's your thinking around that?
Cris deRitis: So I think it's right around there, right around three and a half, right around three and a half, maybe 3.7 If that's ...
Mark Zandi: Yeah. Okay.
Cris deRitis: At the end of the day that shows up. But I keep pointing back to the wages, average hourly earnings, which of course, isn't the best measure, but still a reasonable measure. Still on the high side, so it would suggest a little bit of heating up there. So maybe we're just over full employment here, but tough to make the call.
Mark Zandi: So three and a half percent, it feels like it's a little outside the consensus. I think if you ask the average economist at the Fed, they'd say closer to four, maybe even higher than that. Yeah. Okay. Let me go around the helm very quickly and ask, same question. Do you think three and a half percent is consistent with full employment or is that beyond full employment, Bernard?
Bernard Yaros: Yeah, I would say three and a half is consistent. Yeah. Yeah.
Mark Zandi: Okay. Marissa?
Marisa DiNatale: Yeah, I think, again, I always go back to the job market we were in before the pandemic too, and we had a very low unemployment rate. We didn't have an acceleration in wage growth. So I don't think that this is where we've overshot a ton, we're probably-
Mark Zandi: It was three and a half.
Marisa DiNatale: In the ballpark of where full employment is.
Mark Zandi: Yeah. Okay. And Matt, do you want to take the other side? I dare you.
Matt Colyar: I would. I was planning on it.
Mark Zandi: Oh, there. Okay. Oh, baby. Okay.
Matt Colyar: I'm going to step away from the unemployment. 3.5. Yes, 2019 tells you that's probably not an overheating US economy, but if you look at prime age, labor force participation, all those type of metrics, I don't know where else you're getting, you're above what you were in 2019. I don't know how many more people can be relied on to keep coming into the labor force. So to me, that seems like it leans towards the overheating side of things. 77.6, that's women's prime age labor force participation rate today. And that's the highest ever. So, I stole somebody's number.
Mark Zandi: You stole. [inaudible 00:20:09] Maybe we should stop this line of questioning.
Matt Colyar: Yeah. Maybe.
Mark Zandi: Can I ask you though, and I've wondered about this, overall employment to population is still below pre-pandemic levels. And then that goes to one of the reasons why participation hasn't recovered is older workers, folks that are over the age of 65, they kind of stepped out and haven't come back in, but maybe they're starting to come back in, or that's where we're getting, I don't know. I didn't look at that labor force ... It did not. Cris is saying no, that participation rate did not rise.
Cris deRitis: No, actually it dipped a little, I believe.
Mark Zandi: Oh, did it really? Okay.
Marisa DiNatale: The only age groups that aren't at or above in terms of participation and employment, where they were prior to the pandemic are 20 to 24 year olds where participation is a little bit below where it was prior to the pandemic and people 65 and older.
Mark Zandi: Yeah, 65 and over.
Marisa DiNatale: But every other age group is, so yeah, the 65 and older accounts for the lion's share of the missing workers. Right.
Mark Zandi: But if the question is, as Matt posed, where are those workers going to come from. That that could be it.
Marisa DiNatale: It would have to be them. Yeah, I'm agreeing. Yeah.
Mark Zandi: It doesn't have to be, but most logically.
Marisa DiNatale: Yeah. That's where most people are out of the labor force.
Mark Zandi: Right. I'm wondering if we should do ... Well I was going to say maybe we should do the stats game before we keep going, but before we do that, turning back to you, Cris, the market reaction, holy cow. The equity market is taken off. Now, I know the debt limit agreement's got to be playing a role, but it felt like, I've been watching it took off after the employment report. It really took off.
Cris deRitis: I haven't looked last minute or so, so maybe it's back down again. But I saw the Dow up over 600 points, and the bond market also sold off in that long-term Treasury yields, 10-year Treasury yields were up 5, 6, 7 basis points, which is a pretty big move in a given day. But what's-
Mark Zandi: Well, not anymore, right? Seems like it can go five, six.
Cris deRitis: Oh, that's true. The volatility in the bond market is incredible.
Mark Zandi: And that's a topic for another podcast, by the way.
Cris deRitis: Yeah, that's right. That's right.
Mark Zandi: But have you looked at the futures market for Fed funds? What are markets discounting? Has the discounting with regard to the Fed move at the next meeting changed?
Cris deRitis: Yeah, two thirds of participants suggest a pause or skip-
Mark Zandi: Pause.
Cris deRitis: However you want to say, and a third say a hike.
Mark Zandi: Okay. Well that's interesting.
Cris deRitis: Right. And what, last week, it was the reverse. People, the majority were saying hike and the minority were saying pause. So clearly things are moving.
Mark Zandi: Yeah. What do you think is going on there? You know why? Taken at face value, that employment number would say, Hey, well, I guess there's a lot of cross currents there, but the top line payroll employment number would suggest, Hey, maybe they would tighten, but that's not what the market's thinking at this point.
Cris deRitis: Yeah. Well, my interpretation of the market's thinking this is a confusing report.
Mark Zandi: Okay. Right. I'm not sure how the fed's reading this.
Cris deRitis: Yeah. I think the Fed's going to wait. They don't have a strong case to be made here to hike. I'm assuming that they'll wait and see at this point. But those futures market, they turn around pretty quickly, so if you-
Mark Zandi: Yeah, yeah, for sure.
Yeah, they swing widely in any given hour, they can swing pretty widely. Yeah.
Cris deRitis: But again, I don't see this report as being proof positive or definitively moving the Fed towards a hike.
Mark Zandi: Okay. Bernard, I know you follow the futures pretty carefully. Any view there on what's going on ... The market reaction, is that surprising to you?
Bernard Yaros: Yeah, I think it's just confusion and they probably just, Marcus could just be seeing this confused report as just being further sign of resilience and further odds of a soft landing. That's probably it.
Mark Zandi: Soft landing. Interesting. Okay. Well, why don't we do this before we play the stats game. Let's now talk about the debt limit, which is now deal, which is now, I guess the president hasn't signed the piece of legislation yet, I don't think, but it's gotten through Congress, House, Senate, and it's going to his desk and he's going to sign it just in time. And then we'll come back and play the stats game and then go on to some other topics. But I'll turn to you. Bernard, because I know you and I have been doing a lot of work here along with Cris. What do you think, maybe you can just quickly summarize, don't spend a lot of time summarizing the deal because it's all over the place, but more importantly, the macroeconomic consequences of the deal.
Bernard Yaros: So without going into all of the details, it's still important to say that the most important piece of the Fiscal Responsibility Act, which is the debt ceiling deal that will suspend the debt limit until January, 2025. So it'll effectively remove the debt limit as an issue until after the presidential election next year. The biggest component of this are caps on federal defense and non-defense discretionary spending next year and the year after. So as the law is written, the non-defense budget should shrink by about 8% next year, and in the following year, growth in the non-defense budget will be limited to just 1%. On the other hand, the defense budget will be allowed to grow by 3% next year. But in the following year, its growth will also be limited to 1%. So these caps on discretionary spending are forecast to reduce budget deficits by about $170 billion over the next two years.
These caps are only for two years, and after that, the assumption is that spending is just going to grow in line with rising prices in the economy. An important point though, before I get into the macroeconomic consequences, is that there were a lot of side deals made between negotiators that are not written in the legislative text, but they're going to come into play once lawmakers start writing up the fiscal 2024 budget later this summer and early fall. And so there's a lot of side agreements that essentially shift around money and allow appropriators to adjust the non-defense appropriations higher to be more consistent with current levels. So in reality, when we look at the non-defense budget next year, it's probably going to be slightly lower or flat relative to current levels. So depending on what you assume, you do get different macroeconomic consequences.
So if we take the Fiscal Responsibility Act at face value, I would expect that you see a reduction in real GDP by about three-tenths of a percent. The unemployment rate would probably be higher by two tenths of a percent, and non-farm employment would be reduced by nearly 200,000. And the peak of these impacts would be late 2024, early 2025. So this would be the upper bound, but if we include the side deals which are going to mitigate the impact on non-defense, then you're talking about even smaller impacts. I would say real GDPs reduced maybe closer to 0.19%, a one 10th of a percent increase in the unemployment rate and a reduction in non-farm employment by closer to 130,000. So it's a headwind, but we have to consider this against the alternative, which would've been a Treasury default on the government's obligations and an almost guaranteed recession later this year. So any of us, I'm sure, would take this deal over the alternative.
Mark Zandi: Yeah, I think it's about as good a deal as you could get from an macroeconomic perspective. This is really not much of anything.
Bernard Yaros: And the fact that both sides were equally angry at it, it's a good sign. It just shows that it was a straight down the line bipartisan bill.
Mark Zandi: And my sense is, if they had gone through the normal budget process, they might have landed pretty close to where they landed anyway.
Bernard Yaros: Exactly.
Mark Zandi: So all that drama for nothing really. Again, going back to the points we were making in previous podcast, it might not be a good idea to get rid of this thing, this debt limit.
Bernard Yaros: Yeah, exactly. Yeah.
Mark Zandi: This is really not advancing the ball to any significant degree. That makes perfect sense to me. Thank goodness. And one of the really positive things, just to reinforce it, is the limit is suspended until January of 2025. So we get this thing on the other side of the election, so that doesn't get involved. And it also provides some budget mechanisms. I guess, that's the right way of describing it, to help make sure that the government gets the funding it needs at the end of this fiscal year, which is the end of September, so that there's no government shutdown.
Bernard Yaros: Exactly. Yeah. So I would say the odds of a government shutdown later this year are significantly reduced. Because if they don't provide a full year budget for fiscal 2024 by the end of this calendar year, then you're talking about an automatic limit to all discretionary spending, including defense to 1% growth. And obviously, that would be something that wouldn't be tolerated on both sides.
Mark Zandi: Can you explain the side deals? It's not legislative. Are they binding in any sense?
Bernard Yaros: One example is that they're going to rescind about $20 billion that were funds that were provided to the IRS under the Inflation Reduction Act. And part of that money is going to be just repurposed to other non-defense programs, which will alleviate the cut in the law itself.
Mark Zandi: So it's like a handshake deal saying in the appropriations process, the lawmakers are going to do this.
Bernard Yaros: Exactly. Yeah.
Mark Zandi: Right. Okay. Yeah. Interesting. One other aspect of the deal was around student lending and the president agreed to end the moratorium on student loan payments. I think that is as of September of this year. And there's some hand-wringing about how significant an impact that will have. Now, that doesn't affect the macro impacts you just discussed, Bernard, because we were assuming that was going to happen anyway.
Bernard Yaros: That was already baked in [inaudible 00:30:53]
Mark Zandi: That was baked in.
Bernard Yaros: Yeah.
Mark Zandi: Yeah. That was going to happen regardless if there was a debt limit deal or not. But let's talk about that because I know there are some folks handering about that. And Cris, you spent a boatload of time on student loan issues. Can you just explain this a little bit in more detail and what the macroeconomic consequences of ending the moratorium loan student loan payments might be?
Cris deRitis: Yeah, sure. So there are about 45 million borrowers with student loan, federal student loans. So they'll start repaying come September 1st. It's about $5 billion a month in terms of remittances to the government. So that's, I guess, positive from the budgetary standpoint.
Mark Zandi: But just remind people that most student loans are now government-
Cris deRitis: That's right.
Mark Zandi: Directly from the government.
Cris deRitis: Right. Okay.
Mark Zandi: Yeah, over 90%.
Cris deRitis: Over 90%. Right. So implications would be, of course, well, people have to start paying again so that's going to cut down in terms of their spending, their saving activity elsewhere. But by and large, I suspect that most borrowers are prepared for this. It's going to hurt. They're going to reduce some of their activity elsewhere, but I don't expect that all of a sudden we'd see a wave of delinquencies from them. Now, that's the majority. Even before the moratorium, there was always this pocket of students who got a loan, in particular, who got a loan and didn't complete their degree. So they ended up with a burden that they really couldn't service given their current levels of income. I suspect those people are going to be under pressure once again when the payments restart.
One caveat here is that the Biden administration is pushing this income-based or income-driven repayment plan that they have, which would reduce the amount that borrowers have to pay to 5% of their income and shorten the term for which they would have to repay as well. So to the extent that borrowers do take advantage of those programs, that could certainly help ease some of that financial stress that they're facing. So I see this as certainly having some drag overall on spending in the broader economy, but I don't see this as triggering a wave of defaults or anything of that magnitude.
Mark Zandi: So 5 billion a month times 12 months gets you 60 billion. You divide by 27 trillion GDP. That's two tenths of a percent or something, if my arithmetic is right, and that kind of feels like that might be an upper bound because the student loan borrowers aren't going to cut all the way back on the 5 billion. Some of those folks have savings and other sources of income. And I also think the income-driven plan, they call them income driven plans now, right?
Cris deRitis: Income driven, IDR payment plans, IDRs.
Mark Zandi: The changed the name, but that may also smooth out some of these effects as well. It might not be exactly 5 billion, certainly not right away 5 billion. Okay.
Cris deRitis: That's right.
Mark Zandi: I guess it's kind of a headwind, but it's kind of really very modest headwind to the macro economy.
Cris deRitis: That's right. And the other wild card here is that the forgiveness program, the Biden proposal to forgive $10,000 of debt for most borrowers, 20,000 for Pell Grant recipients, that's still on the table. The Supreme Court will be deciding soon on the validity of that plan, and we'll see what happens.
Mark Zandi: Yeah, I guess the thinking is that the consensus view is that the Supreme Court's going to strike that down, not allow that to happen.
Cris deRitis: Yeah, that's the-
Mark Zandi: That feels like the-
Cris deRitis: ... that's the current thinking.
Mark Zandi: Yeah. It'd be a little bit of a surprise if they said, oh, okay, go ahead and do that.
Cris deRitis: Right. Yeah.
Mark Zandi: But if they did do that, that would even reduce, maybe even completely eliminate the impact of the end of the payment moratorium on the economy because people's debt's forgiven. So that's got to have some positive effect [inaudible 00:34:58] being equal.
Cris deRitis: Yeah, that's right.
Mark Zandi: Okay. We'll see how that plays out. Bernard, do you know when the Supreme Court has thought to be able to rule on that?
Bernard Yaros: By the end of this month, it should be.
Mark Zandi: Oh, by the end of June.
Bernard Yaros: By the end of June, yeah.
Mark Zandi: Okay. Okay. Very good. Okay. Anything else on the debt limit, Bernard? Matt, anything you want to bring up? No. Okay. Fair enough.
Bernard Yaros: Just one last point, looking ahead to 2025. So 2025 is going to be a very busy year for fiscal policy because not only does the debt limit become an issue, again, many of the Trump tax cuts from the 2017 Republican tax law are going to expire then. And then you also have the expansion of the Affordable Care Act under the Inflation Reduction Act, that also expires around that same time. So there's very likely you're going to have a grand bargain between Democrats and Republicans that could potentially increase substantially non-defense spending in exchange for a permanent extension of the Trump tax cuts. And if that's the case, we're looking at significantly higher deficits and debt over the next several years and decades.
Mark Zandi: Wait, when's going to happen? Is going to be happen in the lame duck session after the election and before the new-
Bernard Yaros: Or no, it's going to be during 2025, calendar 2025.
Mark Zandi: Oh, okay. But the debt limit is January of 2025, right?
Bernard Yaros: 2025.
Mark Zandi: So the debt limit won't be part of that grand bargain?
Bernard Yaros: Oh, it will be because towards the end of 2025, that's when the looming expiration of these other policies will be there. And again, remember the debt limit will come into effect in early 2025, but extraordinary measures and cash on hand will push back the drop date for that towards the middle of the year.
Mark Zandi: That makes sense.
Bernard Yaros: So yeah, it'll probably be in the middle of 2025 when we see something big in terms of [inaudible 00:36:52]
Mark Zandi: Oh, so this election is huge. This election is massively huge.
Bernard Yaros: And again, the calculus, it changes if it's a Republican sweep, if it's a democratic sweep or if it's-
Mark Zandi: That's I'm saying.
Bernard Yaros: ... a divided government. Yeah.
Mark Zandi: That's like a big deal.
Bernard Yaros: Exactly.
Mark Zandi: Trump tax cuts, debt limit, what else? You mentioned one other thing that's got-
Bernard Yaros: The Affordable Care Act.
Mark Zandi: Affordable Care Act, right.
Bernard Yaros: Yeah, yeah.
Mark Zandi: Yeah. Affordable Care Act.
Bernard Yaros: Subsidies. Yeah, yeah.
Mark Zandi: Yeah. That'll be very interesting. Yeah. Okay. Very good. Hey, let's play the game, the stats game. Hopefully people still have the stat they can use. I know I blew through one or two of mine already. The game is, we all put forward a statistic. The rest of the group tries to figure that out through questioning and clues, deductive reasoning. The best stat is one that's not so easy we'd get it immediately. Not so hard that we never get it and is bonus if it's apropos to the topic at hand, which is, well, the labor market, I think. But we can go beyond that. So Marissa, tradition, we missed you last week. You would've been very proud of me. I think I did well, reasonably well. Cris actually did very well. Our guest did very, very well. Ben Harris did very, very well.
Marisa DiNatale: Oh yeah.
Mark Zandi: Yeah, that's right. Yeah. But we missed you. So what's your stat?
Marisa DiNatale: Okay, I have two related statistics. 70.7% and 54.7%
Mark Zandi: Labor market related?
Marisa DiNatale: Yes.
Mark Zandi: Is it a participation rate?
Marisa DiNatale: No.
Mark Zandi: Is it in the payroll survey?
Marisa DiNatale: One of them is.
Mark Zandi: Oh, one, and the other is in the household survey?
Marisa DiNatale: Right.
Mark Zandi: And they're related?
Marisa DiNatale: Yes.
Mark Zandi: Okay. It's not a participation rate. Some kind of diffusion index?
Marisa DiNatale: No.
Mark Zandi: Okay, guys, I'm already stumped. It's a tough one. It's a tough one. It says a ratio something to something.
Marisa DiNatale: It's a percentage. Yeah. Yes, there is a denominator.
Mark Zandi: I keep the conversation going
Marisa DiNatale: A numerator and a denominator.
Mark Zandi: That's funny. That's funny. Is it? Can you give us a clue?
Marisa DiNatale: Sure. So this is something that's released every month by BLS, but it's not in the employment situation news release. It's not published in that news release.
Mark Zandi: Oh, is it in Jolts release, the job opening labor turnaround survey?
Marisa DiNatale: No. So it is the household and the payroll survey related.
Mark Zandi: Okay. Okay.
Marisa DiNatale: This is calculated every month, but they don't publish it in that news release.
Mark Zandi: Oh, I see. Wow. That's interesting.
Cris deRitis: Is it the survey response rate?
Marisa DiNatale: Yes.
Mark Zandi: Oh.
Marisa DiNatale: Yes.
Mark Zandi: Oh, very good. Oh man, explain. That's a great one. That is a really good one. Way to go, Cris.
Marisa DiNatale: So are, yeah, that was good, Cris. So 70.7% was the percentage of the household survey sample that actually responded to the household survey in the month of May. 54.7% was the response rate for, this is the initial estimate. We're going to get three BLS conducts, two subsequent revisions to each payroll number that they release. And the reason that they do that is that each subsequent month, they get more and more delayed responses into the previous month's survey. So 54.7 was the response rate or the collection rate in the payroll survey for the month of May. So that should go up into the 90%. That response rate usually ends up being by the third collection somewhere in the 90%. But 54.7 is really, really low even for a first response rate. There was one that was lower recently. It was like 50% or 49% back in November of last year. But generally on the first try, BLS gets somewhere in the range of 70% of respondents. So this was very, very low on the payroll survey side.
On the household survey side, the 70.7% response rate is pretty normal recently. But just to give you some context, it was 75% last May. And before the pandemic, it was well into the eighties. So response rates for all of these surveys, particularly surveys of households, have really fallen to, in some cases, alarmingly low levels. Since the pandemic. In the JOLTS survey, the response rate is down to like 33% or something.
Mark Zandi: I saw that. Yeah.
Marisa DiNatale: It's in the thirties. And so this just makes, to Cris's point about the statistical significance of all of these numbers, it makes those confidence intervals even wider. So for example, in the household survey, for significant job growth, employment growth on the household side, you need something over 600,000 in either direction for that to be significant.
Mark Zandi: I didn't know that. Really?
Marisa DiNatale: Yeah. Yeah.
Cris deRitis: Wow. That's incredible.
Mark Zandi: So the response rate for May has been falling month to for quite some time, but it was particularly low in May.
Marisa DiNatale: For the payroll survey.
Mark Zandi: For the payroll.
Marisa DiNatale: For the household survey-
Mark Zandi: More difficult.
Marisa DiNatale: ... this is pretty normal. It's kind of what it's been since the end of last year, but it is repeatedly down year after year. If you go back 10 years, the response rate was 90% on average. Now, the average is something like 70% in the household survey.
Mark Zandi: So it suggests we might be in store for some meaningful revisions then to the data.
Marisa DiNatale: I think it just underscores as we get confused about how to interpret some of these numbers, you have to remember that these are coming from surveys and there's a response rate problem in a lot of these surveys now. So yeah, we have to take everything with some skepticism.
Mark Zandi: Fortunately, the payroll survey is ultimately so-called benchmark to actual counts of jobs from unemployment insurance records, the so-called, what is it, quarterly census of employment and wages. QCEW. And that has shown some big differences, right? I can't remember which quarter, but there was a quarter back in 2022. Q2 2022. It was, we actually saw job loss.
Marisa DiNatale: Yeah, that was in the summer of last year according to the QCEW.
Mark Zandi: And that's not at all the picture you got from looking at the initial employment estimates. And that still hasn't been incorporated fully into the.
Marisa DiNatale: That's right. That won't until the next benchmark.
Mark Zandi: We will see some downward revisions probably.
Marisa DiNatale: I think so.
Mark Zandi: Of course, who knows? But based on that, all else being equal, we'll see some downward revisions. Interesting.
Cris deRitis: Do you think there's a large firm bias for the first batch of reporting that of those 54%?
Marisa DiNatale: I don't know. I don't know if that is a thing. I'm sure BLS has done a ton of research on response rates, so it could be. Most of this is automated by companies now. They don't even have to have a person physically respond to the survey. There's some automation and that may bias it toward larger companies that have more infrastructure to do that than smaller companies.
Mark Zandi: That was a great one. That was really cool one. Matt, you want to go next?
Matt Colyar: Sure. I kind of do want to make us guess how much PTO Bernard should be allowed to take in 2025 with all the legislation going on. But no, I have one.
Mark Zandi: That's a good point. Did Bernard ever take PTO? I don't think he's ever taken-
Bernard Yaros: I am taking next week for my move to Philly.
Mark Zandi: Really? That's disappointing. I'd have to say. No, no.
Marisa DiNatale: He's going to be closer to you, Mark. You should be happy.
Mark Zandi: Only joking. You deserve a boatload of PTO for sure.
Bernard Yaros: Yeah.
Mark Zandi: Yeah. And he's moving back into the neighborhood, into the hood.
Bernard Yaros: Yep. Yeah.
Mark Zandi: Yeah. Here in Philly. Good to have him back. Yeah. So Matt, what's your statistic?
Matt Colyar: Let's go with 79.
Mark Zandi: The number 79.
Matt Colyar: The number 79.
Mark Zandi: It's not a percent, it's just the number 79. Okay.
Matt Colyar: And it is my first time on here, so I got to shake it up. So keep that in mind.
Mark Zandi: [inaudible 00:45:44] Wait a second. What are the ... oh, it is market related?
Matt Colyar: Labor market related. Not a percentage, not an index.
Mark Zandi: Well, is there units to it or is just 79? You can't tell us the units.
Matt Colyar: I can try to think of a way to do that, but you have to give me a second if I can do that without giving it away.
Mark Zandi: You know what I'm saying, it's like 79 dollars, that's not dollars per barrel. Okay.
Matt Colyar: No.
Mark Zandi: Because is oil is at 76.
Marisa DiNatale: Is it a statistic that was released this past week, Matt?
Matt Colyar: Today. Yeah, it's longer run. So N equals 79.
Mark Zandi: N equals 79.
Matt Colyar: Yeah. It takes this morning's report into consideration.
Mark Zandi: Is it labor market data?
Matt Colyar: Yes.
Mark Zandi: Did someone say that? Okay. Should we know this, Matt, or are you just like ... Oh no, we should know it.
Matt Colyar: It's relevant, but it's not something that you would see referenced in the Wall Street Journal or anything. It's just-
Mark Zandi: We know it. Should we know it or not?
Matt Colyar: It would be difficult to guess, but I think it's relevant. I think it's interesting.
Mark Zandi: Is this something you calculated it or is it actually in the report?
Matt Colyar: Yes, I did calculate it.
Mark Zandi: Okay. Oh, 79. No units, labor market related. Probably don't know it, but we should guess. It is interesting.
Matt Colyar: Yeah.
Mark Zandi: 79. Wait a second. It's not-
Cris deRitis: Is it an index of some sort?
Marisa DiNatale: Is it an age?
Matt Colyar: No, it's a number of times. It's a occurrence.
Marisa DiNatale: Oh, oh, oh, okay. Okay.
Matt Colyar: It's a frequency.
Mark Zandi: Okay. Yeah. 79 times.
Marisa DiNatale: The number of times the unemployment rate has moved.
Mark Zandi: Oh, that.
Marisa DiNatale: More than two tenths of a percentage point-
Mark Zandi: That's a good one.
Marisa DiNatale: ... in a single month or something.
Matt Colyar: Exactly. So the number of times it's been-
Mark Zandi: Oh, oh.
Matt Colyar: Three percentage point.
Mark Zandi: That is amazing, Marisa, way to go. That is so cool.
Matt Colyar: Yeah. Yeah.
Mark Zandi: Where's the cowbell? I thought, boy, that's really impressive.
Matt Colyar: No chance. Yeah, that's great. So it's the number of times, not just in absolute terms, but a 0.3 percentage point increase in one month in the unemployment rate, 79 since the World War II era or post World War II. And the follow up would be how many of those 79 occurred either in a recession or immediately preceding a recession?
Mark Zandi: Oh. Oh, that's a really good one.
Two thirds of them?
Marisa DiNatale: Most of them.
Cris deRitis: Two thirds.
Marisa DiNatale: It has to be.
Matt Colyar: Definitely. Two thirds is great. So 65. So it's a little high. Three quarters.
Mark Zandi: Pretty close.
Matt Colyar: Yeah. Yeah. [inaudible 00:48:38] Sorry, no cowbell.
Mark Zandi: No cowbell. Hey Matt, just a suggestion in playing the game.
Matt Colyar: Sure.
Mark Zandi: That was an admirable statistic, but you might just say this has happened 79 times. That's fair.
Marisa DiNatale: We got there.
Mark Zandi: That would've been fair. Just a little bit of a something.
Matt Colyar: Yeah. I appreciate the uncomfortable play [inaudible 00:48:59]
Mark Zandi: But the fact that Marisa got it like that, man,
Cris deRitis: That was impressive.
Mark Zandi: That's scary good.
Matt Colyar: Very good.
Mark Zandi: Yeah, very good. That's a good one. Okay. Very good. Bernard, you want to go next?
Bernard Yaros: Yes. So I have three statistics. They're all related to labor and they're all related to each other within labor. So it's 1.795 million, negative 110,122, and then a positive, 199,670. The last two numbers are year over year differences and the first one is a total, it's a stock.
Marisa DiNatale: What was the last number you said Bernard?
Bernard Yaros: 199,670.
Mark Zandi: Is rounding going to mess up the-
Bernard Yaros: No, no. I mean, I could just say 200,000, I guess.
Mark Zandi: Okay. Round, because I'm not writing this down.
Bernard Yaros: Okay. So the first one is 1.8 million.
Mark Zandi: 1.8 million.
Bernard Yaros: The second would be negative a hundred thousand. And the third would be a positive 200,000. The first number is a stock, and the second two are year over year differences.
Mark Zandi: And is it like a labor force statistic? Like a household labor force-
Marisa DiNatale: Is it from the JOLTS?
Bernard Yaros: No.
Mark Zandi: Is it in the household survey?
Bernard Yaros: No.
Mark Zandi: Is it in the payroll survey?
Bernard Yaros: No.
Mark Zandi: Oh, okay. It's labor market related. 1.8 million, 1.8 million or something. Is it from the BLS?
Marisa DiNatale: No, no.
Is it UI claims related?
Bernard Yaros: Yes. Yes. The first, yes. Yeah, yeah, yeah, yeah.
Mark Zandi: Continuing claims.
Bernard Yaros: Yes, that's the first one.
Mark Zandi: Okay.
Bernard Yaros: So the last two, they come from the household poll survey. So they didn't necessarily come out this week, but they came out earlier this-
Mark Zandi: Boy, these guys are tough, man. [inaudible 00:51:04]. Man, they're right. I like it. Yeah. Explain the household poll survey. We haven't talked about that in a long time.
Bernard Yaros: The Census Bureau started this survey, which they conduct almost every month or every other month. And they've changed their questions over the course of the pandemic, but they ask a lot of questions about employment status in 2020, 2021 when we had a lot of the federal pandemic relief, there were a lot of questions about stimulus checks, the child tax payments. That helped provide some granularity there. And they also ask a lot of questions about mental wellbeing during the pandemic and nutrition. So it really asks a broad array of questions at the national level and also at the state and metro area level.
Mark Zandi: So can you repeat one more time what those two remaining statistics from the poll survey are?
Bernard Yaros: Negative a 100,000 and a positive 200,000. And they're both year over year changes.
Mark Zandi: And they're related to UI claims.
Bernard Yaros: UI claims.
Marisa DiNatale: Is it the number of people who say they're relying on unemployment compensation to meet their monthly bills?
Bernard Yaros: Very close, but the aspect of these people, it doesn't have to relate with what needs they're meeting with by using.
Mark Zandi: Go ahead, tell us. Go ahead.
Bernard Yaros: All right. So obviously, the first one is 1.8 million. That's continuing claims, which rose 6,000 in the week ended May 20th, but more important, they're up almost 400,000 from a year ago. But there's a lot of shifting, if you look underneath that increase, there's seeing a lot of shifts as to who is receiving unemployment benefits. And that's where the last two statistics come into play. And the negative 100,000, that's the year over year reduction in the number of people receiving UI claims whose household income is less than 35,000. And then the 200,000 year over year increase, that's the increase in the number of UI beneficiaries whose household income is 150,000 or more.
Mark Zandi: Very interesting.
Bernard Yaros: So we've seen a top line increase in continuing claims, but it's starting to really shift drastically towards the higher end of the income distribution. So the lowest income folks are really declining. Their share of total UI continuing claims is declining significantly. Whereas those at the very top end, they're seeing their share of continuing claims increase. And this makes sense because you've had very high paying industries such as financial services and tech, accounting for at some point up to 70% of job cut announcements late last year. So these are jobs that are going to be skewed towards the higher end of the distribution. And I would say the implications of this from a macroeconomic perspective are relatively benign. Lower income folks tend to have a higher marginal propensity to consume any extra dollar that they receive. So if more of them are exiting the unemployment rolls and are becoming employed, I think that's going to be a good thing for consumer spending.
And on the other hand, if higher income folks are accounting for a larger share of UI continuing claims, I think they have the wherewithal to continue to spend more or less during any spell of unemployment, especially since they account for the lion's share of excess personal savings nationally.
Mark Zandi: Yeah, that's a really cool statistic. We never would have ever ever gotten that, but that's a really, really cool statistic. We need to follow that on a regular basis. Can I ask, I look at the initial claims carefully and they're 230, 235,000 per week, which feels it's up, but it's still low.
Bernard Yaros: It's still, yeah. Yeah.
Mark Zandi: Very low. And our break even estimate is, I think, Matt, you wrote this 265,000, meaning you need 265,000 claims to be consistent with no job growth and I think recession would probably be closer to 300K, something like that. Just for the 1.8 million continuing claims, that's up you say, but is that high low? How would you characterize that number historically?
Matt Colyar: It feels low historically.
Mark Zandi: Yeah. It feels low, yeah.
Matt Colyar: Yeah, yeah.
Mark Zandi: Okay. It's up, but it's still low.
Matt Colyar: It's coming from very, very low levels from a year ago. So yeah, I know people have said, look at how high the year over year percent increases in it, but it's coming from a super low base, so I wouldn't read too much in that year over year changes. It's a normalization, not a red flag.
Mark Zandi: I'll have to say I have a statistic, but I'm a piker compared to you guys. Geez. You guys have some really good statistics. I'm not kidding, those are good ones. Cris, do you have a good one?
Cris deRitis: Not really, but I'll use ... The other ones have been taken.
Mark Zandi: Are you going to be embarrassed?
Cris deRitis: No, no, it's a decent one.
Mark Zandi: A decent one. Okay.
Cris deRitis: Not my best work, but passable. 25,000 and 106.
Mark Zandi: 25,000 and 106. And employment related?
Cris deRitis: Yes.
Mark Zandi: In today's job numbers?
Cris deRitis: Yes.
Mark Zandi: In the household survey?
Cris deRitis: Nope.
Mark Zandi: In the payroll survey?
Cris deRitis: Yes.
Matt Colyar: Construction employment?
Cris deRitis: Exactly. Construction employment. And what's the 106? They are related.
Mark Zandi: Oh, oh, oh. There's separate statistics. Oh, I didn't realize that.
Cris deRitis: Oh, sorry, sorry.
Mark Zandi: Oh, okay.
Marisa DiNatale: It's one 106, not 106,000?
Cris deRitis: No, 106. 1-0-6.
Marisa DiNatale: Okay.
Mark Zandi: Okay, so construction employment increased 25K, that's your first number. And now you're saying the next number is 106. What is 106?
Cris deRitis: Exactly.
Mark Zandi: And is that also in the employment report?
Cris deRitis: Yes.
Mark Zandi: Yes. In the payroll survey?
Cris deRitis: In the bowels, yes.
Mark Zandi: Oh, really in the bowels. 106.
Marisa DiNatale: Is it in an industry?
Mark Zandi: Is it employment gain-
Cris deRitis: It is an industry. It's the construction industry.
Marisa DiNatale: Oh, so it's some Is it, oh, is it like heavy civil construction infrastructure related stuff?
Mark Zandi: No, he's got to be cute. There's got to be some cute thing he's doing here.
Cris deRitis: I got this cute bias.
Mark Zandi: Yeah, yeah. Yeah. Well, because if it's 25K, that's pretty boring for him to pick that one, so the 106 has got to be a little juicy in some way.
Marisa DiNatale: Is that a month over month change?
Cris deRitis: No, the month over month change is zero. So it was 106 this month, it was 106 last month. It's construction related. It is one of your favorites, Marisa.
Matt Colyar: Marisa, what's your favorite? [inaudible 00:58:10].
Cris deRitis: What's that?
Matt Colyar: Something related to home improvement. Is it in the construction spending release that we got this week or?
Cris deRitis: No. No, it's the index of aggregate weekly hours.
Mark Zandi: Oh, oh, yes. for construction. That's right. Oh.
Cris deRitis: Construction employment went up 25K, but the number of hours was flat.
Mark Zandi: Oh man. Interesting. Okay. Well that reminds me, going back to the aggregate numbers, aggregate hours worked, which account for employment increase gain and average weekly hours, I think, was down in the month, I believe.
Cris deRitis: Down 0.1 down.
Mark Zandi: And that's an indication that it wasn't that strong when you look at it. As a proxy for output for GDP, it's suggesting a pretty weak number actually, not a strong number.
Cris deRitis: That's right.
Mark Zandi: Back to your point of what your observation, Marisa, the average weekly hours fell now to a level that is quite low.
Marisa DiNatale: Yeah.
Mark Zandi: Even pre-pandemic it's quite low. So that takes an edge off that 339K in payroll employment for sure. Okay, that's good. I'm going to be embarrassed if ... I'm going to do it real quick. Ready, you guys? Lightning run. If you don't get it really fast, it'll be very disappointing, especially you Matt, because you're the newbie. You got to get this right. 80.7%. Tick, tick, tick, tick.
Bernard Yaros: Labor force participation.
Marisa DiNatale: Participation.
Cris deRitis: Participation by age workers.
Mark Zandi: Not labor force participation.
Marisa DiNatale: [inaudible 00:59:44] pop.
Bernard Yaros: Prime age pop.
Mark Zandi: Matt, where was your voice? Did you see what ... I mean, come on.
Matt Colyar: Yeah, I should have.
Mark Zandi: I mean, come on.
Matt Colyar: You're very quick.
Mark Zandi: This is what Marisa does. She goes right for the jugular. She doesn't give you a chance. So it's employment to population ratio for prime age workers, which actually ticked down by the way, and it's elevated consistent with a tight labor market, but not an inordinately tight labor market. This is kind of where it was pre-pandemic, going back to our observations about unemployment labor force participation. So again, in other statistics say, yeah, it's tight labor market, but not screamingly over overdone, and it's starting to ease up, which is what we need.
Okay, we're running out of time. There's so many topics, as I said. I wanted to talk about ai. We're going to talk about ai, but I think we need to do it at in different podcast because there's just so much to talk about. And Matt, I'll get you back on because I want you to talk about the business to business sales data. And maybe real quickly just summarize what that is and what it's saying, because I think that's apropos here as well. But we'll have you back on to talk about that in more detail. But just give us a sense of that B2B data.
Matt Colyar: Sure. Coursely, it's tracking the spending that one business is spending to another, so B2B, and that's based off of really accounts receivable data. So if you can aggregate that, you can roll that up enough and you can also slice it and say what's happening in California or Pennsylvania. You can aggregate all of that spending up to the US level and say, what does business look like? What is the dollar amount happening? Or it's occurring month to month? How are businesses spending money? Even at a top level, that's a pretty good proxy for economic activity. So that was always our goal with trying to manipulate this data to be able to tell a story about the trajectory of the US economy.
There's some oddities to it and some massaging and some manipulation that we've kind of gone back and forth on. But what it is signaling and what we're confident it is signaling is kind of aligns with the story outside of the labor market. So what GDP is saying, which is this, the US economy is slowing down, but it's not contracting. There's a deceleration going on. It's relatively timely. So it's a good way for us to keep our finger on the pulse of what's happening in a pretty timely manner. So yeah, I'd be happy to talk about it more.
Mark Zandi: Yeah, we should come back because it feels, actually, the weakness in March, April was a little more unexpected because this is all nominal dollar. So if you adjust for inflation, it felt like we were getting some actual declines in business to business sales in aggregate across the economy. But we'll watch that in maybe in a month or so. We get another month is worth of data. This is data we collect, Moody's collects based on accounts receivable. We'll come back and maybe we'll spend a little more time on that. Okay. Let's end the conversation. Given all the data, our odds of recession again, and two numbers, what's the probability recession starting this year in calendar year 2023? And what is the probability recession starting in calendar year 2024? Of course, 2023 is we're almost halfway through the calendar year, so probability are starting to come in, I think. But nonetheless, let's go around the horn and get people's perspective on that. And Marisa, I'll start with you. What are your probabilities?
Marisa DiNatale: For a recession starting this year? I'd say 40%.
Mark Zandi: Okay.
Marisa DiNatale: For a recession starting next year, I'd say 55, 60, somewhere in that range.
Mark Zandi: Pick a number. You're on the record.
Marisa DiNatale: 57 and a half.
Mark Zandi: There you go. Okay. And why, can I ask, will probabilities be so much higher next year? Some of it's just arithmetic, I guess.
Marisa DiNatale: Yeah, this year's almost over. We just keep seeing strong economic data, really. And now we have the debt ceiling thing largely behind us. That was the biggest near-term threat to a recession. So I think it's pretty unlikely this year, but we don't think the Fed's going to raise rates again, but they could. I wouldn't argue too strongly if somebody said that there's another one or two rate hikes this year. We're already in a high interest rate environment, relatively speaking. Even if they don't raise rates again, we don't think they're going to lower them. Rates are going to be at the level that they're at, probably at least until early next year. I think that's going to shake some things out of the tree as we saw with SVB.
I think that businesses and consumers are reacting and will grapple with the high interest rate environment. And I just think the longer this goes on without a recession, the higher the odds are that something else happens in this environment of high interest rates and kind of a cooling economy that could push the economy into a recession. So I don't think it would take much for something geopolitical or a spike in energy prices or something like that to come along and derail things.
Mark Zandi: So 40% this year and 57.5% next year.
Marisa DiNatale: Yeah.
Mark Zandi: And can I just ask, has your sentiment improved or darkened?
Marisa DiNatale: I think my sentiment for this year has improved a bit over the past few months, just as we keep getting good data. For next year, it's kind of stayed where-
Mark Zandi: Where you've been.
Marisa DiNatale: ... I've been for the past few months.
Mark Zandi: Okay. Okay. Bernard, what about you?
Bernard Yaros: So I would say 30% odds for this year and then 50 for next year. And it's not just the debt limit that's made me a bit more optimistic. But I'd say the one statistic this week that really made me feel good was the quits rate, which ticked back down to 2.4%, which is, actually, you had a month in 2019 where the quits rate was also 2.4%. And the reason why I focus on the quits rate, I know the markets were a bit upset after the JOLTS report because they saw the increase in job openings. But for me, quits are just a much better indicator of the labor market. There's a lot of issues with job openings. It's very tough to tell how aggressively firms are really recruiting for given job openings. But for me, quits are really more of an unambiguous sign of confidence in the labor market, in the economy, and the ability to find better paying jobs.
And the fact that it's coming down, it's practically where it was in 2019. It suggests that wage growth should start to ease in the next quarter or so. It's still going to take time, but wage growth is going to start to head in the right direction based on the historical relationship between the two. And I focus on wage growth because the Fed is saying they're also plugged into wage growth as the key mechanism by which to bring down inflation, especially service sector inflation. So if we do start to see wage growth coming down as presaged by the quits rate, I think that reduces the potential for a policy error by the Fed.
Mark Zandi: Yeah, I agree. That's a great statistic. Yeah, the quit rate's really important and it feels like wage growth is already rolling over, but that should really help. Is it presaged or precage?
Bernard Yaros: I don't don't know. Previewed or I, yeah.
Mark Zandi: What do you think, Matt?
Matt Colyar: I think he's got that wrong. But anyway, go ahead.
Bernard Yaros: Bursaged.
Mark Zandi: That seemed very, I don't know.
Marisa DiNatale: Sounds very French.
Mark Zandi: It's very, very French. Okay, go ahead, Bernard. You had something else you wanted to say?
Bernard Yaros: And I mean also quits directly contribute to very strong wage growth, especially in this period, because historically, the wage growth premium for switching jobs is about 0.6% percentage points. Whereas late last year, was job switchers were earning two percentage points more in wage growth compared-
Mark Zandi: Yeah, Cris has got his head cocked. I think what he means is that the growth rate is 0.6 percentage points higher points.
Bernard Yaros: Points higher.
Mark Zandi: For a switcher than someone who doesn't switch jobs. Okay. And now, it's 1.2%?
Bernard Yaros: It hit as high as two percentage points.
Mark Zandi: Two percentage points.
Bernard Yaros: Yeah. Which is just-
Mark Zandi: That's a big difference.
Bernard Yaros: Something that you've never seen and that has directly contributed to strong wage growth recently.
Mark Zandi: Right. Okay. That makes a lot of sense. So you said 30% and 50%?
Bernard Yaros: Yes.
Mark Zandi: Okay. Matt, what are your odds?
Matt Colyar: I echo a lot of what Bernard said. I'm probably a little bit less optimistic longer term, but near term, I'd put it at 30% or a third, maybe, chance. And that's a function of household balance sheets. That's the labor market, labor hoarding, all of those things that the firm seems to be doing to avoid any kind of mass layoffs. I think those things are real and they're a buffer on any downturn. Longer term, I am maybe a little bit less optimistic about wage growth coming down. So I would say the Fed may have to push the gas pedal a little harder or tap the brakes, I guess, a little harder. So I'd put that at 55% in 2024. I think there's going to be real pressure to do what they said. And even if 3.5% wage growth is just an estimate about what's compatible with inflation, I feel like they've boxed themselves in and that's where things need to be. And they're not getting a ton of help from productivity growth. So I think there's a lot of pressure to keep raising rates and eventually again set the brake.
Mark Zandi: Got it. Okay. Cris, I'm been waiting for your probabilities with intrepidation. Go ahead.
Cris deRitis: So for 23, I'm with Marisa, I think 40%. It's come down.
Mark Zandi: 40.
Cris deRitis: But still some significant headwinds there. And then for 24, I'm sticking with 65%.
Mark Zandi: Okay. You're still there?
Cris deRitis: Yeah.
Mark Zandi: It hasn't come in. You're still pretty much where you've been.
Cris deRitis: Pretty much.
Mark Zandi: And for the same reasons. Kind of what Matt was saying, that the economy remains strong enough that inflation doesn't come in fast enough and the Fed has to keep on raising rates or something else comes along and knocks us off kilter.
Cris deRitis: Yeah, I worry about the consumer. The student loan repayments are just another weight on a lot of consumer spending. And the excess savings that we've talked about in the past, they continue to dwindle. So I'm worried that we get through the end of the year and, really, consumers start to be tapped out, especially if interest rates are, even if they're not rising, even at this level, I think, it starts to weigh more and more on housing and autos and other spending.
Mark Zandi: Yeah. Okay. Well, I'm at a third for this year and a half for next year, 50% next year. So I think I'm, is that you, Bernard? I think or very close to.
Bernard Yaros: Very close to, yeah.
Mark Zandi: Yeah, yeah. Yeah. And with a bias to lowering it, I hear you. 50%, obviously, is a pretty high probability. That's any scenario pretty high. And I do think the most likely thing that does us in is the Fed, it feels like it has to continue raising interest rates to bring in the slow growth and bring in wages and price pressures. And the current fund rate target of just over 5% isn't in their minds enough that they have to keep pushing rates higher. Our terminal rate in our baseline forecast is just over 5% where we are. So that's what we expect. But the risk is that they feel like they will have to step on the brakes harder, push rates up more, and at some point, they push so hard that it pushes us into recession, makes us so vulnerable to anything else that goes wrong and we go in.
I want to go though, throw a couple thoughts out there with regard to the Fed, what they will do, and this is as opposed to what I think they should do, but what they will do. And that may be that they might not be as aggressive in raising rates, even if inflation remains a bit elevated. So what I mean by that, suppose the scenario is that CPI, I'm using CPI, because everyone's kind of uses that as the benchmark. It's right now just under 5%. It peaked at 9% back last summer. I feel pretty confident. Actually, I'd say highly confident it's going to be close to 3% by the end of the year going into the next, just simply because of the slowing and the cost of housing services, which is tied back to rents which have gone flat to down.
So say we're hanging at three, coming into 2024, here's the thing. Would the fed actually sacrifice the economy at the altar of a 2% inflation rate? When my sense is, if you ask them what inflation target they would pick, if they could pick it today, is not 2%, it's probably 3%. They probably want a 3% unemployment rate. The idea being that at 2%, given the slow real potential growth rate of the economy, if the economy ever gets into trouble, they have to lower rates so fast to get it back down to the zero lower bound pretty quickly. And if that happens, then they've got some difficult choices around negative rates or more likely quantitative easing. And they don't like to do that. So if three percent's their actual, what they desire for their target, would they actually worry too much about getting us down to 2% that fast?
Could they be a little bit more leisurely about it? And here's the other thing. That's an election year. We're talking about an election, and the Fed is independent, yes. But it's a very political animal at the end of the day. And especially it feels like this presidential election's going to be highly politicized and they don't want to get caught up in raising rates and pushing the economy and effectively potentially deciding who the next president is. Would they want to do that?
So given those two things, my thought is they might be surprisingly dovish. Inflation may not be coming in as fast as they would like, but they'd be willing to live with it because, well, for the reasons I just gave. What do you think about that, Cris?
Cris deRitis: A nice theory, but I can't see [inaudible 01:15:23]. It may very well be the right thing to do. But to your point, I think they've committed to this 2%. If they back off-
Mark Zandi: I don't know. I don't know that there's-
Cris deRitis: I guess it depends what the trajectory is. If you're saying, are we-
Mark Zandi: Is it a-
Cris deRitis: It's three and it's like it's with the downward. It's downward bent. It feels like at some point in the future, I don't know when that future is, it's not the next six months, but some point in the future it's going back to something less than three. And by the way, again, I do, I really want to get back to two. I mean really two and a half percent maybe on the CPI. I'm not sure. I don't know. Anyone else have a view on that theory I just threw out there?
Marisa DiNatale: I think they'll look at the trajectory. So if it's three or close to three, but it's been, even if it's slowly moving lower, particularly if core starts moving lower, core's been stuck for the past few months, but if that really starts moving lower, then I think they can say, long and variable lags, it's working. We can stay where we are. I think if core starts moving higher or isn't budging at all, then I think that they act.
Mark Zandi: [inaudible 01:16:44]
Marisa DiNatale: Yeah.
Mark Zandi: I don't know. Actually, the other thing I wanted to bring up, have you guys looked at, Bernard, maybe you have, because you look at these CPI numbers, carefully. CPI, excluding housing services, just take the housing services out. Have you looked at that?
Bernard Yaros: So it was a CPI for core services excluding?
Mark Zandi: No, no. Overall CPI, just excluding services. That's a.
Marisa DiNatale: Excluding housing-
Mark Zandi: Housing services. Yeah. Cost of shelter.
Bernard Yaros: Oh, okay.
Mark Zandi: If you look at it, you should look at it.
Bernard Yaros: Okay.
Mark Zandi: Look at the year over year growth. It's now into the threes year over year and headed south. Headed south. Which again, gives me comfort because I feel very confident that the growth of the cost of housing services, the cost of shelter is going to come in. Is going to come in.
Bernard Yaros: It's already peaked. Yeah.
Mark Zandi: Yeah, yeah. It's already peaked.
Anyway, okay. That was a great conversation. And anything else to add? I guess I'll throw out, we are having a conference, an in-person conference in Wilmington, Delaware on June the 20th, and we invite you, so if you're interested in attending, please get in contact with us and we will set you up with that. I think you'll enjoy it. Cris, you're speaking. I'm speaking. We've got a action-packed agenda as well, talking about lots of different aspects of the economy and different threats to the economy, commercial real estate, what's going on in the banking system, consumer credit. We talk about student loans, so I think people would find it interesting. Okay. With that, I think we're going to call this a podcast. Thanks everyone. Talk to you next week.