INSIDE ECONOMICS

Mark Zandi:                      Welcome to Inside Economics. I'm Mark Zandi, the Chief Economist at Moody's Analytics and today is Jobs Friday, the jobs report for the month of January, 2023 came out today. To help dissect this and talk about the economy and where it's headed, I've got my two co-hosts, Cris deRitis and Marisa DiNatale. Hi, guys.

Cris deRitis:                       Hi, Mark.

Marisa DiNatale:              Hi, Mark.

Mark Zandi:                      And we've got Dante, Dante DeAntonio, who is a regular on Jobs Friday. Good to see you, Dante.

Dante DeAntonio:           Good to see you, too.

Mark Zandi:                      You're looking dapper in your Penn State pullover.

Dante DeAntonio:           Thank you.

Mark Zandi:                      You went to Penn State, right?

Dante DeAntonio:           I did, yeah.

Mark Zandi:                      That's a great university. I love that university.

Dante DeAntonio:           I agree.

Mark Zandi:                      All the kids that we hire from there are just fantastic, great econ department. Dante, you're the poster child for that.

Dante DeAntonio:           There you go.

Mark Zandi:                      Can I have one of those pullovers? Do you have any extra ones?

Dante DeAntonio:           I can get you one. I'll find you one.

Mark Zandi:                      Can you?

Dante DeAntonio:           Yeah.

Mark Zandi:                      I would appreciate that, just saying.

Dante DeAntonio:           I can make that happen for you.

Mark Zandi:                      Very good. I appreciate that. Any news this week, guys? We've got a lot of news, we got a lot of data, we've got a lot to cover.

Marisa DiNatale:              There's some things to talk about.

Mark Zandi:                      We're going to do the statistics game, for sure, because we've got a lot of statistics, this is an action-packed week, and we're also going to answer listener questions. This is a new feature that we implemented since the beginning of the year, particularly when we don't have external guests, so Marisa's going to guide us through that later in the podcast. But obviously, here we are, the morning of Friday, February 3rd, the... Wow, what a jobs number. I'm really, really curious as to everyone's interpretation of this number. Dante, I'm going to begin with you. You want to give us the rundown?

Dante DeAntonio:           Sure. Obviously, a lot to unpack in the report this morning. The headline number is obviously surprising to the upside, but we also got benchmark revisions on the establishment survey, we got updated population controls in the household survey, so a lot to dig through, not only in the regular data, but also in the extra data that we get in the January reports. On the industry employment side, things were strong across the board, although I'll say the outsized gains were concentrated in a few places, really. Retail jumped in January, professional business services turned back around, and leisure and hospitality got back to where it had been in prior months after cooling a little bit last month. We also got a jump in government payrolls as a result of the strike ending and some state government employees back to work. Outside of those, though, it was, I would say, payroll gains as usual across most other industries. Most of those big gains were concentrated in those four places and so not a huge amount of news outside of that.

                                             On the household survey side, the unemployment rate obviously ticked lower again, which was probably surprising to a lot of folks, although if you unpack the details after removing the effect of the changing population controls, the household survey was certainly weaker than the establishment side in January. But other than that, it was obviously surprising to the upside, we were expecting payroll growth to continue to moderate and then that certainly did not happen this month, although I would consider this at this point more of an aberration than some start of a new higher level trend of growth.

Mark Zandi:                      Payroll, employment growth in the month of January, that's the headline number where your eye goes to first. That's the first thing the Bureau of Labor Statistics says in its press release was, "517,000 jobs created in the month of January."

Dante DeAntonio:           Correct.

Mark Zandi:                      It felt like up till this report, underlying job growth, when I say underlying, I mean abstracting from the vagaries of the data, was half that, maybe, 250K. That's what we got in December, that was on a three-month moving average basis, that's the job growth we were getting. 500K, 250K, what's reality? Is it 500K, is it 250K, or is it something else?

Dante DeAntonio:           It's certainly not 500. I think the reality is higher than 250. And

Mark Zandi:                      You do?

Dante DeAntonio:           Well, with this number and the revisions to data from the end of last year, the three-month moving average now is more like 350,000 instead of 250,000. I think that's probably still a little bit high for what underlying growth is, but I think maybe it's closer to 300 than it is 200.

Mark Zandi:                      There's so many things I want to talk about with you and unpack, because you just laid out a cornucopia of issues, most of them very geeky. I'm sure most people who heard what you said go, "What the heck's he talking about benchmark revisions and January controls and all that kind of stuff?" We'll get down and dirty with all that at some point here, but before I do that, let me turn to Marisa. Marisa, fill in the gaps here, what else would you say about this report?

Marisa DiNatale:              It was shocking. I stared at it for a while.

Mark Zandi:                      I have to say, I was eating my Wheaties, not my Wheaties. I was eating, I don't think anyone eats Wheaties anymore, but it was Wheatie-like cereal with my Wawa coffee and I did have to stop-

Marisa DiNatale:              You spit it out.

Mark Zandi:                      I had to stop and then I hit refresh, maybe because I'm on the wrong page or something, refresh, but no.

Marisa DiNatale:              It looks like a jobs report from early 2021, when we were getting gains of half a million jobs a month. I think Dante summed it up, it's broad-based on the payroll side, you have to go really into the industry detail to find any industries that actually lost jobs on net over the month. The household survey is really difficult because we can't really compare January to other months, other than December, the BLS provided a table that lets us look at some of the very top-line household survey numbers in comparison. The unemployment rate at 3.4%, that's the lowest unemployment rate we've ever seen since 1969. It's very strong, but on the other hand, we got a lot of other strong jobs data this month, which was mostly for the month of December, but JOLTS was good. I'm sure, we'll talk about this, too, JOLTS was really good.

Mark Zandi:                      Job Opening and Labor Turnover Survey.

Marisa DiNatale:              The UI claims fell yet again, so there's no reason to think it wouldn't be a strong jobs report, but to this magnitude, I don't think anyone was expecting it.

Mark Zandi:                      Well, we certainly weren't. We were doing a poll last night via email, I was on the high side, I thought, wasn't I? I was at 250K-

Marisa DiNatale:              Yeah. You picked the over on 250 or whatever we were previously at.

Mark Zandi:                      Cris was on the low side, 190, the bear, and where were you, Dante? I can't remember.

Dante DeAntonio:           220.

Mark Zandi:                      220. Where were you, Marisa?

Marisa DiNatale:              220.

Mark Zandi:                      220, okay. 250, it was double even the high for what I expected. It felt like everything was boomy in the report. It was not only a payroll gain, it was the revisions upward in previous months, it was hours worked. Did you see that, number of hour-

Marisa DiNatale:              Yeah.

Mark Zandi:                      Because that had been trending lower and was quite low and now, it just boom, right back up. Temp jobs, going back to professional services, that's where temp jobs are, temp jobs were declining and that typically is a leading indicator of future broad-based weakness in job growth. That rebounded in the month, that increased. We saw unemployment decline. You're right, I guess one point you made, Dante, about the household employment increase, that was large, that was 700,000 jobs in the household employment. That's a survey of households, the other survey that the BLS conducts and releases data for. You're saying that was overstated because of this new population estimates that they do every January?

Dante DeAntonio:           Right. They don't go back and revise the previous year, so you get this break between December and January,

Marisa DiNatale:              There was also a really big gain in December, November to December, which you can compare, that was 700,000, as well, in the household survey.

Mark Zandi:                      If you abstract from the population control issue you just mentioned, what would've the number have been, do you know?

Dante DeAntonio:           84,000.

Mark Zandi:                      Much more pedestrian. But on top of, you're saying, Marisa, this 700K in December?

Marisa DiNatale:              Yeah. That's comparable.

Mark Zandi:                      That's comparable. Cris, what do you make of the report?

Cris deRitis:                       It's confusing, it's confusing.

Mark Zandi:                      It is shocking.

Cris deRitis:                       It does point to strength, but do we buy the number at face value? I don't think so, because of all these other issues here. But every indication is that the labor market was even stronger in 2022 than we originally thought and remains quite vibrant.

Mark Zandi:                      I want to do pushback on two fronts. Let me level set by saying it does feel like the labor market is still strong, it's resilient, and that's most evident in the lack of layoffs. We're getting a lot of corporate announced layoffs, and maybe they'll start to come through in the aggregate data we get on unemployment insurance claims and everything else down the road, but it hasn't so far and layoffs in total remain very, very low. We should come back and talk about what's going on there and why that's the case. But to level set, it's clear the labor market is strong, but I don't think it's anywhere near as strong as the data suggests for two broad reasons. I'm going to throw out one and then we'll talk about them in a second.

                                             The first is it feels like there's all kinds of seasonal adjustment issues all over this report that I believe this January was, and this, of course, is the employment report for January, was the fifth warmest on record. If you have a very warm month in a month that typically, obviously, is very cold and economic activity is depressed, when you try to account for that seasonality, the Bureau of Labor Statistics adds in jobs to account for that seasonality, you get an outsized gain like the one we've got, particularly in the sectors you mentioned, Dante, it feels like. You mentioned retail, leisure and hospitality, maybe even temp jobs, I'm not sure.

                                             Also, just to add to that another layer of complication, there probably is some seasonal adjustment problems related to the impact the pandemic had back three years ago when we shut down, everyone got thrown out of work, we reopened, everyone came back to work, and you get these big swings in the data. As a result, when you use the statistical techniques to tease out the seasonal factors, the seasonality of the data, it gets very complicated to do. I'm sure the BLS has got techniques to try to work around that, but nonetheless, I'm sure it's complicating things enormously. It just feels like seasonal measurement issues are all over the report. Am I off base here, Dante?

Dante DeAntonio:           Yeah. Can I push back on your pushback a little bit?

Mark Zandi:                      Yeah, sure.

Dante DeAntonio:           I saw a few people mention seasonality this mornings, I didn't have a chance to get into the industry detail, but just at a top level, it doesn't really look like there's a whole lot funky going on with seasonal adjustments. January is an unusual month in that, on an unadjusted basis, employment declines by a huge amount every January like clockwork. It's usually between two and a half million and three million jobs lost on an unadjusted basis, and then seasonal adjustment obviously corrects that. The decline on an unadjusted basis in January this year was the smallest that we had since 1995, which fits with your story about it was a warm January, so we got a fewer job losses. That all makes sense, but the seasonal adjustment factor was not unusually large and if anything, it was actually smaller than it's been in the last few Januarys.

                                             If you look at the difference between the adjusted data and the unadjusted data, the impact that seasonal adjustment had this year was actually the smallest that it had been since, I think, 2015. It's not that there was an outsized impact. You could argue that maybe the factor was still too big, even though it is a bit smaller than it's been recently, so maybe there's something there, but it certainly doesn't stand out as a glaring issue like we had seen in some months in the earlier days of the pandemic. I think January, in particular, was not ever hugely impacted by the initial pandemic swing, obviously, because the biggest changes happened after January in 2020 and then by the time we got to 2021, things had normalized, sort of. The unadjusted declines for January don't look all that unusual, even as you look back through the pandemic years.

Mark Zandi:                      Here's another example of difficulty around seasonal adjustment. We know that people bought early for this past Christmas, they seemed to be out there shopping in October, maybe a little bit in November, and they had stopped shopping in December. That means less hiring in December, which means therefore less layoffs in the retail sector, less layoffs in January, and that would, on a seasonally adjusted basis, because the January retail numbers have come in strong. Because we know retail employment's been pretty consistently weak, losing jobs, and then all of a sudden, you get this big increase in January, it just doesn't fit. I know what you're saying, but it feels seasonal to me, no?

Dante DeAntonio:           I'm not saying it's not possible. But go ahead, Marisa.

Marisa DiNatale:              You've got to be careful with retail this month. Another thing that they did this month that we haven't mentioned yet is that they reclassified all the industries on the establishment survey. It was a much bigger than normal reclassification, it affects 10% of employment, they move stuff around. They specifically call out retail as one of the industries that's most affected by this reclassification. I would take that big retail gain with a grain of salt, not only because of what you're saying about seasonality, which I think could be true, but also because of this reclassification is mucking things up, as well.

Mark Zandi:                      Got it. Got it. Cris, you were going to say something there or were you just going to-

Cris deRitis:                       I was going to point out the [inaudible 00:15:37] classification of retail.

Mark Zandi:                      Oh, I see. The shift in industries. Here's the other pushback. We know that this data that we're observing now will be revised. The weird thing in with the January report every year is there's a so-called benchmark revision, where the payroll employment data, the data based on surveys of businesses, it's a survey of a sample of businesses, a large sample, but it's not the universe. Every January with the January data, the Bureau of Labor Statistics goes back and so-called benchmarks the survey-based data to actual employment counts based on unemployment insurance records. Every company, small, large, has to file with the UI Office how many people are working for them and that's a complete count and with every January report, they benchmark.

                                             But that benchmark is March of 2022, so it feels like way in the distance. We do know, based on other more recent data from unemployment insurance records, that the next benchmark revision, the one we get for the next January report, probably, in all likelihood, would show a pretty significant downward revision in all of the employment growth that we've experienced, at least going back to March of 2022. That probably will affect this number, as well. Those revisions look like they could be quite significant. The Q2 data, based on unemployment records, show a very weak job market. The Business Employment Dynamics survey, which came out last week, which is using Q2 data from last year, that showed an actual decline in employment in the second quarter. I don't think it's going to decline, but nonetheless, it's making a point that all this data is going to be revised down. Is that right? Did I explain that right? Do I have that right, Dante, and does that color your perspective on your statement that underlying job growth is 300K?

Dante DeAntonio:           It's certainly right for Q2 of 2022. We don't have that hard count of data beyond Q2 at this point, so that could obviously reverse course in the second half of the year, potentially, in the QCW once we get it. I think Q2 certainly was weaker than what we saw initially, but I don't know that I would lock that in to say that the rest of 2022 was also weaker than what we saw. I don't think that's a guarantee at this point.

Mark Zandi:                      What do you think, Cris? Sorry, Marisa, you were going to say something.

Marisa DiNatale:              There is evidence that certainly, the job market weakened in the summer, but then strengthened again. We were looking at UI claims trended up, they were above 260,000 in July and then they started coming down again, so that is consistent with a weaker Q2, but we don't really know beyond, as Dante said, Q2.

Cris deRitis:                       There are just so many crosscurrents here, it's hard to [inaudible 00:18:47]. I have to say, ADP, I don't know what's going on there, but-

Mark Zandi:                      Oh, that was weak.

Cris deRitis:                       ... that colored was my thinking. It was very weak.

Mark Zandi:                      That came in at 100,000 or something close to 100,000 for the month of January. Just level set, ADP is a private payroll processing company, they construct their own estimates of employment gains in the month based on their records and they showed a small gain, 100K gain, I think, roughly. That's private sector, it doesn't include government, but even if you threw in government, it would be 150K, let's say, so much weaker than... That's well below the consensus, even, kind of number.

Cris deRitis:                       That's right.

Mark Zandi:                      That colored your-

Cris deRitis:                       I put too much weight on that. But I forgot that-

Mark Zandi:                      Dante will tell you firsthand not to put much weight on that.

Cris deRitis:                       They cited bad weather.

Mark Zandi:                      You said they cited bad weather? Oh, that's right.

Cris deRitis:                       They cited bad weather.

Mark Zandi:                      Oh yeah, that's right, they did. Maybe because of the storms up in New York, that Buffalo storm in the survey week or something.

Cris deRitis:                       Oh, it could be, but-

Mark Zandi:                      It could be. There were a couple things in the report that were more, I guess... Positive and negative is all warped now, because at any time in the past if you said 500K jobs, unemployment at 3.4, you go, "Woo hoo, that's good news." But in the current context, it's a little disconcerting because it means that the labor market's tight and that would put more pressure on the Federal Reserve to continue to raise interest rates. If they continue to raise interest rates, at some point, the economy's going to break and you go into recession. We want the labor market to cool off here and give the Fed reason to stop raising rates sooner rather than later. If you took this data, most of it, at face value, you'd say, "Oh, the Fed's going to keep raising the rates," and we'll come back to that in just a second.

                                             But a couple things in the report were more consistent with the Fed should be a little bit more likely to take its foot off the brakes here, or at least not keep pressing on the brakes as strong, and one was average hourly earnings. If I asked a Federal Reserve Board member to rank order all the statistics in the report in terms of what's important in terms of their thinking about monetary policy and interest rates, I don't know if they would put average hourly earnings at the top of the list because it's got its own measurement issues beyond seasonal adjustment, but it would be pretty close to the top of the list at this point, because that's the key link between what's going on in the job market and inflation, because higher wage growth effects, cost of doing business for service-based companies and they pass that along in the form of higher prices for their services, healthcare, hospitality, education, that kind of thing.

                                             The one meaningfully positive thing in the report was average hourly earnings, that came in modestly. I think year-over-year, it was 4.4%, month-to-month annualized, it was 3%, 3.5%, which is pretty close to where you'd want it to be consistent with 2% inflation. Anyone have a different perspective on that or want to add some color to that? Any perspectives on that? Cris, do you have anything else to add to that?

Cris deRitis:                       No, I agree with you. We'll get to the Fed, I think. They're all concerned about the wage implications, but Powell cited the unemployment rate, the actual state of the labor market jobs, as a primary factor that he's watching. I think it's good that this is supportive and could lead to a decision to moderate, but I don't think that's the case. I think they're still gung-ho because of the strength labor market, even if you revise down the 500K, I think that's still going to point to the case to raise.

Mark Zandi:                      Well, let me ask you, Cris, I asked Dante what he thought underlying job growth was. Again, abstracting from the vagaries and measurement issues in the data, what do you think it is, what is underlying job growth?

Cris deRitis:                       Before I saw the report, I thought it was 200K.

Mark Zandi:                      200K, okay.

Cris deRitis:                       Something like that.

Mark Zandi:                      Now, what do you think?

Cris deRitis:                       I still think that's the case, I want it to be that I'm right and this will be revised, but-

Mark Zandi:                      This will be revised.

Cris deRitis:                       But again, I'm discounting data because it doesn't meet my narrative.

Mark Zandi:                      Well, sometimes you've got to do that because, and this is an economist now speaking to you with 30 years of experience, every once a year, maybe once every two years, you get a wacko report that this is out of bounds and it is out of bounds. It's not reality, it's just the vagaries of the data and you have to discount it. There are times when you have to discount what you're seeing because there are measurement problems.

Cris deRitis:                       I'm discounting this, you're discounting the yield curve, I think we're even.

Mark Zandi:                      Very good. Few people understood what that meant, but I did.

Cris deRitis:                       It's an inside joke.

Mark Zandi:                      That was like a dagger into the belly. We can come back to that. Marisa, same question, what is underlying job growth, in your mind?

Marisa DiNatale:              I think it's around 250.

Mark Zandi:                      250.

Marisa DiNatale:              Yeah.

Mark Zandi:                      Very good.

Marisa DiNatale:              I don't know what else to say.

Mark Zandi:                      Well, that's fair. That's my view, I think it's 250K, with a bias lower, because I do think we are going to get downward revisions to this data when it all comes in. It would be very weird for Q2 of last year to be weak and then for things to bounce right back up. Yeah, it might bounce, but it's not going to bounce that much, so I think we'll get some pretty meaningful downward revisions to the data. Do you discount the JOLTS data, I guess this is everyone, as well?

Marisa DiNatale:              And the UI claims data?

Cris deRitis:                       No, no, no, no, no.

Mark Zandi:                      Job openings?

Cris deRitis:                       No. Well, the unfilled positions, I don't know what to make of that, I discount that. But I think it's very clear, layoffs are extraordinarily low. I think that's evident in the JOLTS, that's evident in the UI claims. There are seasonal adjustment issues with that data, too, so we've got to be careful with the UI claims week-to-week, there are seasonal issues. But even with that, we're around 200K in UI claims each week, that is really low. That is really low.

Marisa DiNatale:              In today's jobs report, if you look at the composition of the duration of unemployment, again, I know this is a little bit skewed by the pop controls potentially, but the share of people that were unemployed for less than five weeks, so this is very recently unemployed people, people that just lost their job, it fell quite a bit, so it's consistent with lower UI claims.

Mark Zandi:                      I do think the job market is strong, there's no doubt about it, and resilient. Of course, like most things, I'm looking at it through a glass half full kind of perspective, and that is the lack of layoffs is a reason to be optimistic that the economy will not go into recession. Again, we've talked about this in the past, I have a hard time thinking that the economy goes into recession if we don't have a lot of layoffs, 300K in UI claims per week, not 200K in UI claims per week, so we're a long way from that. That, to me, is coming through loud and clear. I think we are seeing moderation in job growth, not because of layoffs, but because of just less hiring, I think businesses are hiring less.

                                             Here's how I would square the JOLTS data, I'm curious to hear what you think about this. Quits are still elevated, people are still quitting their jobs. When a person quits their job, that generally creates an unfilled position. Most companies will not eliminate that position, at least not quickly. That sits there as an unfilled position. But if I'm reducing my hires, I'm just slow walking hiring and I'm just not hiring as aggressively because I'm nervous about sales, and I don't want to lay off workers because I know I'm going to need them down the road, but that's different than not hiring and filling open job positions. I'm just taking my time to fill open job positions. You could get what you observed in the JOLTS data, elevated quits, elevated unfilled positions, weaker hires, and overall job growth that's still strong, again, it's still strong, but moderate. Does that resonate with folks?

Marisa DiNatale:              Yes.

Cris deRitis:                       It's just that the job openings are soft, they're not hard openings.

Mark Zandi:                      They're not hard, or they're temporary or it takes a month or two or three to work them through.

Cris deRitis:                       Or it's a free option. Why not have those?

Mark Zandi:                      There's no cost to it.

Cris deRitis:                       If someone comes around-

Mark Zandi:                      Why wouldn't you do that, why wouldn't you keep the open position?

Cris deRitis:                       Especially given this uncertainty. If we don't go into a recession, then you're going to need to ramp up quickly.

Mark Zandi:                      I'm speaking as part of a large multinational organization with a big HR department, fantastic HR department, let me just make that clear, I love the HR department. But if you think what we're going through is relatively temporary and it's not going to be that severe, and I think that's generally the case, you don't want to shut down your HR function, because if you shut down your HR function, it is incredibly painful to get it back up and running again. You've got to hire HR people, you've got to train HR people, you've got to get them going again, you've got to get them collecting resumes again.

                                             Stopping the process of HR hiring, that's a big deal in most big companies, so I don't think big companies are doing that. They're saying, "Hey, guys, take your time, don't hire this quarter, let's see what the next quarter looks like," that kind of thing. But we're not going to take those unfilled positions down. We still want to see the resumes, maybe we do want to do some soft interviews, take our time here, but we don't want to shut this thing down. That's what it feels like to me, at this point.

Marisa DiNatale:              Except-

Mark Zandi:                      Go ahead.

Marisa DiNatale:              Sorry. I agree with that characterization. I think you have to look beyond job openings, it's our company's actually hiring people. To say that hiring has weakened quite a bit, it has since the beginning of 2022, but it's-

Mark Zandi:                      Well, let me put it this way, all of the adjustment in the labor market is less hires. We know job growth has slowed.

Marisa DiNatale:              That's right.

Mark Zandi:                      It has slowed. We were 500, 600K a year ago, I don't know what the underlying trend is. You said 250, that sounds about right to me, so we have slowed. We know layoffs are very, very low, relative to pre-pandemic, in JOLTS, in UI, they're very low. All of the adjustments that's occurred so far is hires. That's where the adjustments occurred, all of it's been in hiring. Again, to me, that feels like the most graceful way for the labor market to adjust, a way that would help us avoid an outright economic downturn.

Dante DeAntonio:           The one other note on JOLTS that I thought was interesting, I thought it was maybe just a weird data point at the time. I think like you, I discount the level of job openings at this point. It's not a good comparison to pre-pandemic, but that big increase that happened in December was almost entirely concentrated in retail and leisure and hospitality. Openings across every other industry were basically flat or even slightly down and the big jump was in those two industries, and that aligns with what we saw in terms of a jump in actual job gains in retail and leisure and hospitality. Again, it doesn't guarantee that it's right, but it does give an extra data point to say that maybe demand did pick up a little bit in retail and leisure and hospitality at the end of the year more than we would have expected.

Mark Zandi:                      I missed that. Are you saying a bulk of the unfilled positions are in leisure and hospitality?

Dante DeAntonio:           The increase in December, we saw that-

Mark Zandi:                      Oh, the increase in December.

Dante DeAntonio:           ... [inaudible 00:31:30] in December, it was entirely captured in retail and leisure and hospitality.

Mark Zandi:                      You're sitting at the Fed, obviously looking at all this data, trying to figure it out and trying to digest it and make a decision around what it means for the conduct of monetary policy. What do you do with this? How do you handle this? What do you think about this? Cris, what do you think?

Cris deRitis:                       What do we do or what does Jay Powell do?

Mark Zandi:                      What does Jay Powell do? Or Christopher Waller or Lael Brainard, what do they do with it?

Cris deRitis:                       I'm asking because there's what you might think they should do in theory versus what they've already declared that they will do, and I think they raise. 4.4% on the average reality earnings, while improving, is still too high, so I think they'll continue to step on the brake.

Mark Zandi:                      Interestingly enough, I think the markets, the equity market, the bond market, futures market for fed funds, is in the camp we're sort of in, that maybe it shows the labor market's more resilient than we thought, but I discount it big time, because the stock market's down, bond yields are up, but not a lot compared to where they were. The markets seem to be saying, "Hey, this is not that big a deal," at least it terms what it means for monetary policy. It feels like the markets are still saying another quarter point rate hike in March and then, I don't know, are they starting to discount one now in May, as well? Before this report, they were saying just one more in March.

Cris deRitis:                       I think they're split.

Mark Zandi:                      They're split.

Cris deRitis:                       It moves around a lot.

Mark Zandi:                      If you were on the Fed, would you concur with that one more maybe quarter point in March, another quarter point in May, and pause at that point and see what happens?

Cris deRitis:                       Yeah. I think the market is right here, in terms of the March is almost guaranteed, I would say, and then the next May would be dependent on what the data shows between now and then, so that 50/50 split probably makes sense, too. Pause, but I think the market might be wrong or some market participants may be wrong in the sense that they're anticipating these cuts at the end of the year and I don't see it.

Mark Zandi:                      What you're referring to is that if you look at the futures for fed funds, the rate the Fed controls, they have the rate going up to around 5 and then staying there for maybe the first half of the year and then starting to come back down in the second half of the year going into 2024. Of course the Fed's telling them, "That's crazy, we're not going to do that," and you're saying that doesn't make sense to you either, that doesn't, or-

Cris deRitis:                       They're challenging the Fed, I guess. They must be assuming that the inflation will be coming in much faster than it is or they're anticipating recession. What do you think, Marisa, if you were on the Fed, how would you digest all this?

Marisa DiNatale:              I wouldn't like this report, but I agree, there's certainly... In the minutes from the meeting the other day, Chair Powell said, "We're pretty much definitely going to keep raising rates," he did not indicate a pause. They're certainly going to raise in March, I think. I think it's just really data dependent. See what we get on the CPI and the other February jobs report will come out before they meet again, so I think they're just taking it report by report here. But I would be a little confused by today, certainly.

Mark Zandi:                      Like everyone else. Dante, if you were on the Fed, how would you think about all this?

Dante DeAntonio:           I think the best thing is that they can essentially ignore this report, because they have so much other data coming before they meet again, so I think that's a good thing. I'm glad this report didn't come a week before the next meeting, where they have to use this as their only point to determine off of. We can hope that February turns back closer to trend and gives them more of the same story we've been seeing at the end of last year as opposed to what we saw today.

Mark Zandi:                      I totally agree with that. That's what I tweeted, right, Dante, "Ignore this report."

Dante DeAntonio:           I think you were saying ignore it because you didn't believe it, not because it doesn't matter to the Fed, I think is what you were-

Mark Zandi:                      Oh yeah. Well, I discount it because I do think there's all kinds of measurement issues going on here. Again, it's one of those reports that happens every once a year or twice a year where it's just this happens and it gets washed out in subsequent data. I totally agree with you, thank goodness it happened two days after the Fed met, because now they've got, as Marisa pointed out, they've got at least one more jobs number and two Consumer Price inflation reports between now and then that they can digest and help set policy at the May meeting. But I do agree with you, Cris, that at least a March increase, I should say, the next meeting's in March, another... The one in May, I say that's 50/50. I may agree with the markets there, but that's data dependent, and then they sit and wait. I just don't see them cutting rates later in the year, I just don't see that. That would require a recession, I think, and I just don't see the kind of a recession that would cause them to do that.

                                             Let's play the game.

Cris deRitis:                       We agree.

Mark Zandi:                      It's to 260. What's that? We agree.

Cris deRitis:                       We agree. Wow.

Mark Zandi:                      On no recession? No.

Cris deRitis:                       Wait. Whoa, whoa, whoa.

Mark Zandi:                      That's taking it a little too far. We'll come back to that. We'll come back to that. Let's play the game, the statistics game. The game is we all put forward a statistic, the rest of the group tries to figure it out with questions and clues and deductive reasoning. The best question is one that's not so easy we get it immediately, not so hard that we never get it and of course, if it's apropos to the topic at hand or recent data, that would be a bonus. With that, I think, as I said last podcast, it's tradition to start with Marisa, so Marisa, you're up. What's your statistic?

Marisa DiNatale:              My statistic is +440% in January.

Mark Zandi:                      +440%. Is it in the jobs report?

Marisa DiNatale:              No.

Mark Zandi:                      Is it in a report that came out this week?

Marisa DiNatale:              Yes.

Mark Zandi:                      It's an economic statistic?

Marisa DiNatale:              Yes.

Mark Zandi:                      This is not completely fair, but is it in the JOLTS report, the Job Opening Labor Turnover Survey report? It is not.

Marisa DiNatale:              No.

Mark Zandi:                      440%. Guys? Is it a price, some kind of price measure?

Marisa DiNatale:              No.

Mark Zandi:                      No. Is it a labor market measure?

Marisa DiNatale:              Yes, it is.

Mark Zandi:                      Oh, it is a labor market measure. Is it in the UI claims?

Cris deRitis:                       The claims?

Marisa DiNatale:              No.

Mark Zandi:                      Oh my goodness. She's dug deep into the bowels of some labor market report.

Dante DeAntonio:           Is it in the ECI?

Mark Zandi:                      Oh, ECI, Employment Cost Index?

Marisa DiNatale:              No, it's not.

Mark Zandi:                      No, it's not.

Dante DeAntonio:           What else came out this week?

Mark Zandi:                      Labor market related, unemployment-

Cris deRitis:                       It came out this week.

Mark Zandi:                      Is it in one of the regional manufacturing surveys. No, it's not.

Cris deRitis:                       Should we know this?

Mark Zandi:                      Should we know this? Yes. Is it really something that we should know?

Marisa DiNatale:              It's labor market related, it came out this week, it is covered by us in real-time on economic view.

Mark Zandi:                      Oh my gosh.

Marisa DiNatale:              It is not a statistic we typically pay much attention to or talk about, though.

Mark Zandi:                      I see. Dante, do you have any-

Cris deRitis:                       Not JOLTS or UI?

Mark Zandi:                      No, she said no. Not ECI-

Marisa DiNatale:              Should I just tell you?

Dante DeAntonio:           Is it something from The Conference Board?

Marisa DiNatale:              No.

Mark Zandi:                      The Conference Board survey of confidence came out this week. Goodness. Are we going to be embarrassed when you tell us?

Marisa DiNatale:              I don't think so.

Dante DeAntonio:           Oh, Challenger. You're talking about job cuts-

Mark Zandi:                      Challenger, the Challenger report.

Marisa DiNatale:              Yeah.

Mark Zandi:                      Oh, oh, oh. That's a good one.

Marisa DiNatale:              This is the percent increase over the year in the number of announced layoffs in January. The number of announced layoffs was about 103,000 in January and if you just look a month prior, that compares to 44,000 in December. According to Challenger, there was a huge uptick in the number of announced layoffs in January. That's the largest January reading since 2009, when we were coming out of the financial crisis.

Mark Zandi:                      Over 100K and what was the average monthly increase in 2022? It felt like it was-

Marisa DiNatale:              It was 30,000 a month-ish.

Mark Zandi:                      Of course, there's a lot of tech, I guess, healthcare, I saw, retail, maybe.

Marisa DiNatale:              It was retail, finance, healthcare were the biggest. Actually, the largest contributor was tech. They announced, and this is announced, so this is an actual people being laid off, 42,000 layoffs in tech in the first month of the year, and that was 40% of all of the announced layoffs.

Mark Zandi:                      Well, that's a good one. How do you square that number with the very low layoff totals in the JOLTS and the very low unemployment insurance claims? How do you square all that?

Marisa DiNatale:              I think that the job market is so strong that people are being laid off, but they're being rehired somewhere else very quickly. We talked about this a little bit on the last podcast, but if you look at the labor force flows in the household survey, so you can look at people's status from month-to-month following the same people. The number of people that are remaining employed from month-to-month is trending higher and is elevated from where it was, if you go back to 2018, 2019. Not only are people keeping jobs at a higher pace, if they are getting laid off, they're finding work really quickly, probably so quickly that they're not even filing unemployment insurance claims. I was going to say, if you look at the details of the payroll survey in January, you do see declines in banking, declines in the information services if you go under the super sector level. That is consistent with these layoff announcements that there are net job losses in some of these very detailed industries, but the rest of the job market is so strong, I just think people are becoming employed again very quickly.

Mark Zandi:                      Could it also be by case-

Cris deRitis:                       I have a question-

Mark Zandi:                      Oh, sorry, go ahead, Cris.

Cris deRitis:                       I was wondering on those layoff announcements, this has always nagged me, do we know that those are primarily or exclusively US-based, or could it be if Microsoft says, "We're going to lay off 10,000 people," and they're in India, does that get captured in this layoff data or not?

Marisa DiNatale:              I think it could, because they do say that... The way Challenger does is I guess they see an announcement in the news or something or in a warn notice and they verify it with the company, but the layoff could be either at a specific site or it could just be attributed to where the corporate headquarters are. They don't make that distinction. You see a lot of layoffs, for example, in this report, you see a lot of them in New York, the Bay Area, Seattle, but that doesn't necessarily mean that's where the people are being laid off. I think it could be outside the US, I don't know. Dante, do you know?

Dante DeAntonio:           I don't know for sure. I'm guessing they don't have any way to guarantee that it doesn't happen. My guess is if there's any specificity in the announcements, then obviously, they'll take account of that. But I don't think there's any way they can guarantee that it doesn't bleed over into other-

Cris deRitis:                       I think generally, you just give a number in those corporate announcements.

Mark Zandi:                      This is a more pessimistic perspective, or optimistic, depending on your point of view, but could it be that it's just a matter of time, that it takes time for these announced layoffs to actually be implemented and then because of severance issues, maybe it doesn't show up in the UI claims right away, it just takes time for it to filter through? This is year end, beginning of the next year, that's when corporations tend to make these big changes and then it takes a few months for it to show up in the data. Is that a possibility?

Marisa DiNatale:              I think it's a possibility. But we've now been hearing about massive, at least in the headlines, we've been hearing about massive layoffs, particularly in tech for, I don't know, it seems like maybe six months, four to six months now. You would think that, certainly, that many, many people have been laid off, so we would be seeing it in the data, I think, by now. It could be gathering momentum, they could be getting bigger, and eventually, they'll become so big that these people can't be absorbed very quickly, but-

Dante DeAntonio:           We did see a little bit of an uptick in UI claims back in November, when the first wave of tech layoffs were announced. It was not big, they never got above 230,000, 240,000, and it pretty quickly came back down. I do think over the next couple of weeks, we could see claims come, again, they're below 200,000 now. I think certainly it will come back up a bit, but I still don't think I've seen enough happen that would cause them to spike above 240, 250.

Mark Zandi:                      Dante, you're up. What's your statistic?

Dante DeAntonio:           I'm between two. I'm going to go with -5,000.

Mark Zandi:                      -5,000. Is it in today's jobs report?

Dante DeAntonio:           It is.

Marisa DiNatale:              Is it an industry?

Dante DeAntonio:           It is not.

Mark Zandi:                      Is it in the payroll side of the survey?

Dante DeAntonio:           No.

Mark Zandi:                      It's in the household side. Something in the household employment survey fell by 5,000, jobs?

Dante DeAntonio:           Not jobs, no.

Mark Zandi:                      Not jobs. Ooh, okay.

Marisa DiNatale:              Have you adjusted this for the population control?

Dante DeAntonio:           Yes, it has been adjusted for population controls.

Mark Zandi:                      Oh, okay. Is it something in the labor force?

Dante DeAntonio:           Yes.

Mark Zandi:                      Yes.

Dante DeAntonio:           Change in the labor force, if you adjust for population-

Marisa DiNatale:              Oh, right.

Dante DeAntonio:           The published change is very large, but they're saying all of that change is due to the shift in population controls. If you take that adjustment out, they estimate the labor force actually shrank slightly in January.

Mark Zandi:                      The labor force, it was basically flat in the month of December.

Dante DeAntonio:           It was flat. We have obviously seen pretty strong labor force growth here in recent months. Again, it's one number, it could be an anomaly, but certainly, that would not be a positive sign for how things are going to go moving forward.

Mark Zandi:                      Can I ask, though, and you may not know the answer to it because you haven't done the calculation, but if I look at year-over-year on the population control adjusted labor force, what that increase is? Looking at the unadjusted, I saw almost 300,000 per month. It was 2.6 million divided by 12, I guess that would be 230,000 per month in labor supply. But you don't know what that number is? You didn't do that calculation?

Dante DeAntonio:           No, they don't publish a fully adjusted back year, they just do that one-off.

Mark Zandi:                      That's right.

Dante DeAntonio:           They only adjust the December.

Mark Zandi:                      But if I take that number adjusted for the population control compared to what it was a year ago, that wouldn't be fair to do, you couldn't do that? It feels like you could do that, give you underlying a sense of the underlying growth in the labor force.

Dante DeAntonio:           Well, you can still look at within 2022. Obviously, the underlying trend in labor force growth is still very strong, [inaudible 00:48:52] here in January, that doesn't flip the script of what we have seen for the last year. But certainly, if we get more readings of flatish labor force growth, that's going to make it tough to keep things going, I would think.

Mark Zandi:                      Because I have taken some solace in the strong labor force growth, it's labor supply. Labor demand is strong and I concur. I think underlying labor demand job growth is 250K per month, that's what it feels like to me. That was pretty close to what I thought underlying labor force growth was, so I take some encouragement in that, which means if demand and supplier are roughly equal to each other, the labor market's not getting any tighter. It's not easing up, but it's not getting any tighter, and I took some solace in that.

Dante DeAntonio:           I think that's still the right [inaudible 00:49:39] to take-

Mark Zandi:                      Way of thinking about it.

Dante DeAntonio:           ... unless we get more downbeat readings on the labor force.

Mark Zandi:                      Cris, what's your statistic?

Cris deRitis:                       This one's easy, but it's important.

Mark Zandi:                      Easy. Whoa, whoa. It's easy, but important?

Cris deRitis:                       Yes.

Mark Zandi:                      Is it 3.4%?

Cris deRitis:                       No, it's 3%.

Mark Zandi:                      That's the unemployment rate for... No. 3%.

Marisa DiNatale:              Oh, it's in the ECI?

Mark Zandi:                      In the ECI, is it?

Marisa DiNatale:              No?

Cris deRitis:                       No.

Mark Zandi:                      No. Is it in the labor market? Is it in today's jobs numbers?

Marisa DiNatale:              I guess it's not that easy.

Cris deRitis:                       Now, it's just embarrassing.

Mark Zandi:                      Now, it's just embarrassing.

Cris deRitis:                       No, it's not in the labor market report, the employment report.

Mark Zandi:                      It's not. 3%. Oh, I think I know what it is. Oh, you're saying it's not in the labor market report? I was going to say it's 3...

Marisa DiNatale:              Is it a labor market statistic?

Cris deRitis:                       It is... No, not really.

Marisa DiNatale:              It is no.

Dante DeAntonio:           Was it released this week?

Cris deRitis:                       Well, maybe it is. It is, it is a labor related statistic. It's not related to-

Mark Zandi:                      How can the answer to that question be so hard? Either it is or it isn't.

Cris deRitis:                       It's not wages or employment.

Mark Zandi:                      It's not wages or employment. 3%. Was it in the JOLTS?

Cris deRitis:                       Nope.

Mark Zandi:                      Because you said it was labor market related.

Cris deRitis:                       It's kind of labor related.

Dante DeAntonio:           But he's not sure it's labor market related.

Marisa DiNatale:              Is it a statistic that was released this past week?

Cris deRitis:                       Yes.

Mark Zandi:                      Dante knows it, because he covers it on EV. He probably writes the-

Cris deRitis:                       Oh no, I don't know that he covers it on EV, but we've discussed it in the past, it's one of our debates.

Mark Zandi:                      Oh goodness.

Dante DeAntonio:           Oh, productivity growth.

Cris deRitis:                       Yes.

Mark Zandi:                      Ooh.

Marisa DiNatale:              Oh yeah.

Dante DeAntonio:           Like this jobs report, I'm just discounting it. It's probably not a believable number, so I've put it out of my mind.

Mark Zandi:                      You're right. That was a good one and we should have gotten it right away. Explain, why'd you pick that number?

Cris deRitis:                       3% output per hour, that's the annualized growth for a Q4 of 2022, that was strong. It had been very weak last year overall, but that boost in productivity growth, that's a very positive sign, that gives us a chance to get through this with just a slow session versus a reset. We need that type of productivity growth and that can help to justify some of the higher wages we have. If that is true and that is sustained, that bodes well for the future. The 500K of unemployment though, would work against that. If we're adding a lot of people and we don't expect to add a lot of output, that could reverse things, so yet another reason, perhaps, to be cautious when looking at that 500K number. But 3% productivity growth, again, very strong. I think that's too strong, I don't think that's trend, but it is a very positive sign after a long period of very weak and negative productivity growth.

Dante DeAntonio:           That positive number kept it above the pre-pandemic trend. It looked like it might crash below the pre-pandemic trend and that kept it above, if only by a little bit, so that's obviously a positive, too.

Mark Zandi:                      I think we were talking about this with Jason Furman when he was on the podcast, I guess that was last week or the week before, and he was pointing out that average annual non-foreign business productivity growth since the pandemic hit is 1.5% per annum, and that's exactly equal to the non-foreign business productivity growth in the three years prior to the pandemic. It's been three years since the pandemic hit, in that three years, it's been 1.5%, in the prior three years, it was 1.5%. It feels like pretty strong evidence that underlying productivity growth, abstracting from the ups and the downs and all arounds in this data, is 1.5%, that's what it feels like.

Dante DeAntonio:           I think that's the hope. But you're right, you had this huge spike and then you've had this big decline, and then you've got the first reading where it looks like maybe it's starting to stabilize.

Mark Zandi:                      Because you're in the 1% camp, we're going back to 1% or something, as I recall.

Dante DeAntonio:           Maybe not all the way to 1. I'm in the below 1.5 camp, where I thought that the data we've seen recently is more telling than just volatility.

Mark Zandi:                      That was a good one. That was a really good one. Mine, you ready?

Marisa DiNatale:              Ready.

Mark Zandi:                      I'm sure whether this is hard or easy, I think it might be on the hard side, but it's important. It's two statistics related, 4% and 3.7%. 4% and 3.7%.

Marisa DiNatale:              Are these unemployment rates?

Mark Zandi:                      No.

Marisa DiNatale:              Oh, they're wage-

Mark Zandi:                      Yes.

Marisa DiNatale:              ... growth rates over the year.

Mark Zandi:                      Not over the year.

Dante DeAntonio:           Annualized growth rates.

Marisa DiNatale:              Three months annualized.

Mark Zandi:                      What wages are we talking about? You can spit it out.

Cris deRitis:                       ECI?

Mark Zandi:                      ECI, Employment Cost Index.

Marisa DiNatale:              Is it your wage and salary private workers, ex incentive occupations?

Mark Zandi:                      It is indeed, very good. Which one is of those, 4 or 3.7, is that, do you think?

Dante DeAntonio:           That's 4.

Mark Zandi:                      No, it's 3.7. It's 3.7. It was one or the other.

Marisa DiNatale:              This is what, annualized over the past three months?

Mark Zandi:                      Yep. Annualized in the fourth quarter.

Marisa DiNatale:              In the fourth quarter.

Mark Zandi:                      What's the 4%? The 3.7 is total industry workers, wages, and salaries, excluding incentive pay, which I'll come back and explain to everybody why that's important, but what's the 4?

Dante DeAntonio:           Total comp annualized, isn't it?

Mark Zandi:                      Yeah. Total ECI, total comp, that's wages, salaries, benefits, the whole shoot and match across all civilian workers. 4% annualized, and this is the Employment Cost Index, it's the best measure of wages we have because it controls for the mix of occupations and industries in the labor market. The reason why we don't focus on it often is because it's quarterly data, it only comes out once a quarter, not every month. We just got that data for the fourth quarter of last year this week. I took a lot of comfort in that data. Top-line ECI, Employment Cost Index across all workers grew 4% annually in the quarter. Now, you can't put too much weight on any given quarter and if you look at it year-over-year, it's still 5% year-over-year, but it definitely feels like it's moving in the right direction here.

                                             The one measure in the ECI report that I think is most representative of underlying labor market pressures and wage growth is wages and salaries for private industry workers, excluding incentive pay, so you don't want to include sales bonuses and that kind of thing. That came in quarter-to-quarter, fourth quarter, annualized at 3.7%. Again, year-over-year, it's still elevated around 5%, but it's moving in the right direction. This weakening that we're observing in wage growth, and you see it in the average hourly earnings data that we just got today, as well, is consistent with the theory that we've been talking about, that the surge in wage growth that we got back a year ago is more related to the jump in inflation expectations that occurred when Russia invaded Ukraine and we saw this spike in energy, oil, gasoline prices, food prices, and less with the tightness in the labor market.

                                             The labor market is tight, 3.4% unemployment, but that's kind of where the unemployment rate was pre-pandemic and we didn't have extraordinarily strong wage growth and we thought labor market was tight, but it wasn't overly tight, and this would suggest that that's the case here. If 3.4% was the problem, then wage growth would not be decelerating here, it would be accelerating, and it's decelerating. That's consistent with the theory that, well, now the worst of the fallout from the Russian invasion is behind us, oil prices have come in, gasoline prices have come in, inflation expectations have come in, and wage growth is moderating consistent with that, so I take a lot of solace in that.

                                             Of course, that's what this Fed is most focused on right now, it's the cost of services, excluding housing. They're focused on that because that's what they feel like they can have some impact on with interest rates affecting the labor market, wage growth. These industries are obviously very labor-intensive and price increases in those industries are tied back to the cost of labor. If I add that all up, it just feels consistent with the story, the narrative, that we can get inflation back in the bottle without having to experience a significant decline in employment or a big increase in unemployment. Now, that was a long soliloquy, I think based in data, what do you think? Does that resonate? I know I've said that a few times, "Does it resonate?" But I need confirmation from my guys, my team, that I'm on the right track here. What do you think? Does that make sense?

Dante DeAntonio:           Yeah. I think if there's something you're going to hang your hat on here it's wage growth. Obviously, you want to discount the strength in job growth and assume that that's going to get washed away and the numbers, but if you see wage growth continuing to come in, I think that's even more important than job growth moderating as quickly as we might have wanted it to, so I would agree with that.

Mark Zandi:                      What do you think, Cris? That's not exactly how you would frame it, but what do you think?

Cris deRitis:                       Oh, I think that's the goal, that's the reasonable path. My fear, as usual, is that there's still some other shock to occur here or the Fed is going to be stubborn and wait too long to adjust. But otherwise, I agree, those data points are pointing in the right direction.

Mark Zandi:                      Marisa?

Marisa DiNatale:              I think you're right about inflation expectations, both coming from market signals, but also consumers, who have been much more pessimistic overall about the prospects of the economy and inflation, but even those have been getting better for the past few months, so I agree.

Mark Zandi:                      Let's turn to listener questions. This is a new feature of the podcast, we've tried this out three times, we'll give it another shot. We've been collecting questions. Please, listener, fire away. If you have questions you'd like to post to the group, please do. You know how to get ahold of us through Twitter or LinkedIn, our website's helpeconomy@moodys.com, please feel free. For this part of the podcast, I'm going to turn it to you, Marisa, because I know you're the keeper of these questions. I should say, correct me if I'm wrong, Marisa, we have not seen these questions, these are de novo.

Marisa DiNatale:              No, you haven't. I have, obviously, but you guys haven't.

Mark Zandi:                      Fire away.

Marisa DiNatale:              There's a lot of good ones. Let me package a few of these, because there's a lot of questions about the yield curve and the 10-year yield-

Mark Zandi:                      Oh no. Oh god.

Marisa DiNatale:              ... and how should we interpret the 10-year?

Mark Zandi:                      This is plaguing me, this damn yield curve. Go ahead, ask the question.

Marisa DiNatale:              The question is what's going on with the 10-year yield? How do you interpret it? How can you understand the behavior right now of long rates?

Mark Zandi:                      Great, great question. Cris, do you want to take a crack at that first. I already did my soliloquy, I don't want to give another one. I can do it, but maybe you go first and I'll comment on what you say. This is the yield curve. What is the yield curve actually saying here, Cris?

Cris deRitis:                       The yield curve is saying recession risk is elevated. The difference between the 10-year Treasury and the 2-year Treasury, or the 10-year and the three-month Treasury, or your favorite, I guess, is the 10-year versus the fed funds rate, all deeply inverted at this point.

Mark Zandi:                      The short-term rates are higher than long rates, they're inverted.

Cris deRitis:                       Correct.

Mark Zandi:                      Historically, when that happens, recessions follow.

Cris deRitis:                       That's right. That's right.

Mark Zandi:                      You're saying just take this literally?

Cris deRitis:                       If you take it at face value and you just look at the correlations-

Mark Zandi:                      Should we take it at face value? Should we take the yield curve at face value?

Cris deRitis:                       I think that would be a good starting point. If something has been a good predictor in the past, the onus is on suggesting why it's different this time or what are the conditions that make it unreliable this go around?

Mark Zandi:                      That's fair. I'm going to do that in just a second.

Marisa DiNatale:              Well, Mark can do that.

Mark Zandi:                      I can do that. I can do that. By the way, you're right, I think this is one reason why the majority of economists that forecast for a living think recession, because they are looking at so-called leading indicators, and the inversion of the yield curve is a very prescient leading indicator. It has done a good job. That really influences, I think, the way people, economists, in general, are thinking about this.

                                             One point to make is historically, the lead time between an inversion of the curve and a recession is a long time, it's 12 to 18 months. I guess the 10-year, two year curve inverted probably six months ago or so, so that would suggest-

Cris deRitis:                       Independence Day.

Marisa DiNatale:              July.

Mark Zandi:                      Oh, was it July 4th?

Marisa DiNatale:              It was July.

Mark Zandi:                      Oh, it was July 4th?

Cris deRitis:                       Right around there, early July.

Mark Zandi:                      That's a good way to remember it. That would suggest, again, face value, literal interpretation, recession second half of this year, maybe going into 2024 or something like that. Dante or Marisa, before I jump in here, is there anything you wanted to say about onus?

Marisa DiNatale:              I know what you're going to say and I... You go for it.

Mark Zandi:                      Well, I'd say a couple things. First thing I'd say is the other financial market leading indicators don't seem to be signaling recession. The yield curve is incongruous with the equity market. The stock market fell in the first half of 2021, but since then, it's been basically going sideways. That decline in the beginning of 2021 was all around interest rates, that's when the Fed started to jack up rates and PE multiples, the multiples stock prices over corporate earnings fell. That correction had nothing to do with any expectation about corporate earnings or the economy, because if you look at analyst expectations forecasts, they are still positive, they're not negative, which you-

Cris deRitis:                       Yeah, but... Oh, sorry, go ahead.

Mark Zandi:                      Just let me finish on that one and then the second one, and then you can push back, because then we'll move on, but the second is the corporate credit spreads. If investors were nervous about recession, they would be saying, "We're going to see corporate defaults or we're going to see corporations not being able to pay back on their debt in a timely way, and I demand a higher interest rate on that debt to compensate for that risk." You would see yields on corporate debt rise relative to risk-free rates like the 10-year yield, and that definitely has not happened. That difference in yield is unchanged. What do I have wrong there, Cris, in terms of the way I framed that?

Cris deRitis:                       Well, equity, Mark, there's the old quip, what is it? The stock market has predicted 11 of the last six recessions. It's not a very reliable indicator versus the yield curve itself, which has been, has a much better track record in terms of recessions. You do make a good point about the spreads, but one way to interpret that would be there may still be a high probability of recession, but the severity of that recession seems to be almost universal. The consensus is that a recession, if it does happen, is unlikely to be very severe or prolonged. From that standpoint, the chances of corporate defaults may be still pretty muted.

Mark Zandi:                      The second thing I'd say is the yield curve isn't foolproof, it did invert prior to the pandemic recession. Are we saying that it predicted the pandemic?

Cris deRitis:                       Who knows?

Marisa DiNatale:              You were calling for a recession in 2020, Mark.

Cris deRitis:                       I don't think we're saying that it predicted the pandemic, it did predict the conditions for recession.

Mark Zandi:                      No, I wasn't, that was in jest. My forecast never had a recession in it. The economy was weak because we had the trade wars and parts of the economy were contracting, like agriculture and manufacturing was contracting. You're saying we don't know the counterfactual, therefore it could have been right, we could have gone into recession in 2020 without the pandemic?

Cris deRitis:                       Correct.

Mark Zandi:                      That feels like a stretch, but okay.

Cris deRitis:                       But it's a signal, it's not causality. Just because it inverts, it isn't causing the recession, it's just a signal.

Mark Zandi:                      Then now, let's turn to the yield curve. It feels like we're only going to get to one question, because that was a doozy. But I turn to the yield curve as a predictor and you're saying, "Why would this time be different? Why wouldn't the yield curve be a good predictor this go around compared to times past?" I'd throw a couple things out there, first, quantitative easing. The Fed has bought a lot of Treasury securities and mortgage-backed securities, taken those out of the bond market, they've taken so-called duration out of the bond market. The estimates suggest that is having some meaningful impact on long-term interest rates relative to short-term interest rates. A lot of debate as to how much, but that could be 20, 30, 40, 50 basis points, that's 0.5 percentage points. The yield curve was flatter coming into this period than it typically would be because of the QE, the buying that the Fed did.

                                             Now, it's QTing now, meaning it's allowing those securities to wind down, but what really matters for interest rates, long-term rates, is the stock of debt, Treasury bond securities, that the Fed holds on its balance sheet and that's still very, very large. I think it's, I'm making this up, but 8, 8.5 trillion 4, 4.5 trillion above what it was pre-pandemic. That's one thing to consider that this time feels different. Second is forward guidance. The Federal Reserve is being extraordinarily clear about what it's going to do with policy here. It's trying to tell markets exactly where it's going. Markets know that inflation's going to come in here one way or the other and that if you look out into the future, the funds rate's going to be settling around 2.5%. That's what people think, we think, the Fed thinks is the long-run equilibrium yield.

                                             The amount of guidance the Federal Reserve is giving to bond investors today is extraordinary, it's just more than it ever has in the past. Over the years, it's gotten more transparent and provided more guidance, but what it's been doing in the current period is just very different than what it's done in the past. Let me give you a third reason why this time might be different. This does go to potentially, and there is some causality here, the yield curve is not only a predictor of recession, but it influences the environment and creates the fodder for recession. Because if you look historically in the good times before recession, and they're generally boom times when the economy is operating at a very high level, people are taking a lot of risk, borrowing a lot of money, credit's really flowing, the financial system's providing a lot of credit, and then when the curve inverts, when the Fed steps on the breaks and the curve inverts, those financial institutions, those banks can't make money anymore because they borrow short-term funding rates and they lend at longer, higher rates.

                                             When the curve inverts, that spread turns negative and their net interest margin vanishes, they can't make money, so they stop lending, they tighten down on credit very aggressively, which matters a lot in most times historically, because those folks that borrowed a lot of money in the boom times need the money in the tough times, they've got to refinance. Particularly businesses, they say, "Hey, I've got to refinance this debt, I can't pay you back," but they can't afford the terms and the interest rates when you get into a world of an inverted curve and an interest margin.

                                             That matters a lot less in the current period, because we never saw that rapid credit growth that you typically do coming into a recession, probably because we're on the flip side of the financial crisis and lived through a credit crunch, and so we didn't see a lot of credit growth. If you look at the amount of debt that businesses need to refinance in 2023 going into '24, it's actually quite low, either bonds or loans. That causal link between the shape of the yield curve and the recession is not evident today, it's certainly not to the degree that it has been historically. Those are three reasons why this time may be different, that the yield curve isn't as prescient as it has been historically. I'll stop there, I said a lot. I'm curious as to your reaction in what I said just there. That's for you, Cris.

Cris deRitis:                       That's for me? All reasonable. I'd even throw out another one-

Mark Zandi:                      Oh, okay. Go ahead.

Cris deRitis:                       ... to support that view, which would be why is it so US-centric? If you go to other countries around the globe, you don't see the yield curve having this predictive ability, so that would suggest that it isn't causal and it very well could be these other reasons. I guess what gives me pause is just the extent of the inversion now. You're right, these other factors could make a difference, could lessen the importance, but it's just so deeply inverted, I have a hard time giving it up.

Mark Zandi:                      Can I just throw out one other thing? This is a work in progress in my mind. As I said earlier, this is plaguing me, I'm plagued by this question. If you look at the inversion of the curve since Independence Day of last year, most of that decline has been related to, particularly when you look at a 10- versus 2-year, most of that inversion is related to a decline in the 10-year yield, not an increase in the 2-year yield, that had already happened. If you look at the 10-year yield and decompose that decline into what is behind that, it's one of three things, it's lower inflation expectations, a shift in expectations around the real after inflation federal funds rate target, and the third is the term premium.

                                             I'm not going to go into any of this, because we need a whole podcast. Maybe I should get Campbell Harvey on, he's the professor from, I think he's at Duke, who popularized this measure as a leading indicator, so we can have this conversation in-depth. But most of the decline in the 10-year, and therefore most of the decline and most of the inversion of the curve that we observe is a move in the term premium from positive to negative, and that's not consistent, I don't think, with recession. If it were a recession, that would be around inflation expectations or mostly around the real federal fund rate target and I don't think that's the case. But just throwing it out there fodder for future research, it's a work in progress. I'm not sure exactly why that's the case.

Cris deRitis:                       All of those things are difficult to measure, even having inflation expectations, it's all fuzzy.

Mark Zandi:                      You're right, there's measurement issues. Decomposing is not easy, that's what you're saying. That's very true. That was a great question, Marisa. I think maybe we'll hold the other ones for the next round, because that was a really good one. I do want to quickly end because this is getting, as I am wont to say, long in the tooth, this podcast. Very quickly, just because we've been ending this way, these podcasts now, for the better part of the last six months or so, what's the probability of recession in the next 12, 18 months, let's say, going into 2024? NBER defined, National Bureau of Economic Research, broad-based persistent decline activity, that's the definition of recession. What's the probability and how has that changed over the last week, two weeks, whatever? Dante, what's your probability of recession?

Dante DeAntonio:           I think last time I was on, I was at 50% and I'm going to stay at... Before this morning, I was considering edging a bit lower than that, but I think I'm going to stay at 50, given the report today.

Mark Zandi:                      Cautious man. That makes sense. Marisa?

Marisa DiNatale:              Wait, so the strong jobs report makes you think that there's a greater probability of recession because you think the Fed will just keep hammering and overdo it?

Dante DeAntonio:           Yeah, I think it just raises the risk that the Fed might go too far. Maybe things will reverse next month and there won't be an issue at all.

Marisa DiNatale:              I'm still at 50%.

Mark Zandi:                      50. That hasn't changed that much. Well, you came down earlier, but that hasn't changed for a while.

Marisa DiNatale:              I would say it's more of a soft 50 than a hard 50. I'm thinking the risks are more to the downside than to the upside on the probability of recession now.

Mark Zandi:                      What does that mean, lower probability or higher probability?

Marisa DiNatale:              My bias would be that it's under 50.

Mark Zandi:                      It's 45 to 50, something like that.

Marisa DiNatale:              Yeah.

Mark Zandi:                      Good. You're 50 with an arrow pointed down?

Marisa DiNatale:              Mm-hmm.

Mark Zandi:                      Got it. Cris, what's your probability?

Cris deRitis:                       The groundhog saw it's shadow, so-

Mark Zandi:                      It did? I didn't know that.

Cris deRitis:                       Oh.

Marisa DiNatale:              That's a data point that came out this week.

Mark Zandi:                      That's a good leading indicator, I think. It's kind of like the yield curve, I'd say.

Cris deRitis:                       Apparently, the groundhog's been right 69% of the time.

Mark Zandi:                      There you go.

Cris deRitis:                       Better than most economists. 60%, I'm going to nudge my probability down to 60% from two-thirds.

Mark Zandi:                      Hold on, wait.

Cris deRitis:                       Big move.

Mark Zandi:                      66 down to 60?

Cris deRitis:                       Yes.

Mark Zandi:                      Ooh, interesting.

Cris deRitis:                       But it's really around the timing.

Mark Zandi:                      What does that mean?

Cris deRitis:                       I still am somewhat pessimistic, but-

Mark Zandi:                      In 2100, is that what you're saying?

Cris deRitis:                       No, no. It might be into 2024.

Mark Zandi:                      Fair enough. That's not unreasonable. Maybe we should start saying recession probability in 2023 and recession probability in 2024, because that may be an important distinction, so we'll start doing that in the future. But you're saying over the next 12, 18 months, you're down to 60%?

Cris deRitis:                       Right.

Mark Zandi:                      Got it. I'm still at 50 with a down arrow, I'm with Marisa. I will point out that we run this same question at our macro meeting with all our economists and the group is moving in the same direction this group is. If you go back, I think it was in November, that was at the height of the angst around recession, I think the group was close to 70% probability of recession. It's now down to below 50, it's at 47%, so the group has gotten more optimistic here. Great. Well, that was very good. Well, anything else? I know I was hard on everyone, I apologize for that. I was in a little bad mood because of the numbers. I had written all these tweets under the assumption that this would be down the fairway report and then I had to rewrite all my tweets, so I was in a pretty bad mood coming into this, so I apologize if I was too hard on anybody. No, you're all good with that? Good. Well, with that, we're going to call this a podcast. Dear listener, we'll talk to you next week. Take care now.