Moody's Talks - Inside Economics - Tuesday Twist

Episode Date: December 16, 2025

Dante joins Mark and Cris for an unusual jobs Tuesday podcast to break down the November employment report. Due to the prolonged government shutdown, the report delivered two months of payroll data, which continues to signal that the labor market is grinding to a halt. With the unemployment rate on the rise, the team discusses their updated recession probabilities, and Marisa’s absence is felt as they struggle through the stats game.  Guest: Dante DeAntonio, Senior Director of Economic Research, Moody's AnalyticsHosts: Mark Zandi – Chief Economist, Moody’s Analytics, Cris deRitis – Deputy Chief Economist, Moody’s Analytics, and Marisa DiNatale – Senior Director - Head of Global Forecasting, Moody’s AnalyticsFollow Mark Zandi on 'X' @MarkZandi, Cris deRitis on LinkedIn, and Marisa DiNatale on LinkedIn Questions or Comments, please email us at [e-mail]. We would love to hear from you.  To stay informed and follow the insights of Moody's Analytics economists, visit Economic View. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Welcome to Inside Economics. I'm Mark Sandy, the chief economist of Moody's Analytics, and I'm joined by a couple of my colleagues, my trusty co-host, Chris DeRides. Hey, Chris. Hi, Mark. How are things? Things are good. We're missing, Marissa.

absolutely so this episode will have an asterisk next to it if we play the stats game yes indeed uh she's quite good at that stats game so she is she is we've got someone filling in dante dante d'antonio dr d'antoneo i should say hello dante and you're joining us because it's jobs tuesday it's an unusual jobs tuesday yeah right uh this is tuesday December 16th, and we got a bunch of employment data today, and we're going to talk about that. I also got retail sales. We'll talk a little bit about that.

We'll play the game. We'll answer a few listener questions. It'll be a relatively short podcast, I think. Although every time I say that, it never is, but it just feels like it might be this go-around. And I'm at HQ again, as you can see. My artwork behind me, that is definitely HQ artwork. Did you notice that, Dante?

I wasn't sure where you were. I knew you weren't at home, but I couldn't figure out where you were. Oh, is that right? Am I supposed to know by now what that background is? I think so. I've been on the 52nd floor of Seven World Trade a lot recently, so that's where I am. And I'm camping out in the same conference room, and that's my favorite artwork on the wall there. What do you think?

I like it. It's nice. Yeah. I like your map better, though. I was insane. At least I can look at my map and learn something. Exactly. Exactly. Exactly. Well, maybe we should just dive right in, huh? Let's go. Yeah, let's go. Tell us about these job numbers.

Sure. So it's technically the November employment report, but we got two months of data on the payroll side of things. We got data for October and November. On the household survey side of things, we just have November, right? We skip October altogether, so we'll have a permanent hole there in October. On the payroll side is a little bit of a mixed bag, right? We had a big decline. in the first print of the October data, so down 105,000 jobs almost entirely due to decline in the federal government, right? Federal government declined by about 150,000 in October. That's finally the rolling off of federal employees who took the deferred resignation offer earlier this year, so it's not a new story, but it's finally showing up in the data.

In November, we got a little bit of an increase in payrolls, up 64,000. The story is much the same in that most of that game is concentrated in health care, if you're looking at an industry basis, some minor gains elsewhere, but healthcare added 64,000 jobs. The headline gain in November was 64,000 jobs. So the net of everything else is zero in November, which has been a pretty consistent team here over the last three or six months or so. Before I dive into anything else, usually stop me with a question by this point. Yeah, I know. I've been unusually quiet, but let me ask. Do you think,

just a general methodological question. Do you think the effects of the government shutdown have had an impact on the – I know they're late, and we're not going to get October data for the household survey. But in terms of the surveys and the quality of the data that we're getting – is there more noise in the data than typical? Can you tell? So on the payroll side of things, I think the answer is no, right? If anything, they reported that response rates were actually higher, you know, for the first print of October and November.

than usual because they lengthened the response period, because the report was delayed. So if anything on the payroll side, you should have sort of more accurate first estimates for those two months because we have more sample in than we usually do when we get this report. On the household survey side of things, they did note they had to make some methodological tweaks because they don't have the October data. Typically, they use the previous month data and some of the re-waiting that they do, and so they had to make some tweaks to the methodology, and they noted that the standard errors on the November data were a little bit larger than normal because of that. I don't think it's anything

meaningful in the sense of telling us what direction things are headed, but yeah, I think that should all roll off here in the next couple months as we get sort of back to normal data processing. Yeah, there was a couple weird data points, weren't there? I mean, I know you, particularly in the household survey, yeah, you know, part-time for economic reasons as an example. Yeah, I think there's some questions there around, you know, is there some noise from the actual shutdown itself, right? So one of the survey period still captures the period of the shutdown. And so workers that were impacted by the shutdown either directly or indirectly,

you know, maybe they were working part-time because they were working at a contractor of the federal government and, you know, they were shuttered for part of that period. So I think it's possible that the shutdown itself impacted, you know, sort of worker behavior in some of the data. And so that could affect some of those part-time readings that we saw. I think obviously we'll get a better sense of if that's all it was as we get December and January data moving forward, but I think it could be part of the explanation. Yeah, I know some of the unemployment rates for some of the demographics bounced around a lot.

Like for unemployment for people 20 to 24, that fell a full percentage point. That just feels like noise to me. It doesn't feel real. Yeah, and a lot of those tend to bounce quite a bit month to month. And now you're looking, you know, over a two-month period. And so some of the movements seem kind of large, but maybe they wouldn't have been, you were sort of put out as much if you would seem half that movement in October and another half in November. But yeah, I agree. It seems like there were some larger movements if you get into some of the demographic cuts of unemployment rates that it would be curious to see how they'll shake out here moving forward.

But bottom line, you feel pretty good about the data. I mean, we're going to start talking about what it all means. And you're saying we're getting reasonable information here. Yeah, I feel fine about the quality of the data. I don't feel great about what it signals about the labor market. Yeah, but we'll come back to that. Chris, any questions on the methodology that you can strike you about that? You come to the same place that this is, you know, feels like the data is a good enough quality

that we can make inferences around what's going on with it. I think it's okay. There's still a bit of an asterisk next to it, right? So we want to, I mean, you highlighted some of the kind of outlier, so I don't think we want to too much into it, but I think directionally, it seems reasonable. Right. Right. Okay.

And so, Dante, you talked about the payroll survey. I guess, and we'll go talk about the household survey in just a second. But on the payroll survey, I think the bottom line, at least from my perspective, is that once you abstract from the kind of vagaries of the data, there's been literally no job growth. since liberation, in the wake of Liberty, Liberation Day back in April. You could begin in May, June, July, August, September, October, November. One month is up, one month it's down, but the net is we're not going anywhere. And that's before all the revisions in, correct?

Yeah, that's my feeling as well. I mean, job growth seems like it's awfully close to zero. And yeah, like you said, if anything, later revisions will likely push that further down. And it feels like if we're at zero now with revision, because Chair Powell was talking talking about the Fed research on this issue, and they're calculating that the revisions, the upcoming benchmark revisions, which will get in early February with the January report, overstate the monthly job growth by 60Ks over this period. So that would suggest we are firmly, if that's accurate, and we'll, you know, we'll see. But if that's accurate,

it feels like we're firmly in the red here. No? Yeah, well, and that revision only goes through March, right? So that's really from, you know, April of 24 through March of 25. You talk about shaving, you know, 60,000 jobs a month off of that. But that job growth was much more firmly positive in that period, right? So you're still talking about much weaker job growth. It's not quite as clear what that means for the revisions to job growth for most of this year yet, right? I thought he was, Chris, correct me if I'm wrong, though. Wasn't he commenting on job growth this year? Are you saying, did I get that wrong? I thought he was suggesting that the job numbers were looking at now for 2025 could be overstated by as much of 60K per month. And that would come out

with the revisions. I think that's what he suggested in his words. In terms of the precision, yeah, okay. They might have a point that maybe he's generalizing. Yeah. It's overstating the case. There's a case here. The revisions are going to be down, but maybe not 60K. And also, who knows, really, because. Right. Right. Okay. Yeah, I think it's pretty clear the directionally they're down. And I think, you know, maybe we can quibble about the magnitude. But to your point, we're already basically at zero. And so any down revision at this point is not a good thing. Right.

And I know we've talked about this in the past, but I'll just say it again. Doesn't it feel like this all goes back to Liberation Day and the tariffs? I mean, the break and the data coincides exactly with – Liberation Day was April 4th, 2025. by the May survey period, mid-May, that's when we, since then we've gone flat. It feels like you can connect the dots. It certainly feels like the timing lines up, right? I mean, and if anything, job growth was even weaker in sort of the three-month period immediately after that, right?

And if anything, prior to payroll growth has maybe creeped back up a little, you know, it's not strong, but, you know, maybe it's a little stronger in the last three months than it was in the first three, which I think would also fit, right? We had the sort of the most uncertainty in May, June, July, and then maybe things have gotten a little more stable here over the last three months. So I think that fits that narrative. Yeah, Chris, right, would you agree or disagree? When I calculated the six-month change was exactly $100,000, right?

Growth rate. A growth change in payrolls. What about it? I don't know. From six months ago, was up $100,000. Oh, over that six-month period, if you net everything out, it's up $100,000. K in six correct right so that's 16k a month okay well within the top right that's within rounding of zero yeah yeah especially with was we're going to get some revision it does feel like

it's going to be lower so right whatever it is it feels like we're zero okay i mean i think we can stipulate it flat i stipulate two things one no job growth and two it's because of the tariffs No? I mean, at least at first. Yeah, I'm going to stipulate it. Yeah. Anybody disagree? The uncertainty caused by the tariffs?

Oh, okay. There's a lot of channels. That's why I heat and hawed for a second. But yeah, there's a lot of channels. Right. Yeah, I think that's reasonable. Heed and haught. Is that a phrase?

That feels right. It sounded right, but that doesn't mean. Yeah. Man, that came out of nowhere. Where did that come in the deep recesses of my mind? Heed and haud. Is it hemmed and hawed or he? Oh, hemdenhock.

Oh, that's what it is. It's hemdenhod, not he. That's so funny. I thought I had it. I thought I nailed it. Yeah. Hemdenhod. By the way, thinking about that for a second, that even sounds weird.

Hemdenhod. Where is that coming from? Someone's got to do some etymological, that's the right word, isn't it, etymological study? It's hemdenhod. I just checked. It's hemdenhod. to hesitate or be indecisive, hemmed and hawed.

Okay, then I got it right. I got it right. Somehow it all worked out. What about he and Hodd? He didn't odd. He didn't odd. He's just made up far away out. That's funny.

Okay. Anything else on the payroll survey, surveyed businesses? Yeah, one of the thing that stood out, and it could just be noise, right? Average hourly earnings growth was only a tenth of a percent in November, and that brought year-over-year wage growth down to. three and a half percent, which is the lowest that it's been since all the noise around the pandemic. And so I think there's maybe a little bit of concern there. Obviously, some of the story here has been job growth is slow, but productivity growth may be picking up and

sort of filling the gap. But having wage growth weakening when productivity growth is supposed to be accelerating doesn't seem to fit that story very well. So, you know, again, could just be a little bit of noise in the data, but I'm curious to see where that goes. Yeah, I think if I got the data right, year over year, we're down to three and a half percent on average, generally earnings. CPI is three, so we still get real wage growth, but it's pretty de minimis at this point. And that's, that's for the typical worker. So some workers are now experiencing outright real wage declines.

If you buy into the data, if you buy into the data. Yeah, and certainly if you believe that inflation's headed higher here, at least in the near term, and if you get wage growth, it keeps slowing, then I think that because, becomes more problematic over the next six months or so. Right. You know, the other thing I noticed was going back to the composition of the job creation, you mentioned health care. Construction is also adding, and I'm not sure I believe that.

I mean, even residential construction, I think, added to perils. Now, this is seasonal adjustment becomes pretty difficult in the month of November, so October or November. So that might be what's going on. But it just feels like the data is even weaker than this would suggest. It just doesn't feel right that we get job growth in construction. I know we're building data centers, but, you know, I don't think that's enough to offset the loss of workers on single-family homes. And we know that's weakening.

But anyway, Chris, anything else on the payroll survey? No. No? Okay. All right. Okay. The household survey, this is a little more squirly. This is where the asterisk says, no October because it couldn't conduct the survey.

But what is the November data say? Well, I was going to say something that would color your perspective, so I won't say it. What did it say? It's already colored, but I think it reads as pretty negative, right? I mean, the headlines, obviously, the unemployment rate is up to 4.6%. It was 4.4 in September, so you assume sort of a straight line here to get to 4.6 in November. That's the highest that it's been in quite a long time, I think, since late 2021, if my memory serves me correctly. We're getting a little bit of labor force growth, right?

So if you look over the two-month window from September to November, labor force was up by about 300K, which again is sort of bucking the pattern that we saw in the first half of the year where there was basically no growth at all in the labor force. And now over the last three, four, five months, we've seen some of that growth pick back up. And certainly I think that's contributing to the unemployment rate creeping a little bit higher. It's not the only contributor, right?

The actual number of unemployed workers was up as well between September and November. So some of it is just outright increases in unemployment. But I do think the return of some of that labor force growth is also putting some upward pressure on the unemployment rate as well. Like you alluded to before, I think there's a little more volatility than normal here without being able to see October. It's hard to get a sense for some of these. I'm not putting a whole lot of stock in the sort of demographic cuts of the data here. The part-time for economic reasons you mentioned had jumped again. I think some of that could be just shut down related and sort of

workers that were affected by the shutdown in some way. So I'd hesitate to put any real stock in that. So I think the big story here is that the unemployment rate is still definitively moving higher. Participation is still, you know, basically stuck where it's been, you know, we're getting a little bit of labor force growth, which is, you know, better than what we had early in the year, but it's not a lot to write home about. Yeah, what do you think underlying labor force growth is, you know, because it's obviously very weak start of the year. Now it's a little stronger. Net, net, net, net. What do you think it is? It's a hard question. You know, you look the pattern of labor force growth and it's messy, right? I mean, you had some growth early in the

year and then you had declines, you know, sort of in the middle of the year and now you've got what looks like strong growth. I find it hard to believe that underlying growth is more than like 50K a month, but it just, it feels like it's hard to tell given all the ups downs and all around. Normally labor force growth is a little more steady. It sort of wobbles a little bit months to month, but here it's been on a much bigger roller coaster, it feels like this year. Right, right. I mean, What was I going to say? Oh, the other complicating factors, the population controls, right?

I mean, you get a, every January, they benchmarked a new population, and they don't go back and revise the historical data, so you get this discontinuity. So there's a big discontinuity as of January, so you really can't only look at the data since January. It feels like it's 50 to 75K break-even. That kind of is like a half a point on labor force growth. That feels like what it's saying to say, sank in the data. So 50 to 75K break-even, that's the rate of monthly job growth necessary to maintain stable

unemployment. So if we're at zero, that's our, we concluded we're creating no jobs, zero that we stipulated, and 50 to 75K would be consistent with the rising unemployment rate, pretty much what we've observed. So the unemployment rate was 4% in January. It's now 4-6. You have to round up. You get up to 4-6. But that's consistent with those kind of numbers.

Chris, do you agree with that arithmetic? Yes. Yeah. Yeah. Again, if you go back to May, if you look at the average since May, it's around 73,000 per month. Okay. Okay, 50 to 75K.

Kind of right in there. Yeah. Okay. Anything else in the household survey that you would call out? Not specifically from the household survey, but, you know, it's probably time. We can start talking about the SOM rule again if you want to go ahead. I was going to ask about that. Yeah.

We didn't quite trigger it this month, but we're sort of on the precipice. If you use unrounded, if you use the rounded unemployment rates, you know, if it stays at 4-6 next month, we'll trigger the SOM rule again. It will? It will. Yeah, we'll be exactly at the threshold, just like we were back in the middle 2020. Okay, tell us the SOM rule and explain it. Sure, you take the three-month moving average of the unemployment rates. And you look at the change relative to the lowest point in the last year, right?

So if you get another 4-6 unemployment rate in December, the current three-month-moving average is just under 4-6. And then the lowest three-month moving average that you have is just under 4-1 back towards the beginning of the year. And so you get that half a percentage point gap or increase in the unemployment rate within the year if it holds at 4-6 again. Okay. Oh, that's interesting. I didn't realize that. Wow. And historically, when that you trigger that some rule, we're already in recession typically. It's not going into, you're already in, a couple, three months in, right? Yeah, I mean, and obviously we violated that rule last year, right? Did we violate it? Did we actually violate it last year? It did trigger in July and August. We got exactly to the, you know, 50 basis point threshold.

So we didn't go over 50 basis points. We got exactly to 50 basis. Oh, okay. And then the unemployment rate obviously came back down a little bit. You know, it had gotten up to 4-2 at that point, which was up, you know, from three and a half within the last year and then came back down some. Right. Now we're on a similar trajectory here in the back half of this year.

Right. Okay. Chris, anything else on you want to bring up? Again, kind of need to. caveat this with the data here, but the African-American unemployment rate rose to, if you take it at face value, it rose to 8.3%, which is a pretty sizable jump again from October. It increased. Increased. It was 7.5 to 8.3.

That, too, has typically been associated with recessions as kind of one of the first cohorts that seized deterioration in unemployment. Right. Well, it's interesting because I mentioned the unemployment rate for 20, 24 years old. That declined to 8.3%. And you're saying the black unemployment rate rose to 8.3%. They would jump, but in different direction. I think we need to wait. I think so. I don't want to. Yeah. Yeah. The other one would be multiple job holders also. So pretty significant increase. Again, if that holds, that's kind of a negative sign that folks are having to take additional jobs. And that might also be government shutdown related or something. Right, exactly. Yeah, exactly. We don't want to read too much into it. Right. Right.

Okay. All right. So let, let, just gauge your level of concern with the data. Tell me, we haven't done this a while, what is your probability of recession starting at some point in the next year, let's say through the end of 2020. I think that is a good way to really quantify your level of angst. have any ax? Chris, what's your probability recession? I've got some angst. I'd say 40%. Really? Yeah. Wow. Forty percent. These data are not looking, the trends are not our friends here,

right? Interesting. Because you were much lower than that. Meaningfully, I thought, right? Was that lower than that last time? It's been a while since we asked. It's been a while since we asked. Yeah. All right, 40%. Okay. Dante, what's yours? I'm at 40%. I think that's probably roughly where I've been recently. You know, I don't think it's changed the whole lot in the last six months or so. Right. Well, I'm at 40 to 45%.

I'm right with you. Maybe I'd said the 45% just to make a little bit of difference between the two of the two of the best. It seems like we're all on the same page. 40% seems high. I mean, that's pretty high. I think it's higher than consensus. The consensus is probably closer to this.

30 percent, 25, 50 percent. Although that might change after today's data. I don't know. I'm not sure. I think the market reaction, the stock market reaction was somewhat negative. But again, I'm not sure how much to read into that. And bond yields were down a little bit. All suggestive that investors are thinking this is a pretty weak report.

You know, it shows the economy is pretty weak. Yeah, and here's the thing. I know I've said this before, but this goes to the recession. Oh, by the way, I should mention Shandor, our colleague, Shandor Witcher, who developed a leading indicator of a recession based on a random force algorithm, machine learning algorithm, was unable to run his model for quite some time because of the government shutdown the lack of data. Well, he ran it today. You know what it said? Did you guys see the email? No.

Jandor? 41.5%. Well, there you go. Right there, right? Very consistent. Yeah. Yeah. Here's the thing. that makes me nervous, it adds to my level of angst. And I think I said this before, but let me say it again and get a reaction. So one reason why job growth is weaker is less labor supply, right? I mean, because of the immigration policy and labor supply. And that gets you to the 50 to 75K break-even monthly job number. That by itself, if nothing else was going on, is already pretty pretty weak, right? But that goes to lack of bodies, lack of people able to work. And now we're at

zero. We're saying that's kind of where we think we are, the actual rate of job growth, underlying job growth. And that goes to demand. So we're at 50 to 75,000, 5,000 acres of supply, and we get down to zero because of a lack of labor demand. And that feels like that, again, we can trace it back to the tariffs. We can trace it back to some of the other de-globalization efforts that the administration's engaged in, including immigration policy, right? Because immigrants, they're consumers. And if they're not here, they're not consuming, and that's a matter of demand. But the other factor is AI. And there, the evidence would suggest that, yeah, maybe it's having some impact on demand for labor. But at least so far, it's very, very modest. You know, it's on the margin. You know, young people

maybe, the hiring rate might be a little lower because of weak because of the AI. But what happens when AI kicks, productivity gains start to kick into a higher gear, which feels like we're, at least that's the betting in the stock market. Stock investors are buying AI stocks thinking these, they were going to see big adoption rates by business. This is going to raise productivity growth. This is going to raise profitability if I'm going to buy these stocks, if they're half right or even a quarter rate and we get more productivity gains because of AI, then we're in a world of outright job decline, all else being equal. Is that right? Is that narrative? Am I missing something, Dante there in that narrative? I don't think so. I mean, that's how I've been thinking about it as

well, yeah. Okay, Chris, same, same thing. Am I missing anything? Is that at least in the early stage, right? Yeah, yeah. Yeah, yeah, the immediate impact, yeah. The immediate, Debt ahead, next three, six months, that kind of thing. Yeah, okay. And if you go, if you have actual outright job loss because of, you know, supply but also demand, and you have weak consumer spending or soft consumer spending, I think we need to go next to the retail sales numbers because they got released today by census for the month of, I believe for the month of October, you know, you see what they say.

But if you look at real consumer spending total. The whole shoot match, it includes retail sales spending on services. Since the beginning of the year, that's up 1.3% at an annualized rate. That's pretty weak. I'd characterize that as weak. You know, that's below the 2% kind of rule of thumb for a kind of typical growth and consumer spending in a typical well-functioning economy. That, all that feels like a fodder for a real problem, right?

Job loss in the context of already, without accounting for AI to a significant, degree in the context of already weak, relatively weak consumer spending. That just feels like the fodder for, you know, potential recession. And thus the 40 to 45% risk of recession. Is that, is that storyline hold, Chris, for you? Anything you'd add to that? It does. So even if those productivity gains lead to higher stock values for those, for a few companies, for specific companies, that's, that's. submit that, is it sufficient to have that upper 10% of households, right, drive additional spending through a wealth effect? And I think we're getting to the point where that's not

the case. I think that you kind of get some diminishing returns there. And if you have the bottom falling out or the middle of the distribution impacted, right, I see that as having a negative impact on it in terms of spending and then overall growth as well. Yeah, I guess that's an interesting point. I mean, my interpretation of what you're saying is also, I'm taking as given that the benefits of the implied productivity gains are narrowly distributed. They're going to a small group of high net worth households, that it's not being distributed broadly. Because if it were, that would mean higher wages. So people, you know, would still be able to spend, even if we're not creating jobs, even for losing jobs potentially, if their wages are rising more quickly. But that's

If they had savings, they could, or assets that were appreciating. Or the wealth, yeah. But we're not observing that. We're not seeing that at all. Just the opposite. We were talking about wage growth earlier. It's decelerating, not accelerating. Yeah.

Right. Dante, anything to add to that? No, I think that's, yeah, I mean, that, yeah. It leaves early. I think that's the big thing you'd want to see. If AI is going to have this big impact, you don't want the only sort of benefit to come through wealth effects and through the stock market, right? You want it to come through workers and higher wages, right?

That's the thing that would help keep, consumer spending and drive consumer spending higher moving forward. I guess the thing, and maybe this is the fodder for another podcast, I guess what could tip the balance towards no recession, you know, continue kind of weak growth in 2026, is all the policy support that's in train or, you know, debt ahead. And you've got the Fed who's lowering interest rates could lower them a lot more. We have three rate cuts next year, but it could be more. than that. There could be quantitative easing. We've got the one big beautiful bill act, you know,

with all those tax cuts for businesses and households kicking in in 2026. And the president is obviously putting forward other bennies for taxpayers talking about a $2,000 stimulus check or aid to farmers or a, what do you call it, the Trump savings account or something, you know, something along those lines. So there's a bunch of stuff that, oh, and there's also the deregulation, the banking system is being deregulated, or the regulation on the system is being scaled back pretty quickly here, it feels like, and that could lead to more lending and more credit growth. So the policy support that's coming, you know, it's all temporary, it all feels temporary, and there's side effects to all that, higher budget deficits in debt,

maybe the lending becomes a problem down the road in terms of credit conditions. But for the immediate future for the next 12 months, which is where our recession probabilities are focused, you could get more juice from economic policy and that kind of saves the day. Does that sound right, anybody? Chris? It does. That's the reason why my probability is only 40%. 40%. Right? Without those possibilities or those supports, that'd be probably closer to 50. Really interesting. You too, Dante? Yeah, I think you'd have to assume a higher probability recession if you didn't have that coming in 2026, yeah.

But you wouldn't say over 50%. I don't think I'd go over 50%. No, but I'd be much close for 20. Much close for 50. Actually, I think I'd go so far as to say we'd be over 50% without that policy. I mean, it just feels that way. But anyway. Okay, let's, anything else on the data?

Anything else to point out? before we move on to the game and the listener questions? No? Okay. All right, let's play the game. We each put forward a stat. The rest of the group tries to figure that out through clues, questions, deducted to reasoning. The best stat is one that's not so easy.

We don't get it immediately. One that's not so hard we never get it. And if it's apropos of the topic at hand, obviously jobs, so it can be broader than that, all the better. Dante, you want to go first? I can. Let's see. Let's go with 238,000. Well, it feels like that's in the payroll job numbers, no?

It is not payroll job numbers. Household job numbers. It's in this household number. 238K. Geez, is it in the labor force data? It's in labor force data, yeah. 238K. Is it a month-to-month change?

It's an average change over a period. of time. Oh. Okay. 238K. Related to the labor force. So it's some demographic group within the labor force? No.

No. I don't know. Chris, do you know? 238K in the household survey? It's not adjusted to payroll concepts. Oh. No, that's not that. That's good.

I didn't check that this morning. Oh, yeah. I'm interested in that. in that. Oh, the other thing I'm interested, oh, well, I might take someone's stat, so I won't say it. But 238K. I don't know, Dante, what is it? So it's average labor force growth over the last five months, right?

Just over the last five months since June, we've been adding 238,000 people to the labor force every month, which is a huge. Yeah. And if you go just a couple months different than that, the number changes dramatically, right? So Chris had mentioned earlier the 6,000, if you just go like one month further back. Oh, yeah, because there was a big. There was a huge decline in the middle of the year. So all of this is to say, you know, this, the increase in the unemployment rate that we've seen, right, the solemn rule trigger, right? It directly corresponds to this period where labor force growth is all the sudden, you know, shot out of a cannon, right?

For June to November, the unemployment we went from 4-1 to 4-6, but that's because we've had labor force growth that is, you know, almost 240,000 a month, right? It's kind of the same thing that happened last year when the unemployment rate was rising. It's mostly because of this outsized labor force growth, right? And if you look at the first half of the year, January to June, labor force growth was flat, right? There was basically no change in the labor force. The unemployment rate was four, and it went to four one in June, right? There was no change in the unemployment rate because we had no labor force growth. Now, nothing's really changed on the job side of things, right?

It's just labor force is now rowing again since June, and that's pushed the unemployment rate higher again. At least that's my read of the situation. And so I don't think we can expect that sort of labor force growth to continue, right? So I don't know that we have to be so concerned that the unemployment rate is taking off here and it can't be stopped. I think it's just a question of does labor force growth sort of settle back down again? And if it does, then the unemployment rate likely stabilizes. So that's despite the crackdown on immigration. So that's – well, yeah, that's – I mean, you see this, you know, in April, you had this peak in labor force, and then it declined in May and June.

And I think even in July a little bit, it still declined, right? So that feels like sort of the immigration crackdown. But then since then, it's really taken off again. And whether that's just volatility of the data or that's a reversal of some of that change that happened earlier in the year, it's not entirely clear. But we have these like two distinct periods of labor force growth, which are having, you know, a big effect on the employment rate, I think. So what do you describe? Yeah. Yeah, what do you describe that?

Is that people staying in the labor force longer or not retiring? Or is it, has there been a surge of graduates? I'm not sure. I mean, I think some of it is graduates, probably. I think some of it is probably older workers staying in the labor force longer. I did not look this morning at foreign-born labor force. Obviously, that had been declining pretty sharply in the first half of the year. I don't know if that's at least stabilized or come back at all, which might be providing a little bit of support.

I will weigh in on that, but I'm taking my stat, though. So I'm just saying there were 32 point, I think I have the data right. right, 32.2 million foreign-born in the labor force as of November. That's down from 33.3 million, you know, at the start of the year, January. So that would suggest, you know, weaker labor force, not stronger labor force. I didn't look at it month to month. I didn't look at the pattern. But it looked like I did a chart. It looked generally downward trend. So that is, now some of that may be people don't want to say I'm foreign born. Right. Right. In this current environment, so that I'm sure is playing

a role. But nonetheless, I mean, directionally, it would suggest that the immigration policy is having an impact. Yeah. And I think even in the best of times, right, the sort of the population controls that are used to formulate these estimates throughout the year are tough, right? You have to assume that things are sort of trending in the same way. And obviously, I think it's a safe assumption that things have changed a lot over the last year. And so I think we'll get new population controls with the January data in a couple months. And I'll be curious to see how much that, you know, we obviously have to say this big level shift

this past January. Do we get a similarly size level shifts one way or the other here? So I think I still attribute some of this to noise, but I think just the sharp difference in the first half of the year to the second half of the year makes things look very different, right? It looks like the unemployment rates rising dramatically and it is. But I think it's mostly because of this shift in what's going on in labor force, not because of some big deterioration in job growth.

Job growth has been weak basically all year anyway. Okay, so that mitigates some of the concern, doesn't it, then? I mean, my sense is, given the volatility and the data, the seasonal adjustment issues, all kinds of stuff, probably still best to take the entire period and say this is what, that's where I got to have percentage point growth in labor force, 50 to 75KK break, even on a job growth. I use the entire period. It feels like that's still what we should be doing, but you're saying there's an asterisk there as well.

Yeah, right. If you go back to January and look, right, labor force is up an average of like 80K per month from January to November, but it's, you know, it was flat in the first half of that and then a lot in the second half. So, yeah, I think you can make an argument that you just average it out. And that means, you know, this increase in the unemployment rate shouldn't really have just happened over these last five months. It should really, yeah.

Right, you spread it out over the whole year, and then maybe it doesn't work as bad. Right, okay. Yeah, it feels like the more we talk about the household survey, the more noise there is in that, and the more we have to wait a little bit here to get some more data. Yeah, I agree. Chris, you want to go next on the stat? Sure. It's a twofer.

47.6 and 53. Are those diffusion entities? They are. Which ones? Say them again, 40.6? 47.6. And 57.6. And 53.

Oh, is one, the first one is a one-month diffusion in the, no, three-month diffusion? No, six-month diffusion? They're both six-month diffusion. I could see Dante smiling. So I think he knows. There's one of them manufacturing, one of them's had one of them. Oh, no. I don't know what many, maybe, but this is October, November.

Oh, okay. October was 47.6. Oh, I see. And then it, uh, bounced back in November to 53, but 47.6 is quite low. Right. So you know like the diffusion index? What is it? Yeah, it's basically a measure of the breadth of job growth, right? So we can, we can talk about job growth overall being weak or strong and, you know, Dante covered that. But then we can also look at the share of businesses or industries, I should say, that are expanding their payrolls

versus those that are contracting them or holding their payrolls, uh, constant. So 47.6 is, is below 50. And historically, anytime you've had that diffusion index below 50, it has signaled recession. So that's the relevance here. Now, it did bounce back in November. And yeah, maybe there's some noise in the data here. So we don't want to overreact. But once again, it's suggesting some weakness in the breadth of job growth. It's not only few jobs being created. It's the ones that are being created are really concentrated in health care and a couple other industries. Yeah, that's a good stat. That's a good one. Okay, why don't we turn to any listener questions? Just become a new feature when we don't have a guest on Inside Economics, where we'll take a few questions. And please, listener, feel free to fire away.

We value those questions and we'll, you know, as you can see, we're going to respond to them as many as we can on air. So, you know, feel free. Chris, do you have a good one? I've got it. I usually does this. I know you're stepping in for her. I know, big shoes.

Big shoes, yeah. Hopefully you'll give me some leeway here. So this one is a question that came in. Actually, I got this question over email. And it's about the ACA subsidies. So with the ACA subsidies likely to expire at the end of the year, would this pose a risk to health care jobs? This has been the sector driving job growth,

so wondering how big of a risk to the labor market this would be. That's a great question. What do you think, Dante? I mean, I think, yes, I think it's bound to have an impact. I'm a little bit torn on what, you know, that's obviously part of the impact. We also have, you know, sort of Medicaid cuts that are coming also, right? So I think there's, you were talking about policy supports in 2026, but I think health care, specifically there's reasons to think that it's going in the opposite direction, right?

Policy is likely to be a headwind to health care, or at least in your term. Yeah, I don't think that happens immediately, right? I mean, especially on the ACA subsidies, I mean, because the channel to me seems to be, okay, fewer people are likely to be insured, fewer people sort of renew their health care policies in 2026. That eventually leads to probably higher uncompensated care costs at hospitals. And that has some financial ramifications, which could lead to less payroll growth. But that's not, that doesn't happen immediately, right? Those things all sort of play out over a longer time horizon.

I think the same thing's true with the cuts to Medicaid, right? they're phasing in over a period of time. So in my mind, I don't see it as a huge, you know, I wouldn't expect to see health care payroll growth slow dramatically in 2026, but it feels like you've got some headwinds to growth as you think about, you know, 12, 24, 36 months out maybe. Yeah, these are all provisions of the one big, beautiful bill act. And most immediate is the curtailment of the enhanced ACA health care subsidies

that were provided during the pandemic. under the Biden administration, it just expanded out the support that was provided to people using the Affordable Care Act subsidies and brought down their insurance premiums and allowed for a greater number of people to have health insurance. And those subsidies are now going away, unless lawmakers figure out a way to stop that, but it doesn't feel like that's going to happen here, at least not anytime soon. And the Medicaid cuts you talked about, and I think a lot of that's related to work requirements. requirements, that kind of thing. But I don't think they don't kick in right away, right? I think they kick in not this year, not in 2026, but maybe 27, 28. I think they're a little longer run. But the ACA health care subsidies, they expire and they're going to have an impact right away. But as you point out, that's going to have an impact on health care demand over a period of time and then impact on jobs over a period of time. So I don't know that that's a 2026 event, you know, maybe that's something down the road.

But it's a good question. It's a really good question. And, of course, it goes to the heart of the only sector adding to jobs. Are there any other threats, Dante, that you can think of, Chris, to the health care sector in terms of jobs? I mean, if we're so reliant on that one industry, I guess we should be really. really focused on that to make sure that we're not missing something in terms of because kind of the logic is for the growth is just purely demographic. We're aging and we just need more health care services and demand continues to rise. But, you know,

is there anything else out there that could jeopardize that growth? It's taken on such added importance now. Just an open end of question. Any thoughts there, Dante? Nothing that immediately comes to mind. I think your point, to me, it feels like, like the tailwind, right, the demographic tailwind probably, you know, far outweighs, even the, you know, headwind that you might get from the ACA subsidies expiring or, you know, future Medicaid cuts. It feels like that, you know, that demographic tailwind is going to be hard to stop, right? You're going to need, health care demand is going to continue to increase, even with some of those changes just because of the aging of the population. So it feels to me

like that's still going to be the bigger story. Yeah. You know, I don't know if I said this on the podcast before, but if you look at year-over-year job growth, the absolute level of job growth, and the last data point we had before the government shutdown, I think was through September. Philadelphia, our home, at least, yeah, we're all, we're all Philadelphians, right? Our hometown had some of the strongest year-over-year job growth of any metropolitan area in the country, you know, even stronger than like Dallas, you know, I don't think I'm making that up. I'm speaking from memory, but I think that's right.

You know how unusual that is? I mean, that is like incredibly, I don't think that's ever happened in my professional life, in my entire life. And that goes to health care. Philly is a health care center. You know, we've got all kinds of research centers and hospitals and Penn and Jefferson, chop, you know, just a major health care center. And that's just generating a lot of jobs and driving a lot of growth. I think Pittsburgh also is a health care center, isn't it?

I think they've got a lot of health care jobs. It's going to be just apportionate, yeah. So PA is really benefiting. Pennsylvania is really benefiting from that. Okay, that was a great question. You want to do another one? Chris, give another one? This is a fun one.

Do team members consider themselves economic optimists or pessimists? And where do these meanings come from? Oh, whoa. That's interesting. Should I go first? If you'd like. Dante shaking his head. You do, yeah.

Set the tone. I consider myself a long-term optimist. You know, never bet against the American economy kind of optimist. I mean, I do think we have fundamental inherent strengths that will drive economic growth in the long run. We can certainly screw things up in the short run, but in the long run, we kind of figure out a way. and we solve problems. I mean, we're a very, you know, market-oriented economy. You can make a lot of money if you are focused on a specific problem and you solve it.

We have historically been open to bringing the best and the brightest here and having them solve those problems and make their fortunes, and that's driven a lot of economic growth. You know, so, you know, we've got basic fundamental strengths like the rule of law. You know, we have laws. I'm kind of hesitating a little bit in the current environment because all those things are under some pressure in my humble opinion. But, you know, long run again, you know, I do think that those things will still prevail. So I think there's good reasons to be optimistic in the long run. Don't bet, don't bet against the American economy. That's a losing bet.

But, you know, having said that, we look at business cycles, the ups and downs in the economy, and depending on where we are in the cycle, and there's also a credit financial cycle that's related and also plays a role. But depending on where we are in those cycles, determines whether I'm an optimist or a pessimist. And at the moment, I'm, as you can tell, feeling somewhat pessimistic about what's going on.

I don't think the efforts to de-globalize the economy are constructive. They're very counterproductive, and you can see it in the data. I keep going back to, you know, we started, the job market went sideways right after Liberation Day in April. I mean, I think you can connect the dots very easily. So I do worry about the economy near term, and thus recession risks are high. But long run, I consider myself to be an optimist.

What do you think, what do you think, Chris? How do you, how would you answer that question? Well, I'm going to answer in a way that you don't like. Oh, really? Okay. I like to think of myself as an economic centrist, right, kind of down the middle here. However, I'd say probably, if I'm doing honest assessment, lean a little bit more on the pessimistic side in terms of seeing the risks. And that's just because that's the industry I grew up in. Or I got my first job in.

I've been in the risk, credit risk analysis for a long time. And there, you know, nobody wants to talk about the upside. or the bright side. It's always about what's the next shoe to drop, what could go wrong here, right? So I think that's still perhaps my bias. As I look at the data, I might be, well, what are we missing here to the downside, right? What's the downside? So, again, I like to think right down the middle, but probably a little bit more.

That's an interesting point. You know, many of our clients are financial institutions. financial institutions are all about mitigating risk. They don't generally participate on the upside. It's not like their equity, their debt. So if things go better than anticipated, they're not going to benefit from that. They still get their interest payment, the one they agree to contractually. It's not going to rise.

But if things go bad, they take it on the chin, right? They're on the hook. So tend to be more pessimistic. That's interesting. That's a good point. Dante, what about you? And why did you say I wouldn't like that, Chris? What would you think that?

Oh, centrist Chris, always down the middle. It got some teems aside. Well, no, no, there's no side. This is about optimism, pessimism. How can you be a centrist? You're saying I'm equally optimistic and as pessimistic. Yeah, balanced. Balance.

Yeah, in terms of my pessimist-optimist-optimist quotient. I'm kind of right down the middle. Got it. Dante, what about you? I think before you answered, I was going to describe myself as, I was going to call myself a big picture optimist. But I think in describing it is probably more about time horizon, kind of like you were talking about, than it is about big picture, narrow picture.

I think generally my view is that everything will be okay, right? Like, it doesn't mean that in the six months, everything will be great. But generally speaking, I think positively about the economy. I think certainly I'm a pessimist on some narrow topics. I think maybe I was more pessimistic. 10 years ago, yeah, I was trying to find my first job in the middle of the financial crisis. I feel like I came out in a little bit of a pessimistic bent, and then I feel like I've gotten a little more optimistic over time. I think certainly on specific topics, you know,

productivity growth. I can be a little bit pessimistic, but I think big picture, long run, I tend to be more optimistic. Yeah. Would you characterize Dante's up that way, Chris? Is that your perception of him? Not based on his productivity outlook. He's a pessimist on that, isn't he? He seems pretty pessimistic about AI. Oh, is that right? We haven't talked about that in a while.

We should talk about that again, but that's actually the next question. Maybe skeptical. Maybe not pessimistic, skeptical. Skeptical. What about Marissa? I mean, she's not here to tell us, but how would you characterize her? Do you have a way to characterize her? Dante, how would you characterize her kind of?

worldview, optimistic or pessimistic, or down the middle? I think my guess is she would lean closer to you and I than to Chris is my guess. I think she's a little more optimistic, but I don't know if she would agree with that. That's interesting. That's a great question, though, a thought-provoking question. Okay, I don't know. If we keep this short, then we should, if we're going to keep the podcast on the short, so we should end it right here.

Should we take one more question? Chris, what do you want to do? What's your pleasure? One more, I kind of teased it up. Okay, go far away. I feel like I got to go through with it. Okay, this refers to the November 17th podcast, so jog your memory. We were talking about the disruption in white-collar jobs during that episode.

Is there a concern that entry-level jobs will go away due to AI takeover? In my profession, the basic skills I learned were valuable to my growth and understanding of the industry, even if I were plopped into a higher level job early in my career, I would have, I would have flailed not knowing the basics. So I've heard this from other institutions as well, that they're worried about kind of the long-term consequences. Sure, they can not hire today because they're going to get these AI productivity growth, but are they going to undercut themselves by not having the management team, you know, five, ten years from now that is being trained today in the entry level? Any thoughts around that? So the thinking is that AI is going to do more damage to hiring among younger people. AI is going to do damage to hiring because younger people are more likely entering into the workforce.

They need to get hired. They're the ones they're going to pay the brunt for it. And also, I guess you could argue AI is more likely to substitute out the kind of tasks and skills that younger people. I'm stretching, but that's kind of the intuition. Yeah. And then you're saying, okay, if we're not hiring young people into our institutions, into our companies and into government, what's the world look like 10 years from now because you lose that cohort, then you lose your middle managers, and then 20, 30 years from now,

you lose your CEOs and your CFOs and everything else. That's kind of the thought process. That's right. Oh, wow. Do you subscribe to that view? What do you think, Dante? He's a pessimist on productivity anyway. I don't think that's going to happen.

Now he's an optimist. Now he's an optimist. If you want me to take that, okay, if we assume that that is true, right, there will be some disruption to entry level. In my mind, it would make more sense that just what the definition of an entry level job will change, right? The types of skills and the types of tasks that entry level employees do might evolve over time. But I don't think there's a world where you just don't have.

new workers entering firms and entering the workforce, right? It just feels like sort of the types of things that those entry level or new workers will do may evolve if AI becomes, you know, as much of a health and productivity boost as we think it could be. So, yeah, I don't think they're arguing there's no hiring. I think there's this, right, the potential concern is that there's just less higher, significantly less hiring. And, you know, it does feel like the current data suggests that that's what's happening. If you go look at younger people, their hiring rates are way down and they're having a hard time getting into the workforce. I don't know if that's AI or other stuff, uncertainty, everything else, but it is suggestive.

But I think it is a reasonable concern. The other concern kind of related is allowing younger new entrants or facilitating younger new entrance use of AI. Because if you use AI, you may not learn the basic skills that you need to succeed going forward. You need to do the blocking and tackling. This is an argument I've heard. Even in our own world, I've heard this, you know, in managing our own company. And I've even heard one of the management went so far to say, well, maybe we shouldn't allow them to use they being new entrances of the workforce don't allow them to use AI because if they use AI, then they're not going to learn what they need to learn to be good economists down the road. My immediate reaction

to that was, well, if that logic holds, maybe we should take the internet away and let them, give them in Sac, Pletia Britannica, maybe that would help. That was my reaction. I actually said that. I probably shouldn't have said that. It's blurted out. I don't know. Do you have a view on all this, Chris? I mean, what is your sense of things? So here I'm a little bit more optimistic, right? More optimistic. I look historically, we've been through rough labor markets in the past, right?

We've had, you know, we've had generations. Yeah, good point. Very difficult time entering into the labor market, high unemployment, and everything kind of worked out, right? They learned the skills. Eventually, they picked up, right? Maybe they got a boost from some additional technology that came along. So I don't know, I believe that even if there is some weaker hiring today among the entry level that they'll still pick, there'll be sufficient number of them that do pick up the skills. And things are going to evolve as well.

So the skills that a middle manager or senior manager or CEO are going to need, I don't know, 20 years from now, they may be quite different from what they may be learning today at the entry level. kind of to your point, I don't know that we always have to go all the way back to first principles. When someone learns a new field, they can get an edge up, right? My kid, or other students today, they're using scientific calculators, right? I don't know that that makes them terrible mathematicians, maybe some of them, but I think there's also certainly some real advantage to be able to leverage that technology. So, well, just the obvious, I'm less pessimistic. Just the obvious point that if you want to advance, you've got to take some of this stuff as given, right? I mean, because if you go back and learn everything, then it's going to be pretty hard to move forward, I would think.

Yeah. Yeah. But going back to the question about optimist or pessimist, I do sense, and I say this with no data, I just sense it talking to younger people all over the world, that they're just generally more pessimistic. I just, I get that sense of things, that they're worried about really big things that are so overwhelming that they can't get beyond it. Like, you know, climate change or, you know, what's going on with regard to autocratic governments around the world, these things matter. And I worry about birth rates and, because, you know, you have to be an inherent optimist to have

children, don't you? I mean, fundamentally, you have to be optimistic to have a child. right to bridge on. You have to be naive, I think. Or naive or both, or both, or both, or both. Anyway, these are great questions, thought-provoking. Yeah, very, very good questions. And we are going to call this a podcast at an hour, so this is a little shorter.

Not a whole lot shorter. It's a little shorter. And we'll be back. I think we record another podcast this coming Friday. We'll have more data. Consumer Price Index. CPI comes out on Thursday. I think we got UMIS, University of Michigan Survey, Philly Fed Survey.

What else do we have? Anything else coming out that's a consequence? No. The government's still behind, and we still don't have release dates for a lot of the data, but it should be coming where else to lose soon. Okay. Well, we, I think we'll call it a podcast, dear listener. I hope you enjoyed the conversation, and we will catch you next week. Take care now.

Thank you.