Moody's Talks - Inside Economics - Pulse on Payrolls and Prices

Episode Date: February 13, 2026

Together at last. In a rare joint performance, Dante and Matt join the Inside Economics crew to unpack January’s jobs and CPI reports. The brief federal government shutdown delayed economic data releases a few days, which made for a loaded slate this week. Dante shares his impression of January’s seemingly strong jobs report and then the team plays the stats game. A brief hiatus did not affect Marisa’s ability to dominate. Matt then goes through the first inflation data of 2026, and where it looks like inflation is headed in the coming months.View the Full U.S. Macroeconomic Outlook Webinar here: https://events.moodys.com/ta6186-2026-bank-odwbn-mau28334-us-economic-outlook-q1View our AI generated paper here: https://www.economy.com/getfile?q=165AB685-ED95-43E8-8533-DA2CE131A01A&app=downloadHosts: 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' and BlueSky @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 Zandi, the chief economist of Moody's Analytics, and I'm joined by a few of my colleagues, of course, my two trusty co-host, Marissa Dina Talley, Chris Doretti. Hi, guys. Hi, Mark. Good to see, Marissa. Nice to see you, Chris.

Good to see you guys. Marissa, you weren't here last week, but you're here. I know, I missed you. Yeah. But I did listen to it the other day. It was a good podcast. And I appreciate, I have to say, that you refraining from the game in my absence and in memoriam.

I like that. Yeah, of course. I mean, we were completely rudderless without you. I mean, it was, I'm surprised we even pulled off the podcast at all. Yeah, I know, but no, we would, you know, you're the heart and soul of that game. So, which we will play today because you're here. We're going to play that the stats game at some point, but good to have you back. And we have two other colleagues, two regulars.

We've got Dr. De Antonio, Dante Di Antonio. Hey, doctor. How are you? I'm doing right, Mark. How are you? I'm very, very well. Of course, Dante joins us when we've got jobs numbers and we got some jobs data this week, the payroll employment data, the household employment data. I meant to ask, did the population controls come out yet for the? No, next month. Next month. Darn, next month. Okay. We'll have to wait for that. And then we've got Matt. Matt Collier. Hey, Matt. Hey, Matt. Hey, Mark. Hey, everybody. Has it, has the stars ever aligned where we've had the two of you on the same podcast? Because we got CPI inflation,

consumer price inflation today, and you're always on for that. But historically, the inflation numbers and the job numbers don't coincide. They're not the same weeks. We have you on separate weeks. But is this the first time we've had you both on the same podcast? We just talked about yesterday. I'm convinced I remember we did a Saturday morning podcast one time, and I think it was just you, me and Matt, Mark, for some reason. I don't know why. No one else was available. I'm convinced that happened. Maybe I'm making it up. I don't know. I don't remember it. I was told Dante I was going to pretend I didn't know. This is my first time meeting him. He's like, oh, no, we've already done this pot.

So they died. Finally, nice to meet you face to face in person. But I don't remember. I don't remember that Saturday podcast. I don't, yeah. I don't remember that either. It's a fever dream. I don't know.

It was a dream in Dante. I don't even remember what we talked about last week. So that doesn't mean anything of a, you know, I don't remember. But you apparently remember Super Bowl commercials from 10 years ago. Well, no, it was longer than that wasn't? 20 years ago, almost. Yeah. 20 years ago.

Yeah. Right. Yeah. So that's an inside comment. You have to go back to the last podcast because we were talking about Super Bowl ads. And my favorite Super Bowl ad was a FedEx commercial. I didn't know this, but you guys figured this out. 2006, right?

2006. It's where the caveman, well, I won't tell it. But I highly recommend 2006 FedEx, YouTube it, cavemen. You'll find it right away. And to this day, 20 years later, I'm still laughing when I think about that. It's a metaphor for life. Did you guys watch the Super Bowl? Did anybody watch?

Yeah. It was part of it. Yeah. Well, what about this year's commercials? What did you think, Marissa? Lots of AI commercials, not a surprise. Yeah. Yeah, there were some good ones.

There was like an anthropic, dystopic, AI commercials, weren't they? I mean. Yeah, it's kind of weird to like point out. as you said, the dark side of AI by an AI company that's saying, but we won't do that, right? Everybody else will do that, but we won't do that. Those other guys, they'll do it, not me. But we're obviously all thinking about doing it, hence the commercial.

Exactly. Yeah, I found that a little weird. I'd like to talk to the CMO of Anthropic and see what exactly the thought process was. But anyway, I came away more nervous and scared, not less nervous and scared. Anyway, okay, well, we got a lot of ground to cover. As I said, we got the job numbers this past Wednesday. We're going to go over those with Dante. And then we got the consumer price inflation this morning.

This is Friday, February 13th. Oh, Friday the 13th. Ooh, okay. Ominous. This dawned on me. And we'll play the game. We've got a few, I think a couple listener questions related to jobs and inflation will take those, and then we'll call it a podcast.

So with that, Dante, give us a rundown on the jobs data for the month of January 2026. Sure. So I think, you know, headline was that the report was better than lots of people expected, certainly better than I expected. We can talk about the details maybe weren't quite as favorable as the headline might suggest. But we got 130,000 jobs added in terms of total payroll. The private sector was actually stronger than that, just over 170,000 because there was a pretty sizable loss on the government side. Three-month average gain is 73K now at a top line number.

So we finally rolled off that big October decline that we had. So the three-month average is positive again. On an industry basis, it's the same story that we've had for a long time where most of that growth was in health care, right? Healthcare added just over 123,000 jobs of the 130 headlong. line. A couple other industries were sort of notably positive. Construction had a big jump in January, up 33,000. Professional business services was up a little over 30,000 as well. So it wasn't, you know, just health care adding jobs. On the downside, you had obviously the public sector down about 42,000. And then you had a couple of industries with declines of, you know, sort of more than 10,000

jobs, finance, information, transportation warehousing all had, you know, sort of reasonably sized declines. So sort of a mixed bag in terms of the details, but still, you know, health care doing most of the heavy lifting. Wage growth, you know, maybe... It's not going for a second. So how did you begin your description? What was the first sentence? Does anyone... I said better than expected, right?

So, I mean, if you look at it... I mean, immediately my... I go, no. I think I caveated it immediately by saying the details maybe, you know, aren't, you know, quite so favorable. but just in terms of headline number. Consensus, I think, was at 70K. My forecast was 40, right? So coming at 130 is certainly better than that.

You were at zero, man. Oh, yeah, Mercia was negative. Right, right. But it was actually negative. I mean, because you're going to get to the punchline here, right? I mean, you're benchmark. Yeah, January was up 130, but what, you know, would your characterization of the job market change based on that 130?

No. No. Why? Because why? Well, I mean, there's a couple reasons, right? One, I mean, obviously the benchmark revisions paint a much more negative story about the previous data and the 130. I think we can talk about it. But healthcare, there's clearly some seasonal issues going on in healthcare. My read is more that job growth is, you know, basically should be 70K in November and December and January. They should all basically be the same, right? Healthcare just the seasonal adjustment sort of shifted a big gain to January when it shouldn't have been there.

So I don't think there's not an actual uptick in January in terms of job growth. I think things are still sort of holding where they were. All right. So we got this big jump in January job increase, 130K, and you're saying almost all that's in health care. You know, you had some other sectors add a bit. You mentioned construction, I think manufacturer added 5K. And then you had some industries that lost jobs. And the net of all, those were basically nothing, you know, zero.

Okay, so it's all health care. But all the history got revised down, big time, right? That's what you, when you say benchmark revisions, that's the annual revisions that come out of, and we described this in some detail, you know, what that is in the last podcast. And if you look at, you look at, take a step back and take a look at what's happened over the past year or so, that doesn't, that my. my perception, my sense of the job market isn't better.

It's worse, isn't it? I mean, is it worse, though? I mean, you already were pretty pessimistic, right? I mean, you already thought job growth was zero. Yeah, so I agree that it shouldn't be better, but it can't get much worse, I don't think. Right, because, you know, if you look at the January employment level, payroll employment level, it's basically where it was back last April, right? I mean, it really hasn't changed much at all.

Maybe up a little bit, maybe up 100K, you know, 150K over that nine-month period. But since last April, which, you know, to remind everyone, that was Liberation Day, you know, when the reciprocal tariffs were put in, we've effectively, even with healthcare, created no jobs, no jobs, right? Would you agree? That's right. I mean, yeah, if you look at, you know, just the balance, all of 2025, right, we only added, after. all the revisions, we added 181,000 jobs in total in 2025, right? Right. So, and most of that came in the first three or four months of the year. Okay. And then here's the other thing you alluded to it, but maybe you can flesh it out. Are you suspicious of that January number? Do you think that

also was affected by seasonals or might be revised? Do you think that 130 is going to stand? I think it might stand. So, I mean, most of it was in health care, right? So, I mean, Health care is the thing that I think we need to focus on that gain in health care is the largest gain ever in health care outside of the immediate aftermath of the pandemic, right? So it's the largest monthly gain ever, which stands out as unusual. So if you go and look at some, I'd never really look to this. Healthcare actually is a bit seasonal in terms of employment. So you typically get a little bit of a run-up in health care employment in the fourth quarter.

And then on an unadjusted basis, health care employment typically declines in January. So the seasonal adjustment factor essentially is providing positive lift to health care, right? It's adjusting for that normal decline that happens in January. That didn't happen this year, right? So the unadjusted healthcare employment was essentially flat in January. It was almost exactly zero. So the seasonal adjustment is what boosted it from zero to 123. But the reason for that is that the seasonal pattern was much different this year.

So if you go back and look at October, November, December, the buildup in health care was much smaller than it's been in recent years. And so you actually had smaller gains in October, November, December, and then you got this big pop in January because now the offsetting decline didn't happen. So that's why I made the argument that basically the gain that we got in January should really have just been spread out over the last couple of months, right? Healthcare was about 20K two week in October, November, December, and now it all showed up in January. So I don't think it would get revised away, but it's just, it's showing up all in one month when it, you know, should have been spread out if the seasonal pattern was like previous years. interesting. And I guess that also argues that we shouldn't count on these big outsized gains going forward. Healthcare will continue to add in all likelihood, but not like what we got in January. Yeah, I mean, there might be a little, you know, so if you look, health care, the sort of growth in health care in the fourth quarter was about 90K, two week.

And you only really got about 50 or 60 of that back in January. So there might be a little bit of sort of residual seasonality that provides some lift in February, March. But most of it happened here in January. Okay. Got it. Got it. So when you look at, I ask you this every month, and since you've been on now two weeks in a row, I ask you every week. Given what you've seen here, and we'll come back to the household employment data and the unemployment rates in a minute, but based on this data that we just received, what is your sense of what I call underlying monthly job growth, abstracting from the vagaries of the data, the seasonal patterns, everything you're just discussing. You know, what is, what's reality here? It's the question I like the least every month, by the way.

I think it's, I still think it's around 50,000. I think that's probably my answer for the last couple months. Yeah, around 50K? Okay, 50K per month. Of course, we didn't get that, we haven't gotten that for quite some time on a consistent basis, but you're still think it's 50K. You mean, over the last three months, right, Jobger's actually stronger than that. It's 73, right?

Okay. It depends on how far back you take the ruler, I guess. Yeah. Correct. Okay. And what do you think so-called break-even monthly job growth is? That's the rate of job growth necessary to maintain stable unemployment. And again, we'll come back to the unemployment numbers in a second. I still think it's around 50K as well. I think we're sort of, we're right there in terms of break-even. We've

seen the unemployment rate go up a little bit. Now it's come down a little bit. So I think we're sort of hovering right around break-even. Okay. Before we go to household, let me turn to Marissa and Chris. Looking at just at the payroll numbers. Marissa, anything you want to add there in terms of your sense of what's going on? I agree with you. I'm not, this doesn't change right. I'm different from Dante.

I mean, do you agree with Dante or you agree with me? I agree with you, Mark, the one asking me. Okay, then go ahead. Feel free. That this doesn't change my mind about the job market at all. I think this will get revised lower as every single number has for the past almost to year. I didn't know that about the health care seasonality. That's strange to me. I wouldn't have thought health care was a seasonal industry. We also got to pop up in construction that was kind of

big on the non-residential side. I'm not sure what to make of that. I don't know if that there might be some seasonality going on there. The BLS had put a note in the news release that said that the weather in January didn't, didn't mess with the seasonal factors or didn't play a role in any of the, you know, the numbers up or down. I feel like they always have that note whenever there's a bad storm and yet there's always something wonky in the numbers. So yeah, I mean, I don't, I don't think that the number is 1.30, not at all. I mean, the other thing I would point out is we continue to see government losses, federal government job losses. And We had sort of been given a heads up actually by a listener about this a few months ago that some of the doge cuts that were supposed to be finalized on October 1st.

Some people actually got even more deferred resignation that took them through the end of 2025. And I think we saw that in this report as well. Interesting. Okay. So what is your estimate of underlying job growth? I think it's around like 40K, like 30 to 40 somewhere in there. And what's your sense of break-even? I think break-even is probably 50-60.

Similar to Dante. Yeah. Okay. Okay, Chris, anything to add here on the data? Did your sense of the job market change as a result of all the ups and downs and all around here in the data? Not dramatically, but the number was stronger. Even though I do believe there will be revisions, you know, we started off the

year a bit stronger than I would... Then certainly I would have anticipated them what we said last week. So it doesn't change my long-term view. Bigger than I thought they'd be, didn't you know? Well, for January specifically, right? I'm talking, you know. I'm saying your general sense of the labor market. No, my general sense hasn't changed that long-run view, the break-evens and all that.

But January was, you know, even accounting for some of the noise here, I suspect that it's going to come out stronger than we had anticipated last week. Mm-hmm. Mm-hmm. Okay. Okay. So your underlying is what? Probably around, I'd say it's probably 40K for both. For both. Okay. Okay. So we're all, you're all in the same kind of ballpark. Yeah. Yeah. Okay. Can I ask though, and I read this, I didn't investigate myself, Dante, and this goes back to the seasonals, but the overall seasonal adjustment, if you look across all industries, was pretty large, pretty significant. Meaning in January, you always see seasonally weak employment because it's in the middle of the winter.

And so for the Bureau of Labor Statistics, tries to account for that with the seasonal adjustment. And that adjustment was a big part of that increase. It was just very large. You saw in other industries as well. Right. I mean, just, right. Broadly speaking, you always get a very large decline in unadjusted employment in January. So the seasonal adjustment is always, you're doing the, you know, there's always outright declines.

clients, usually by more than a million jobs in January, if you just look at the unadjusted data. So the seasonal factor is always very large in January. I think health care was unusual, right? Other industries, I think, based on my quick look, we're sort of more normal in terms of seasonal pattern. But yeah, you always get this big adjustment in January because of the seasonal drop that happens. Okay, okay. And, you know, often when there's storms or winter weather, the BLS will publish, you know, these are the folks that might have gotten affected by the weather. I didn't see that. Did you see it?

Was there anything that stood out there? No, I mean, I think the worst, the big storm in January happened after the reference week, right? So, I mean, that shouldn't have had any impact. Obviously, I think there was some smaller weather issues maybe earlier in the month, but the big storm that affected, you know, lots of the country didn't happen until after. Right. Right. Okay. Okay.

Mercy, you said there was a question from one of our listeners about the jobs numbers or jobs data in general. Did you want to fire away on that while we're here? Yeah, sure. Let me pull it up here. It was actually about the difference between the jobs report and ADP and Rivello. You know, you guys talked about this last week, right? This is data that we look at from these private sector sources to inform our forecast for the jobs report. So ADP and Rivellio were both quite weak and would have suggested, you know, if you kind of average those two private sector numbers, right, would have suggested.

something, I forget, something around zero, right? That's why I said zero, because that's what I used, right? So this listener is asking about the, one, he says, one of the major divergences between the BLS and ADP was professional business services. BLS reported plus 34K jobs in professional business services, whereas ADP reported minus 57,000. And he said the net difference between the two accounts for almost the entire difference between the two top line numbers. And he says, and I didn't verify this, but I'll believe him. This seems to be a consistent trend throughout 2025 as well with the BLS showing stronger growth in professional business services while

ADP shows losses. What do you make of this? The question is, what do you make of this divergence and do you think it's significant and do you have any guess as to which number is, more accurate. Dante, do you have any views on that? I have a couple thoughts. One, I mean, I think we always have to be careful in any given month, right? The sources are not going to perfectly align. I think we've shown that if you look at a three-month average or even better, a six-month

average, you know, they tend to align pretty well in terms of, you know, job growth over a six-month period for BLS versus ADP versus Revelyo. So I think if you look over a slightly longer time horizon, the trends and the signals tend to point in the same direction. But I think in any given month, we have seen, and I think we'll continue to see these big divergences. I would say particularly in a month like January where we know seasonal adjustment plays such a big part in the actual headline number, all those other sources have to deal with those same issues, right? They're having to do their own seasonal adjustment and the underlying unadjusted data that they have maybe is likely different from what BLS has. And so I think in those months where seasonality plays a particularly big role, I think it's much easier to see why you would have divergences because of the sort of, you know,

sort of mechanical differences that pop up there. So I would say one month doesn't make me, you know, sort of feel less good about those other sources. I still think they're useful. But again, I think, you know, in terms of predicting job growth in every given month, they're not always going to be accurate. Yeah, I think the seasonals might play a role, particularly in that January number. That might be part of it.

And professional business services, a big part of that is temp help, right? I mean, that's a big component. and, you know, maybe ADP and BLS that aren't, you know, consistent with each other when it, when it comes to measuring temp employment. But that's the point. And you would also think that that would be very seasonal around this time of year, the temp portion, right? Because a lot of, a lot of workers that are hired might be temp agency workers to do seasonal work. So, yeah.

Yeah. Hey, Matt, did you want to say something? But if you do, then we have to ask permission from Dante, you know. Yeah. He's cool. He's a nice manager. He's been pretty, yeah. He's your manager?

I didn't know that. Yeah, he is. Right. Which would have made my joke about not knowing him funnier, I guess. But the, Marissa, did he say that they diverge all or did the, the value is always lower or that they're just always different? They're always diverging. He's talking about. I'm sorry, AP, okay.

Yeah, yes. It's always consistently lower. That BLS has shown stronger growth in professional business services during the year than ADP has showed it to be weaker. Gotcha. Yeah. Although it's been quite weak, right? I just saying special service has not been, no. BLS is not saying it's doing well.

Yeah. Yeah. And a lot of it has been temp. And that's been going on for years, the temp component, the decline in temp jobs. Let's go back to household employment, the basis for the unemployment rate. Anything there you want to call out? I mean, the unemployment rate dipped to 4.3%. It peaked at four and a half a couple months ago. We're back down to 4.3. What do you make of all that? Yeah, I mean, I guess I would throw a little caution. It certainly looked strong, right? Unemployment was down. The labor force grew. The number of unemployed workers were down. You know, household survey employment was up by a large margin. But I think all that has to be with the caveat that we didn't get the population controls, and they're going to go back and revise that January data with the updated population controls at some point. So that, you know, I think until we get those updated controls, it's hard to read too much into particularly the levels.

I think the decline in the unemployment rate is good. And, you know, those updated controls don't tend to have a huge impact on the sort of rate estimates in the household survey. So, yeah, I still certainly can't read it as bad news, but I would caution, you know, sort of being too optimistic based on that household survey data right now. Yeah, and you said the population control, so just explain with the population controls and I'd like to know when are they supposed to come out from Bureau Labor Statistics. Sure. So once a year they take updated population estimates from the Census Bureau and they use that as the basis to update the population controls or the weights that they use in the household survey, right? So they have to take the sample responses and they have to weight them by something to get those sort of population labor force employment numbers and those weights or controls that they use. are based on those population estimates from the Census Bureau.

Typically, those new controls are implemented in this report, right? In the January report, which we just got, because of the government shutdown in November, that process was delayed, right? So they had already previously announced that those controls wouldn't be implemented until the next report, right? So the February report released in early March. And they also noted in that report that they will, typically they don't go back and revise old data with population controls, right? They just implement them in January and you end up with, you'll oftentimes, level shifts between December and January in terms of the actual level measures like labor force

or employment. But they did say that because they sort of missed January, that they will go back and revise the January data at some point. They didn't necessarily guarantee that those revisions will come out next month with the updated controls. That's my hope is that they will, but they didn't commit that to paper. So we will get new January data at some point. We'll definitely get the revised February data with the updated controls next month when the report comes out. Yeah, the one thing in the household, this is the household survey data that stood out for me, if I've got this right, is the continued increase in unemployment for college educated workers. That keeps moving higher here. Yeah, I think if you look over the year, you know, like January 2025 to January 2026, it was up, I think like 0.6 percentage points, which, you know, it's like the biggest, if you look across the educational groups, it was definitely the biggest movement of any of those groups, yeah.

Yeah, I think we're at 2.9. I mean, I hope I didn't take anyone stat game statistic, 2.9%. I mean, do you think, I know I'm really pressing here, but AI? Is that kind of what's going on here? I mean, I think that's certainly what we've been talking about. And I think you can't ignore it as at least part of the reason, right? Is it the only reason? I mean, I think just the general slowdown in hiring also hurts.

So it depends what you attribute that slowdown and hiring to, right? If that's AI, okay. But just the slowdown and hiring AI aside. affects those new graduates a lot, right? People that are trying to enter the job market for the first time get disproportionately affected by that slowdown in hiring that's been happening. Right, okay. Except the educational attainment unemployment rates are for people 25 years old and older.

Yeah. But I think if you look at, I mean, they do publish a 20 to 24 year old. Right. I think that also has been up. I thought the one we're referencing, but I think that also shows some weakness, which would get more at the sort of new entrant measure of that. Yeah.

Yeah, you're right. That number, the 2.9 is not affected by new entrants, probably. And actually, wasn't there a Stanford university paper, used ADP data and showed that the increase in unemployment is very, much more pronounced for younger workers in the tech sector that are obviously going to be more educated. So kind of connect, and they connect the dots back to AI. So it feels like AI's imprint is,

is starting here. We're starting to feel it. We're starting to feel the effects of it. Okay. Marissa, Chris, anything else on the household numbers before we play the game and then we'll come back to Matt and talk about inflation?

Chris, anything? Not on the household, but did his comments reminding me that there was a change to the birth death model as well this month? I'm wondering what his view is in terms of the impact. It's a very nerdy podcast.

We are really laid down into the bowels of this. But that's a good point. You want to explain that, Dante? Yeah, well, I think there's two things. And Marissa can correct me if I had missed one of these. So one thing that happens is that with the benchmark revisions, right? So the benchmark typically only goes through March of the prior year, right? So the actual benchmark to QCW was through March of 2025.

But you saw that there were still big revisions to the data beyond that, you know, from April to December of 2025 as well. So those revisions beyond the benchmark point happened because, one, you know, where they update the seasonal adjustment factors, and two, they update the birth death model adjustment in those months based on the new benchmark data that they have. And then the other announcement that they made is that they're moving, they're shifting to a new birth death model where they're using current month sample information to estimate that birth death model. So they've gone through a couple different iterations of the birth death model over time. I think if you go way back,

they used to do, they used to create those adjustments just once a year. Right. So at the benchmark point, they would estimate the birth death factors and adjustments for the entire next year based on the data they had at that point. At some point, then they shifted to a quarterly process where once a quarter, they would incorporate new QCW data. They would sort of estimate those birth death adjustments every quarter to try to get sort of better estimates. And then now they're moving to a monthly process where they're actually going to use some monthly sample information in conjunction with the QCW quarterly data to again try to provide better sort of real-time estimates. of that birth-death adjustment process. So they're trying to get that more accurate. They know their issues with the time lag

and sort of the further out you're trying to forecast those birth-death adjustments that likely less accurate they're going to be. And at risk of butchering this, I mean, we need the birth-death models to get out companies that are forming and companies that are... Dissolving out of business.

Going out of business. That's right. This is the words I'm trying to describe. So the survey is, you know, companies that are around. and are still operating. And that's the bulk of companies. But every month, the economy is very dynamic.

You have new companies form, obviously, and you have companies that fail. And to try to account for the job implications of that, they have this model, the Bureau of Labor Services has this model called the birth-death model, and that's what you're talking about. That's right. You're right.

The monthly estimates are based on what they call a matched sample, right? So the change over the month is calculated based on, you know, companies that I observed last month and this month, and I can see their employment change. But that obviously there's two problems, right? I can't, so if a company doesn't report this month, I don't know if they just didn't respond or if they went out of business, right? So I can't use that match sample to bring in the effect of business closures. And on the opening side, right, new businesses can't be selected into the sample

immediately. That takes time. And so they, you know, they need some way to estimate the impact of those new businesses that are opening and the businesses that are closing because they're not baked into that sort of match sample estimate that they produce. Got it. This put a pin in this podcast because, you know, this, we're going into the bowels here. This is really good for people. People really want to understand how this sausage is made. This is a pretty good podcast to listen to. Chris, were you going to say something?

Did I? I was going to continue. One quick question. Do you expect that this will reduce the revisions or the size of the revisions going forward? Or is it just more around the edges in terms of the accuracy improvement? I mean, I think time will tell. I certainly don't think it'll hurt, right? I don't know that it's going to solve all of the problems.

And certainly in some years, you know, it'll be better than others. I think it's one of those where when the economy is pretty stable and things aren't changing a lot, even if you estimate birth death a year ahead, it does a pretty good job. But when the economy is very volatile or you're near a turning point, that's when that, you know, sort of that forecasted birth death has always struggled. And so this should help, I think, at least a little bit, but I don't know that it solves the problem completely. Thanks. Okay. Let's move on.

Let's do the stats game before we come back to Matt. We all put forward a stat. the rest of the group tries to figure that out with clues, questions, deducted reasoning. The best stat that's not so easy, we get it right away, not so hard that we never get it. And if it's apropos to the topic hand, all the better, but not necessary. And we always begin with Marissa. Marissa, what's your stat? My stat is minus 57,000.

Minus 57K. In the jobs numbers? Yes. In the payroll survey? Yes. Is it a change in employment in a certain industry? No. No.

It's a change, though, in employment. No? Yeah. It's a change in employment. Oh, boy, this is what she does. You know? She pauses. She delays.

She head fakes. So it's not an industry. It's like the AI, right? It's not an industry. You're like the AI, like the commercial, right? Pause. Yeah. Throw us off through it.

Yep. Taking it all in, calculating, answering. No, it's not a specific industry. But it is a change in employment. Yeah. See that? See that? Damn.

Because that isn't, I'm hesitant. My hesitation is because that's not the way I would. Characterize it. Characterize it. But technically, you're right. Is it an average of something? Is it an average of some kind? is an average something.

Are we going to get this if we try? Eventually, I'm sure you'll get there. Eventually, we'll get it. It's enough to do with the public sector. No. No. I don't know. I give up.

Do you guys want to continue to torture yourselves? I don't know. Matt hasn't. Matt, you don't want to stay on mute. I am inflation mindset, so I cannot go back. This is difficult. I thought we were going to do this at the end, so now I'm scrambling to find a jobs-related number. No, you don't need a jobs-related number.

You can do CPI. You can do whatever. We'll come back to you. Go ahead. It's good leader. Yeah. It's okay. No worries.

Okay. Mercer, tell us. Give up? Yes. We give up. This is the average revision. The average first month revision throughout all of 2025 was minus 57,000. Wow.

Well, do you remember Powell, Chair Powell, said 60K? He thought the numbers were overstated by 60K. I wonder if this is kind of sort of what he meant. That's what it is. That's where he got it from. Yeah. That's where he got it from. So you're saying after you go back to the data a month ago, and you look at the data today after all the revisions,

on average, the data has been revised down by 57,000 jobs. jobs per month. Yeah, it gets revised a couple times, right? So I'm not looking at the first to the final. I'm just looking at the initial revision, the first month's revision, which tends to be the biggest one because the response rate is the lowest in that first month. And as you move through time, you get more responses. So the revision's kind of lessen. So the difference between the first report and then the subsequent month, the average throughout 2025 is almost 60,000. Interesting. That's big. Yeah, it's huge, right?

And if you look over history, that's the third largest number we've ever seen outside of, A, the pandemic or B, the financial crisis in 2008. So that tells you something, I think, about the state of the job market. And more reason to be skeptical about numbers that are coming out that might look really great initially, you know, we could get, you could easily, that 130 could. easily be revised down to 70 next month. And you said this is the largest downward revision except for 2008, the GFC. And what was the other year?

The pandemic. So 2020, yeah, 2021. 2021. Wow. And, you know, clearly it goes to the, this happens generally when the economy is throttling back. And obviously in those two previous periods, we were in recession. So it's so interesting that the economy kind of hung together even despite that pretty massive downward revision.

Yeah. I mean, it usually, right, it's usually indicative of a bad recession. Yeah, bad recession. When you see numbers like that. Now, and I'm only, I'm not talking about an absolute value term. So I'm talking about the biggest negatives, right? There are also, on the flip side, very big positive upward revisions when the economy. economy is coming out of a recession and recovering and BLS is missing all that dynamism of new

companies forming or whatever it is. Right. This is an unfair question, but it's generated by this observation. What do you suppose is going on? Why didn't we suffer? You know, we saw this huge downer revision. Typically, that would signal the economy's falling apart. We're in recession. What, you know, What's different about this go-around? I think this go-around, we have a massive pullback in labor supply at the same time that we have a pull-back in demand. Typically, you have steady supply, employers pull back on labor demand, unemployment shoots higher. This time around, we got this immigration reform that massively pulled back millions of people out of the labor market at the same time employers were facing all these headwinds from tariffs and policy changes and a slowing, economy. So you got this sort of matchup, right, that we're in this now low kind of steady state

where, as we said, what we think, the break-even jobs number is very low compared to probably what it was a year or two ago. Yeah, the key is unemployment, isn't it? That's how you go into recession. It's when the unemployment rate rises. And what you're saying, it makes a total sense is, yeah, labor demand's been weak, but labor supply has also been weak. The unemployment rate really hasn't risen to any significant degree. Therefore, that's how this can happen without a recession. That's what you're saying. Yeah.

Very interesting. Okay. Let's do another one. Dante, you want to go next? Sure. 3.3 percent. It's an unemployment rate? Oh, is it a CPI number?

CPI. It's in jobs. That would be tricky. That would have been tricky. The ECI wages? It is not, what did you say, Marissa? ECI wages, wage growth, year-over-year wage growth. Yeah, private sector wages in the ECI, the employment cost index. Oh, see, this is why we can't do this

without Marissa. Geez, otherwise we'd be still at it, you know? Yeah. So this is the employment cost index year over year for private sector wages and salaries. That's right. Yeah. And it's quite a bit different than the jobs report measure of wage growth, right? So they tend to align pretty closely, but the, you know, So in the payroll report, wage growth was still 3.7% year over year as of January. And it was actually, I think, 3.8% in December. So the ECI goes through the end of 2025. And it was only 3.3%, right? So quite a bit lower than sort of reported wage growth in the payroll report.

And I think it also goes to that. You know, the unemployment rate is still pretty low. You've had this big reduction in labor supply. If anything, you think that would argue for sort of a floor under wage growth that should help support wage growth, you know, if the labor market is still, you know, sort of somewhat tight given that reduction in labor supply. But wage growth looks to be falling, right? I mean, the ECI has been on a pretty steady downward trend here over the last year. And I think it, you know, gets back to those questions of, you know, real wage growth. You've got inflation that's still

sort of stubborn and you've got, you know, wage growth that's falling. And do we get back into a period where real wage growth is negative like we saw in 2021 and 2022? And that would certainly, you be an even bigger problem for household finances than consumers are already under pressure. And so does that add to that pressure in 2026? Yeah, this kind of suggests that wage growth for lower wage workers may actually be less than inflation, right? Because if you're at 3-3 on the aggregate across all workers, and we know from the Atlanta Wage tracker data that folks in the bottom, a couple quintiles of the wage distribution or seeing much weaker wage growth than the folks in the top part of the wage

distribution. This would are in inflation is kind of around three. We'll come back to that in a second. That feels like we're saying that real wages after inflation wages for for low wage workers is is actually declining. And that that would be consistent with, you know, what we're hearing, you know, in consumer sentiment surveys, that kind of thing. Does that, does that, I know I'm piecing a lot of data together and drawing inferences, but that, does that ring true? don't they? Yeah, I think so. I mean, I think we're at least getting to that zero point. And certainly, if average wage growth is three, three, there are some workers that are much below that and where real wage growth is negative already. And we may be headed for more people entering that group here as the year goes on. Just as a sidebar, the ECI employment cost is thought to be the most accurate measure of wages because it controls for all kinds of stuff, occupation, industry mix, that kind of thing. Right. That's right. Yeah. I would add to just the CPI, for lower income households that we can look at. It's an approximation based off the different to the CPI basket. The average household spends X percent on something that's different across

the income spectrum. So we look at it for lower income households. Their inflation is hotter too, which would strengthen that argument that they're seeing declines in real wages making it. Do you know what the number is? I know what it was in December and it was about 3.1%. So we'll see what it looks like with updated data in January. We do that by quintiles. Yeah. Got it, got it. Okay. Okay. Matt, do you have a good stat? Do you want to do one?

I got a great stat. I got a great stat. Okay. Okay. Yeah, 13. What is it? 13. 13.

Today's date. That's right. That's it. Good job, Dante. Baker's dozen of very overpriced eggs. 13. That prices have come down. Time something has happened.

that's right that's right in the inflation statistics or i know i couldn't that would be disjointed i had to stick with the jobs market but so it's in the job number jemma yeah yeah something in the job numbers has happened 13 times in history what is it that's what you're that's that's that no that's narrow i mean you're you're you're too narrow but jett or too broad too broad uh yeah too broad number of times employment has declined over the last two three years I don't know what the timeframe would be say that again no because we if I think we have four jobs declined just in 2025 no that's not it okay uh Chris Chris you've been kind of silent yeah I'm off today I don't know go it's a hit Matt the streak is over it was a streak of

13, it's no longer. 13 months. Monthly job gains for a certain sector, certain industry, Matt? Inverse of that. Something was declining for 13 straight months. Oh, is it manufactured? Oh, it's manufacturing.

Yeah. With benchmark revisions, it was... It was a seven or eight month streak, but benchmark revisions pulled industries down, and now you have 13 months of decline from November through December, and now January with the... with a modest gain ended that streak, but it was 13 consecutive monthly declines in factory or manufacturing payrolls.

Wait for revisions. Yeah, right. The street could continue. Yeah, the streak could continue. I think we're going to end the game now because we're going to move on to the inflation statistics. Unless, Chris, you had a great, you kind of looked disappointed. Did you want to give your stat?

No, let's continue. You did I, you're saying, damn him. I've done all this work and now I can't. All this work, yeah. Define the number. Let's go. All right, but let's move on. So, Matt, let's talk about inflation, the Consumer Price Index, CPI.

A lot going on there, too, in the weeds. So you want to give us a rundown? Yeah. January came in as expected. So the first inflation view we got for 2026 was without surprise, 0.2% increase in the headline CPI. That's exactly where we, it's 0.17 unrounded. We were at 0.16, not usually that close, but we were this month.

So, so mild, not so bad inflation, year over year growth gets a lot of attention here because of base effects, but we had a drop from 2.7% to 2.4%, which sounds like a lot of progress, but it was foreseeable given what we knew inflation did last January, which is a big jump that pushed up year ago rate comparisons. that's no longer the case. We got a nice drop there. Behind that weaker growth is a 1.5% decline in energy prices, which again, we saw coming. Wasn't that decline also the result in the year-over-year growth rate because of what happened in the government shut down the October survey, that effect? To an extent, I mean, the decline would have happened and we have our alternative measure that we're

looking at, and that has the same 12-month comparison, even if it's just a little bit higher. So that would have happened to a degree the 2.4% I think overstates things, but it's our alternative measure that takes into account the government shutdown effects went from 2.9 to 2.7. So there is still a sizable reduction. No, no, no. But it's about the actual, but you're saying the top line CPI year over year through the month of January is 2.4. But if it not for the way the BLS handled the lack of data in October when the government was shut down, it would be what, 27? 27, yeah, which I'm speaking about the change, but you're right. So the 2.4 overstates it, but the decline, that reduction is happening regardless of the federal government. It's just a matter from where. Okay, okay, okay. You know, accounting for the measurement issues is 27. Yes, that's still a step down from the 2009. It was last month, but it's still 27, right? That's right. That's right. Okay, okay, fair enough. Okay. And then on core, what was core of the increase, the monthly increase? 0.3%. So a little bit stronger, doesn't have, we're excluding food and energy, so we're not looking at that negative contribution from energy. 0.3% increase in core CPI.

Base effects still exist, just not as dramatic. So 2.6 to 2.5% in the year-over-year rate, we argue we're closer to still in that 27, a little above two points, rounded down, but 2.74 for core CPI. So still high, but again, taking that shift down. Hold on, hold on, hold on. A lot of numbers there. So the month to month increase in core CPA, what was it to the second significant digit? It was right on the nose. Okay. And then you're saying in the data, the year over year growth through January of the core CPI was two and a half percent. Down from 2.6 in December to 2.5 in January. Okay. Two and a half.

And you're saying if you correct for what the BLS adjustment back in October, we're at 27 or 28 somewhere in that ballpark. Yeah. That's right. Okay, fair enough. Still, the rate of inflation year over year in January is down, even counting for the bias created by what happened last October. But the inflation rate as measured by CPI is still probably closer to three than two. That would be the way to put it. Yeah, I would argue that. I would say 0.275 is where we are for both core end CPA. No, 2.75, I'm sorry. 2.75.

That's right. Yeah, yeah. On round. Yeah. All right. Okay, fine. I appreciate the specificity. Okay, got it. Okay, very good. So now, if I look into the bows of the report and different products and services, what's lifting inflation and what's restraining inflation? What components? Lifting inflation has been a pretty consistent sort. Shelter has eased, but it's still a positive contribution, gets a ton of weight. So it's still there, not as dramatic as it was. healthcare, I think, is worth watching more and more. Health medical care services rose solidly again in January. So that's delivering a little bit of upward pressure to inflation, not dramatic, not runaway. It's a slow-moving component.

Some of our tariff-sensitive goods that we look at, so appliances took a big jump in January. In general, I would say that the tariff effect in January was mild, but there are some components that you can point to and say that's a trade-dependent good. There's some pass-through going on. On the negative side, for headline, you got the big drop in energy prices. Vehicle prices. Would be electricity. Did you mention electricity?

I don't think you mentioned. So the big drop is, no, I haven't. The big drop is a function of gasoline and fuel prices, but within, among the components for energy, electricity is one of them. Electricity ticked down a little bit, so it wasn't a big change. 0.1% decline in January. But utility gas keeps climbing, and electricity over the past year really has been on a pretty consistent upward climb.

I think that's something. That's right. That's right. And that's something that I think the best argument is that we're just getting started in a lot of ways. So a lot of that is just demand driven. AIs is the top of mind when you think about what's changing the demand structure for utility companies. So expectation is that's going to continue to rise. That's, you know, acutely felt by households. That's utility bill. That's affordability. That's something that's going to say top of mind. And we're just starting to see even if January's increase wasn't all that strong. Okay. So you mentioned car vehicle prices also seem to be a weight on inflation, both new and used. What's going on there, Matt? I mean, I know every product has its own story. What's the story with new and used vehicle prices? Why are they so? I know part of it's hedonic, you know, measurement of quality that always pushes down measured inflation. But what, is there something going on there? It's perplexing and it has been for a few months. And that's a, you know, we have 15% tariff in the United States on cars, on car parts. That's going to make new vehicles. costs more. People don't want to buy new vehicles, so they flock to use vehicles. So those prices should be rising theoretically, and they just haven't. We're at 0.08% year-over-year growth for new vehicles. We had a slight 0.1% increase in January. So it just hasn't shown up. There's theories why. I think a lot of car manufacturers still think of the tariff negotiations as unfinished. And they're not, look, they think that there's negotiations ongoing. And then I talked to our Japanese analysts, and that's kind of the

prevailing theory that they have is this isn't this isn't over I mean so you're you're talking about high volume low margin cars over there they don't you know price sensitive customers that they're shipping to the United States there's a real reluctance to raise prices I think it's you know surprising that we're now into 2026 and we haven't seen a lot of those cost increases embedded into new vehicles so it's confusing but you know for for people buying cars that's been a positive story and even if you look at motor vehicle insurance now is negative year over year so if we go back the podcast of 2024, 25, that was a big story of how much people's car insurance premiums were rising. And even that as a knock on effect of mild car prices has shown up in a real way, which is great

for consumers, but for forecasting and to think about the effects of tariffs, that's probably been the most confusing. And it's keeping inflation rates month to month changes lower. Got it, got it. So my sense of, I'm just going to throw you throughout my sense of things and just get your your your your uh any pushback or thinking about it is that uh inflation underlying inflation you know and i'm looking broadly beyond the cpi consumer price index other measures of inflation most notably the consumer expenditure deflator uh and by the way i think you probably have an estimate for that as well you know what is that by the way what do you based on cpi and pPI do you have a sense of what the pce the consumer expenditure deflator

inflation will be when that's released? We're still waiting on January's PPI, so we don't have a PCE forecast, but because of the shutdown, we don't even have December's PCE data yet. So our forecast there, which we can put together, is in the 0.34% increase in both those measures. So we're in compositional changes. There's little differences between the CPI and the PCE. There, the year-over-year rate is actually holding up and ticking up at the end of 2025. 2829 and

PC we're at 29 our forecast for December others are 3, 3-1 even so there's a lot more strength just based off of the difference different weights those are measures and the less of the impact of the downward bias from the federal government

shutdown in the PC estimate. Right and the PC of course is the inflation measure the Fed uses for setting its 2% inflation target. They deem that to be the air quotes better measure of inflation And I think one of the differences, the key difference between PCE inflation and CPI is housing the weight of housing in the, because in the CPI, what is it, 40% of the index?

Something like that. Third, yeah. Oh, is it a third? Oh, 40% on the core. Yeah, third of the total. Right. And then what is it in the PCE? It's kind of half that.

Half that. Yeah. Right. And because we're seeing this deceleration in the cost of housing going back to weak rents, that's pushing down the CPI relative to the PCE. and that's one of the things we're observing. Is that correct? Yeah, you're seeing, yeah, what we reference is, you know, the wedge between those measures is growing and it's clearly a function of the weight placed on shelter.

Actually, interestingly, you know, typically CPI runs a little higher than PCE because of, but right now it's, they're basically the same, you know, because of this quirk that we're observing in the cost of housing. Is that right? That is right through the end of 2025. I think that widens, and again, the relationship is different flip than it's normal relative position, but I think that widens when we get December's BC and January's further. Got it. And this is my way of pushing us to this sentence I'm going to say,

and that is inflation writ large, you know, taking all the inflation measures, feels like it's closer to 3% to me than 2%. The 2% is the target, roughly speaking. But we're kind of sort of around three. Would you agree with that sentence, that statement? Yeah, I don't think you could disagree with it. The only way you could make that arithmetic is if you take the 2.4 headline inflation reading at face value. And that's, I think, I'd reasonable, definitely. Right.

So that's the only way somebody could conceivably. I think your 2.8 is probably your underlying inflation right now across all the measures and, you know, taking all the idiosyncratic effects into consideration. Okay, got it. And the other thing that is in my mind's eye, and I'm not sure if I'm right, I haven't checked, but maybe you know, it feels like the rate of, if the overall rate of inflation is close to three, it feels like the rate of inflation for, let's call it a bucket, a basket of, let's call it necessities. You know, that feels like that's on the high side of three. No? Or am I, do you have a sense of that? That's electricity. Homeowners insurance, child care, food, food, housing, things you need, medical care, you know, that stuff. Yeah, I'd say that's, I wouldn't want to overstate that. You're probably right, but I wouldn't overstate that dramatically. Food inflation is 2.1% year every year, grocery store prices down.

I mean, that's... Food inflation, but it's... It is, then it's a huge... It is probably the necessity that people are focusing on, and certainly has the most political salience, but the, yeah, broadly, the essentials, the affordability, I mean, all those things are referencing necessities and those prices are, have risen to an effect. Here's the other thing. Did you see that, going back to the tariffs, you know, because my sense of things is that inflation is closer to three. We want closer to two. And the key difference

is policy is the tariffs and to a lesser degree, the restrictive immigration policy. And did you see that study from the New York Fed that just came out? not showing that 90, 95% of the tariffs have actually passed through to consumers and businesses. Now, they don't make a distinction between consumers and businesses, so it's not to the consumer, but to American consumers. An American company or household is paying the tariff. Did you see that? I did, yeah, which is not a provocative take generally.

I mean, that's the theory about where tariffs end up. Right. Yeah. Yeah. So when I do the arithmetic. again, in my mind's eye, just curious, you know, if you push back on this, is that without the tariffs, if tariffs had stayed where they are a year ago, 2% effective tariff rate, right now we're at 12, but 2% effective tariff rate, and immigration policy was not quite as heavy-handed,

that the inflation would be closer to two, that we'd be back to the Fed's inflation target. Does that, does that resonate with you, that statement? Yeah, I think that's a matter of what we were talking about. in 2024, shelter inflation keeps disinflating, keeps going the trend we thought it was. It has. That's kind of the path that's walked. It's hard to admit. There's, of course, there's other shocks that could have happened. And I would argue, too, that the inflationary effect of tariffs has been maybe 80% of what we expected, but it still is an 80% that's keeping us away from the 2% target that I think we would be at now. And yeah. Okay. Marissa, you heard everything I just said.

You know, I've been leading the witness here, obviously. I'm sure I'm going to make someone out there unhappy about leading the witness. But Matt, I didn't lead you anywhere you didn't want to go, right, just to make that clear. I didn't feel I was being led, but that's my opinion. Okay. Beautiful. But Marissa, does that all, everything that was just articulated there resonate with you, or would you push back on anything? No, I absolutely think that inflation is closer to 3% than 2%. It does seem like there are some components

that are a little puzzling, like the auto vehicle prices. I'm not sure why we're not seeing more inflation in new vehicle prices and then used as a downstream. Yeah. No, I don't think anything that you said was controversial, really. And my view is on tariffs and immigration

policy and its impact on inflation. Your thinking is consistent with that as well? Yeah, I think there's still a lot of evidence that companies are facing tariffs and are deciding how to handle that, right? A big company with wide profit margins is able to keep eating the cost of tariffs before passing it on to consumers. A smaller company maybe that has thinner margins isn't going to be able to do that for as long. So I think it's still happening. It's just being obscured by some of the other things. going on with inflation, like Matt said, the ongoing disinflation and shelter, which is something

we've talked about a lot over the past few years. I mean, that's come down quite substantially. And when it makes up a third of the CPI basket, you know, it's going to overwhelm a lot of the other components in it. Yeah. Chris, what do you think? Anything I said that you took an exception to or would push back on? I'd say the general contours I agree with. I'm a little, I would push back a bit in terms of the degree of how close we would be to 2% given what we're seeing in services inflation. I see. Right?

So services inflation still remains, what, 3.1, 3.2% still high on the high side. And that's not terra-well, presumably that's not overwhelmingly tariff-related. Potentially, it could be some of the immigration effect, but I don't see that as a major factor just yet. So I don't know. I think there are other structural, factors here that may be propping up that

services inflation that could take longer to correct. What do you think? Are you asking me or asking? Yeah, I'm asking you. Push back to your pushback. Yeah, yeah, yeah. No, my pushback to your pushback, I wouldn't argue strongly with that. I mean, my sense of the service price inflation, the sticky goes back to immigration, but I may be overstating the case there. I don't have strong. It's hard to connect those dots.

Dante, you want to weigh in or not, if you want to weigh in, you've got to ask Matt's permission. You're okay with that, Matt? Yeah, I mean, I think I agree with Chris. I mean, I think at a headline, we're closer to three than two. I don't think there's any debate in my mind about that.

I think certainly you would flip that if the tariffs hadn't happened. We'd be closer to two than three, but I don't know that we would be at 2% even despite the tariffs. Okay. Okay, very good. Hey, Matt, before we move on,

anything else you want to bring up that I didn't allow you to say? No, nothing really jumped out in the report one way or the other. Okay. It was pretty much down the strike zone in terms of relative to expectation, right? To our expectations, consensus was a little hotter than us. Yeah. Yeah. And I do think it's quite interesting that PCE inflation is now running at or higher than CPI inflation. Is there ever been a time that we might want to go? That'd be a good stats game. Has that? Has that? ever happened in the past, I'm not sure. Yeah. PCE weights, you know, medical services and health care much higher than in the CPI. So if we have medical services and medical care price inflation that's above average, then that could go some way to explaining that, right?

Yeah. I guess we could go back in the 70s and 80s when medical care inflation was a lot higher. Maybe we had a similar situation with the PC and the CPI. but be curious to go take a look. Okay. Mercy, you said there was a listener question on inflation, too. Did you want to throw that? Yeah, there's actually, I had one,

and then I was looking back at other ones, and there's actually a really relevant one. So there's two good ones. Okay, so let me pull it up here. Okay, so this goes to the missing data in October. So this listener said he was, talking to his wife about it, and as they were talking through it, they couldn't understand why that matters because they were saying if the CPI basket prices are empirically measured, then the missing

data should be irrelevant. To take a simple example, if a candy bar costs a dollar in September and a dollar five in November, we know inflation was 5% over that period, regardless of what the cost was in October. If you have time next week on the show, if you could explain why this matters, that October is missing, that would be great. I was thinking maybe it has to do with parts of the CPI that aren't easily observed, like OER, but that's just a guess. Matt, you want to take a crack at that? One, I wish my wife and I would talk about CPI components more than we do.

This guy, he doesn't know how good he has it, this guy. Second, it's a great observation, especially in November, data came out. Can we just observe the prices over the two-month gap? And isn't that going to capture whatever inflation did occur in October? That is the case for a lot of components, but shelter, as the listener alluded to, is not calculated that way. So in October, we're not looking at what rent change looked like from September to October and saying, here's OER and here's rental prices and all that.

Instead, what's happening is they're looking six months prior. So by saying that change, and because we didn't have any date in October, the BLS is looking at April and saying, okay, we don't have October data. Let's just say nothing happened. So now you have the zero there that really had nothing to do with September. And because, as we mentioned, shelter is 33% of the CPI, 40% core CPI. That zero, I mean, normally it's whatever that changes to the sixth route and they go, okay, that's the percent change over, you know, because we're looking at a six-month window. Plugging in a zero there ends up weighing, just kind of shifting the index lower in a way that you wouldn't be able to.

to just jump from one month to month three and be able to observe that price. So it really comes down to the way that shelter inflation is calculated on that six-month panel that can't be addressed as the way a candy bar could be. Excellent. Excellent. Any other question? You said there was another one, Marissa, or there was a... Yeah, there was another one about tariffs effect on inflation, which is also relevant given our discussion. So this listener was noting a Wall Street journal article where it was talking about different company earnings, and it was saying Estee Lauder had changed their earnings outlook, and they're particularly, they're taking $100 million hit to profit over the last quarter because of tariffs. And he was saying,

I keep seeing articles like this with companies saying that they're facing all this headwind from tariffs. So how do we get to this argument that a lot of people, people, including Jerome Powell, make that tariffs are a one and done kind of thing when this seems to be an ongoing thing that companies are calling out every month. So I think the distinction, well, I'm sorry, I didn't mean to cut you off, but I think the distinction is the tariffs are obviously in effect every month, right? But I think the one and done is a reference to companies choosing whether or not to raise consumer prices. That's my take on that argument. And that that may be a one-time adjustment instead of ratcheting up prices over a period of time. Do you think that that's

the way to interpret that? I mean, you say that kind of in a loose way. I mean, there might be a period of time over which companies will adjust their pricing in the context of the tariffs. So it might not be all in one month. It may have a period of time. But once they adjust to the tariffs, that's the end of it. At least that's the idea. I mean, there's concern that there might be some more persistence in pricing, you know, because you raise prices that might affect inflation expectations and gets into wages and prices more broadly. But that doesn't feel like that's what's happening here. It feels like it is going to be more one and done. One in the sense, though, that one adjusts, you have this adjustment to the tariffs, and then you're done with

the price increases related to them, I think. But it's more of a, one and done doesn't feel like it's, makes it sound like it's an event. This is a process, not an event. Right. Yeah, that extends, not for one month, but can extend for a period. It may extend longer this go around that we've talked about in the past. One of the reasons why we haven't seen the same pass through at the same pace, is because the terrorists are all over the place. You know, they're on again, they're off again, there's exemptions, there's carve-outs. So when businesses see that,

they don't pass through as quickly because they just don't know where the terrorists are going to be. And then there's also concerns about getting caught up in the political maelstrom, you know, if they raise prices too quickly, so they're going to raise them, you know, ultimately they'll pass through, but they're just going to take their time doing it

so they don't, you know, they're not caught up in, you know, in the political debate. I think that that's what's going on. Does that sound right, Matt, what I just said? Yeah, and I think too, it's easy to think of like, okay, April 3rd, whatever Liberation Day was, that that was the quarter, that the following quarter, all these prices were going to rise. It matters the good we're talking about, how big is your margin. Can you absorb it if you're a luxury car manufacturer or if you're cosmetic company?

That's going to be a different time horizon that you're able to absorb prices or not absorb prices. So one time is messy, but it's a couple, you know, we're still talking a few quarters ago, and that's happening now, but it's not the perpetual inflation that you get in a kind of normal business cycle that would concern the Fed. And that's why they take in a stance they have. Got it. Okay. Well, guys, you know how much time we've been at this? One hour and 10 minutes.

It's a law of podcasting inside economics, an hour and 10 minutes. I don't know how that happens, but that's, that's the, that's the hour and 10 minutes. the way it works out. So before we call it a podcast, anything else anybody want to say? I do think, I did want to mention, you know, we had a webinar on the Outlook, Marissa, Chris and I back a couple weeks ago. We took the transcript along with the PowerPoint deck and gave it, Chris did this to Claude. We gave it our style guide, said, put it in the voice of Zandi, and we produced a document that kind of summarizes things. And I think we're going to just put it out there in LinkedIn

and maybe some other venues and get your feedback where you think this is a useful thing or not. You know, we read it. Chris, what do you think? I thought it was pretty good, right? Pretty good. It's like an annotated transcript. Let's call that.

Entated transcript. Yeah, so we want to get your feedback. So we're going to put that out there. I think we're going to put that out there. I hope I'm not jumping the gun. But we like to get your feedback on that when we do, just to see if you think that's a useful thing or not. I feel a little weird about it, but, you know, maybe I got to get used to it.

I got to get over it, you know. Brave new world, Mark. Yeah. What do you think, Dante, is that weird you out, too? I mean, yeah, a little bit, but I think there's utility there. I think it's useful. Like, you know, you got to balance this too. Okay, Matt's the youngest in the group, but are you weirded out by that or not?

Not weird it out, but I don't know. I'm not sure. I'm not sure. Yeah. I like my own voice and my own writing. But maybe that says more about me. I know. At what point do I lose my voice and I become AI? You start trying to sound like what you think AI sounds like. As soon as that goes up on LinkedIn. Maybe it's already happened. Maybe it's already happened, right? Who knows? No, no. Every. everything I've written to date is Zandy. You know, old Zandi. I haven't even put it into an LLM, and maybe I should. I don't know.

Anyway, with that kind of metaphysical discussion, we are going to call this a podcast, dear listener. I hope you enjoyed it, and we will talk to you next week. Take care now.