Is the housing market finally flagging a recession?

HousingWire

On today’s episode, Editor in Chief Sarah Wheeler talks with Lead Analyst Logan Mohtashami about mortgage rates, residential construction labor and Trump’s firing of the head of the Bureau of Labor Statistics.   Related to this episode:   ⁠How much lower can mortgage rates go after Friday’s jobs report? | HousingWire⁠  https://www.housingwire.com/articles/how-much-lower-can-mortgage-rates-go-after-fridays-jobs-report/  ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠  The HousingWire Daily podcast brings the full picture of the most compelling stories in the housing market reported across HousingWire. Each morning, listen to editor in chief Sarah Wheeler talk to leading industry voices and get a deeper look behind the scenes of the top mortgage and real estate stories. Hosted and produced by the HousingWire Content Studio.

Transcript

00:00Welcome, everyone. My guest today is lead analyst Logan Motoshami to talk about mortgage rates,

00:12residential construction labor, and Trump's firing of the head of the BLS. So much news to

00:17cover. But first, I want to thank our sponsor Optimal Blue for making this episode possible.

00:23Logan, welcome back to the podcast. It is wonderful to be here. I have to say

00:28this was one of the most enjoyable weekends I've had being a nerd. First of all, I just want to say

00:37that I am so happy that maybe one of my friends, Neil Dutta, who's one of the top Wall Street

00:45economists, there are whispers of him possibly filling in the role for Adriana Koger, who's

00:53going to resign in a few days. She was going to resign anyway in January, but they've got to fill

01:00in that spot. I cannot stress to you, that would be the best governor pick I've ever seen.

01:07Neil's like the chart daddy on the East Coast side, and he's very fluent in terms of talking to his

01:17people about how economic cycle works, and he's very objective. And that's what you need. That's

01:23always been, you know, that's to me is what the Fed needs, a kind of a market-oriented data person

01:29out there more than an academic. And I think he would just be the perfect fit. So I am crossing

01:36my fingers. You know, Neil's a little bit more conservative than I am, so that should maybe go

01:41to his favor. But yeah, that would be a very good thing for the housing market if he's there.

01:49Well, okay, speaking of the housing market, is the housing market finally flagging a recession? And

01:55what I mean by that is, you know, residential construction labor, because you know, you've

02:01pointed out for a couple of years, this is what to look for if we're looking for a recession.

02:05So this was also part of the big topic over the weekend, and it was so much fun to talk about this.

02:14Now, for those that are new here, one of the things that we've said, we believe in models,

02:20and we believe in teaching people how cycles work. And there is no yes, no answer. There is always a

02:28model to give you directions. I believe in a progression economic cycle, recessions, expansions.

02:34I never stop. It guys goes in and out. That's it. It constantly goes. There is no days off.

02:40But one of the reasons why a lot of recession people have been wrong since 2010, outside of not

02:46being versed on macroeconomic trends, they don't have a coherent model to go with them. So they assume

02:53a lot of things. A lot of people do it for attention. But in this case, when we went into 2025,

03:01we said in December of 2023, the home builders are going to be in a supply and demand equilibrium

03:08problem, right? That was the article we wrote. And the reason I use this year instead of the

03:14past few years is that the completed units of sale, this is the key data line. Not a lot of people know

03:20this. They look at the monthly supply and they think that's the actual active inventory. That's not true.

03:27This is, you know, one of the recession theories in 2022 was that monthly supply is high, but then

03:33you have to break the inventory channels into their categories, units completed in construction or

03:40hasn't started construction. Well, the units completed were growing to a level that in 2025 is usually the

03:49peak of the economic cycle. So if rates didn't fall, this supply and demand equilibrium can break.

03:56So now we have that in play, which wasn't the case for a very long time. So here we are. Housing permits

04:03have been in a recession for some time. Housing starts have been in a recession for some time.

04:08Completed units are now up to levels that usually that's the end of the cycle for them.

04:14But the missing piece was always the construction labor. So we have two charts that we show

04:19always. There's the specially contract data line, and then there's the residential construction

04:25workers, which remodeling single family and apartments. They are now both fading together.

04:34So we always want to believe it. We want to teach economics so people don't get suckered by

04:40crazy men on the internet. So point blank, connect the dots, be the detective, not the troll. Here is

04:46a data line that's starting to flag a recession. Now, a lot of people said, well, is four months

04:52enough? And I said, no, and I'm going to explain why. And let me ask you, Sarah, why would you think

04:58four months isn't enough? Seasonality in construction is one reason.

05:04Well, here's the thing. We actually had four months of decline back in January of 2023.

05:13And the declines back then were actually a little bit more than what we saw in the last four months.

05:18But back in 2023, mortgage rates were at 6%. The builder's confidence data started to pick up.

05:26And then pretty much that kills that story. So you're going to have permits grow. You're going

05:31to keep the employment. There's a lot of backlog. So there you go. So the backlog is still there.

05:36So the builders are trying to hold on to as much as labor. If rates fall down, then they can do this.

05:42But now you have both coming down together. I shortened the duration of these long charts

05:51that I usually show and showed it just for one year. It's a very, very mild decline right now.

05:57But the difference now is that the units completed are up. The specialty trade are always falling.

06:04And this is a very early trigger flag for a recession, which I really have not been able

06:10to say with all of these components put in together, right? Also, back in 2022, when single

06:17family permits were falling very aggressively, multifamily permits were rising. So there was

06:22an offsetting effect, which never happens historically. This is why I always say this housing cycle is

06:27the most unique. If you are a housing nerd, like an authentic, non-trolling, look at my YouTube

06:33page nerd. This was as glorious as you can get. So now we have to talk about hypothetically, let's say

06:41mortgage rates go up to seven and a half percent. I know it's not part of the question. And the

06:48builders just, their profit margins get worse and worse. There you can see a more fluid downtrend

06:55in the labor market there. And then that filters itself into the jobless claims data, which is still

07:00very low. And then as the jobless claims data rises up, the question is, does the deficit spending that

07:08we're doing in America now, because we have a lot of older people right now, so mandatory payouts,

07:12social security, Medicare, all these things are very heavy on the budget side. But AI spending,

07:20right? We're almost at the point to where AI data center construction is actually going to surpass

07:27what we usually construct in non-residential construction data, which is crazy. It's like,

07:35right when ChatGBT was created, AI data centers just went, who knows?

07:40We actually, John Delosier wrote a feature article for us on the Arizona housing market where they

07:48have closed down building of single-family homes in this one area, but they don't have their same

07:55restrictions on a data center. And the data center is going to take up as much water as 100,000 single

08:00family homes. And there's no restrictions on the data center, but they can't build a single-family

08:05house there. You know, when it's all said and done, the amount of energy and resources AI is going to

08:14need to make this work, it still does not get enough attention, but it's a lot of electricity.

08:22Electricity bills are going to start to go higher, but the amount of resources needed to not only get

08:31this going, but to keep it going for this amount is going to be historic. So yeah, it's just one of

08:37these things, but the AI construction data is actually, the data centers are about to surpass.

08:43So there is one form of construction demand that's non-residential that is very unique to this cycle.

08:49And then, I mean, I can't even, we are spending hundreds and billions of dollars on AI that's going

08:55into the economy that would not have normally been here. So it's a very unique cycle in that sense.

09:01That the deficits are very elevated for an expansion. And also, we do have the AI factor.

09:08So when you're in a normal cycle, when you're looking at that residential construction labor,

09:12are you looking for a certain timeframe? Like four months is enough, you're looking for six months,

09:17or are you looking for a certain level of drop-off?

09:19For me, I'm looking, I'm looking for a certain level in that data line to crack down with a little

09:26bit more velocity and a little bit more duration. You know, when you have four months of a very small

09:32decline, that's not, that's not big enough, right? You need to have this string out and be a little

09:37bit more forceful. But then you also have to put, is construction labor going down in general?

09:44So I don't believe the AI factor is going to be, you know, totally offsetting everything,

09:50but that is a variable that is unique in the cycle that you have to adjust your models to. And,

09:56you know, we go into the money that's being poured into the economy through AI. So it's very unique,

10:02but we're, we're finally in the, what we call the slope of the curve going down, right? We always

10:07talk about our slope of our curve and our tracker data, you know, and, and I always explain to people

10:12what I share to all of you on the weekend tracker is just a glimpse of the general macro data because

10:18I track economic cycle first. Our really good stuff is states and cities and zip codes in a more

10:26prolific fashion. We're not what we call third party regurgitators. We're not people

10:32taking Redfin's data or realtor.com's or Zillow's data and just, you know, putting it out there.

10:38We have our own data line set. I show you what the tracker is on a national basis. Mike does that

10:44as well. And then you get to the internal sides and it becomes a much more prolific data line set.

10:52Like I can't do this with every city in America. So I give you the broad picture, but it becomes sharper

10:57out there, but the slope of the curve. And the reason I say this is because Redfin,

11:03right? Remember Redfin seller buyer, it's really confusing everyone now. Cause now that seller

11:08buyer gap has got worse, but their median prices keep on hitting all time highs and it's firming up

11:15in the last. So people are just like, what's going on here? Something's wrong. I said, no,

11:20I understand what Redfin is doing with that. It's just worded in a way that's really confusing to a lot

11:26of people. So I went on Twitter today and made a video and did a chart. And I said,

11:30look at our slope of our curve and our data line. If you look at Redfin's data, I mean, in 2022,

11:38we actually saw seller buyer stress that came with the fastest crash in home sales ever.

11:43Prices were authentically falling in the second half of 2022, not the normal median sales price

11:49declines that we see. You look at Redfin's data, there's nothing going on there.

11:53Like, how did that happen? Like here. And then I go, look at the pricing curve of the second half

12:00in 2022 versus Redfin's data versus what's happening right now. And everyone's like,

12:06wait a second. So we are big believers of slope of the curve of economics. And now, you know,

12:12of course the tracker has a ton of, a ton of data lines, but the slope of the curve in the residential

12:17labor data is now curving lower. So when you have that happen, that usually is a peak of the

12:24economic cycle. So now we have to be mindful of, do rates come down low enough, fast enough to change

12:30this? Because one of the things I've highlighted is that when we get to 6%, CBS, Nightly News, we did an

12:37interview with them last night. And I was trying to tell them that, you know, higher inventory is great

12:42for the existing home sales market, you know? And they were saying, well, for sellers, isn't great.

12:47Oh, trust me. It's great because sellers are going to get a reality check and they need one.

12:51Okay. But that seller, most of the time is going to be a buyer. Well, the builders, on the other hand,

12:56they don't have that seller buyer thing. They sell homes as a commodity, right? So to them is that all

13:02they care about is to get a buyer in. And every time mortgage rates get down towards 6%,

13:07the builder's confidence data in the last few years picks up. And that's the smaller builders

13:12that don't have the profit margin. So the question is, can rates get down to 6% fast enough to change

13:19this slope of the curve? Oh, see how beautiful this is? You love slope of the curve. You're always

13:24doing that. Yes. I'm in heaven. What if I could just do this my whole life, just doing slope of the

13:29curves. It's great. But here now it's like, okay, can we get here fast enough to have the builder's

13:34confidence and then permits rise? That's the race right now out there. And that is something I know.

13:41Neil Dutta, team Neil Dutta, right? Hashtag not Kevin Warsh. I'm going to be so mad at it. See,

13:47Kevin Warsh to me would probably be the number one pick to fill that void in. And I'm just like,

13:52no, Neil will be so much better. We can listen, we can do merch. We can get like shirts and stuff

13:57and just, uh, who's our pick? I know, chart daddy, east coast, chart daddy, west coast. I think that'll

14:01work. You know, um, we got to have Neil on, on the show, uh, one time, but, uh, in this case,

14:09uh, uh, it's an interesting dynamic to go forward, right? And how we use, how we take it is we take

14:14all the economic cycles in the past and we, we do the curves and we do the permits and the starts and

14:19the completions and all that stuff. And when you put them together, it's like, whoa, people get it.

14:24And that's, that was a good part of this weekend that when you show people the correct charts and they

14:29connect the dots there, that housing, uh, that economic equilibrium theory that, you know, has

14:34been here for 2000 years, you know, you could connect it to where people are like, oh, okay,

14:39that's, yeah, there you go. It was so beautiful to watch people actually go, oh, I get it now. This

14:44is yes. There we go. That is good. My daddy is happy, right? I have so many questions. Yes. I have so

14:53many questions. So one of them is about the recession, but I'm going to hold off on that about a

14:58possible recession, um, and ask you and follow up on like, could we get rates down low enough to

15:03change the slope of the curve? Because last week rates fell, right? I mean, was it a 0.18%? Was it

15:0920? It was almost 20 basis points move, uh, in, in one week. So we're, we're, we, we finally got to

15:16the level to where you start to see, you know, if you go from 6.64 down towards six, that's when I've

15:25seen the data actually authentically get better. And then it was just like, like trying to explain

15:30to CBS news about purchase application data. It's like, you know, it's a funky survey and

15:36their year over year growth, but the weeklies aren't that strong. So what's happened in the

15:41previous times where rates get down to 6%, the weekly data is really pick up and it really picks

15:47up from a negative trend. It's like last year is a really good example. Uh, last year, a lot of

15:52people, uh, didn't know that, you know, we had like an 18 week positive week to week trend

15:57because they saw no year over year growth. So they didn't think anything was happening.

16:01It ended up being a couple hundred thousand home sales. Uh, we ended up taking new home

16:06sales to the, uh, uh, you know, the, the, one of the highest prints of the cycle, you

16:12know, outside of, uh, after they crashed. So it works. Uh, but I've only seen it really

16:18from 6.64 to six. And that's why we don't throw up the 6% number out of just random things.

16:23We have to show why we don't, we have to create a, a model said this happened, this happened,

16:28that happened, all this worked with this rate range out there. Uh, uh, so that's, it's worked

16:34before the question is now going out. And this is why I, I really want Neil as part of the

16:41fed. Uh, the federal reserve has ignored the housing market for three years. Like we've always

16:46said, they're not going to care. They absolutely do not care that existing home sales are here.

16:51They do not care that housing permits are a recession. They just, they give some token

16:55answer that, well, even if, you know, after this is all done and rates fall where the supply is

17:00going to be okay, whatever, that's not it. The economics of it is different. Uh, so it,

17:05it is interesting that they do have the federal reserve. If they wanted to, if they finally believe

17:10they needed to pull, they do have one sector of the economy that has not enjoyed the expansion

17:16on its data line and they are depressed. Now the builders of course have outperformed the

17:21existing home sales market by a long shot, just because they live in a sub 6% world, but their

17:25corporate profits are starting to get hit a little bit hard. So it's, it's fascinating. We're finally

17:31here. Right. And you got to remember, this is like since 2010, I've not been able to have this

17:35discussion for a long time, right? COVID was an anomaly factor, right? You know, and even with

17:41COVID, when we wrote the COVID-19 recovery model on April 7th, 2020, we use tangible variables of

17:48like economic output was so crushed just because everyone stopped that this was going to be a

17:53V-shaped recovery when everybody's like, you know, I'm alive, I'm working. We plugged the holes with,

18:00uh, uh, checks, you know, so we're going to have a full force and there we go.

18:05Well, here's going to be a little bit different because, you know, employment levels, the total

18:09employment levels are still high, but the one sector that actually can benefit from rates can

18:14maybe boost up a little bit. I, is it going to be enough, uh, out there to keep the labor market

18:20in it? We will see, but it's just, it's a, it's finally here. It's a fascinating experience. So you

18:26never, you never knew me in the last decade, man. Last decade was awesome. We had the longest

18:30economic and job expansion ever again. Guess who? The anti-central bank movement of the society.

18:36Angry old men were always talking about a recession every single day. They didn't have

18:39proper models, right? If you don't have proper models, you're just throwing stuff up.

18:43So I know that you're very careful about recession. So you published like, you know,

18:48I think you raised your fifth red flag for the recession. You have six, uh, red flags. You

18:52actually presented this to the conference board, right? That was the most nerdiest. Oh my God.

18:57If you, I could have died on that day. No, the conference, that was, um, I forgot, was it July?

19:04It was July 22nd, 2022. Yeah. You know, the conference board and the conference board,

19:09the conference board for those who people, the conference board is the one who created the IMF,

19:13the world bank, Ethan Hutt, mission of possible. They also, uh, give their information to the federal

19:18reserve and the white house teams, but we created a six recession red flag model. When all six

19:23recession flags are up, this is where we want to look at for those labor triggers, jobless claims

19:27and, uh, uh, uh, residential construction workers, August 5th. And then November 9th came the 10-year

19:34yield right in October, 27th. We said the next move in the 10-year yield should be lower. When it

19:38happens, let's look at the housing data. Permit starts to rise, builders confidence starts to rise.

19:44The builder stocks actually had a, you know, the builder stocks have still outperformed

19:49the S&P. You know, people are saying, oh my God, the builder stocks are crashing to it. No,

19:54they're not. Actually, you do it. If you compare to the S&P, like Toll Brothers,

19:57they're still on a percentage basis. But that was November 9th, 2022. What we harp was a very

20:03key point. Rates went down, builders confidence picked up, permits picked up. So that took away that.

20:08Now it's a little bit different. Permits have been in recession for some time.

20:11times starts have been in recession. Uh, they're holding onto as much labor as they possibly can.

20:16Cause if they believe if they could get rates to 6%, they could start moving this thing. So

20:20it's, it's getting that equilibrium. Sarah, it's so fascinating. I love it. You know,

20:26there's the last cycle was so boring. You know, this one is so much fun.

20:30It was so boring. Okay. I have, uh, the last question I want to ask you is on our podcast,

20:35our last podcast, which came out on Saturday, it was really fun. Um, we talked about the,

20:40uh, labor data and what the risk, uh, revisions meant. And because I said, you know, there's a lot

20:46of talk about, um, sort of a conspiracy theory about the revisions and like that someone's cooking

20:52the data. And then what happens after our podcast, after we, uh, record it, Trump fires the head of

20:58BLS. And I know you've got a ton of questions about that this weekend. So I wanted to ask you,

21:03what does that mean? What, what do you think as an economist, who's like relying on that data?

21:09What do you think? So we have different data line sets to get an idea of what's going on around that

21:19isn't just tied to the BLS number. Uh, of course there's private sector data and there's corporate

21:25profit data. So we're not totally in the dark. I, Trump is just being Trump, right? You know,

21:32it's like, remember the whole hurricane thing he did with a marker. He's like, Oh, the hurricane

21:36was here. And then he marketed out. Oh, no, the hurricane's here. Trump has given us some awesome

21:42stuff over the years. In any case, I want to see who is taking that job. But if there's any crazy new

21:50divergence in the data, there are ways that, you know, all data miners can check to see if it's,

21:57if it's valid. I think it'd be a very hard stretch to say that they see a negative number and they

22:03erase it and they add something because, you know, you're, you're really going into banana Republic land

22:09if you're like erasing data and putting other numbers in there. So I will reserve my judgment

22:15until I see a whole data line set to actually come out and see if there's anything crazy about that.

22:21But I mean, part of the problem is still the, the, the surveys are not getting as enough responses

22:28and we're kind of defunding the BLS to a degree. So I'm going to reserve judgment to see who's in

22:37and what does the data line look like before anything. But I mean, personally, I would say,

22:41Hey, listen, let's get some of that tariff money back into the BLS. You know, let's find ways to get

22:47the surveys responses to grow. And then I think everyone benefits in that regard.

22:53I think it's important to clarify that it's not one person who, who does that data analysis or comes

23:00up with that number, right? There's a, there's a whole group of like 40 analysts who look at the

23:05data and then they deliver that result to that person who. Yeah. I mean, you, you get a survey response

23:11and with that survey responses, you know, it's a team working there and they come up with their best

23:16number they possibly can. And the survey response. Now, if you actually take the aggregate data going

23:21back decades, we we've actually done a better job of actually getting these numbers in. But,

23:27but the survey responses have been coming in a little bit lower. So to me is, you know,

23:33if I, if I see that it's really apparent that, you know, they're actually authentically fudging the

23:40numbers now, you know, that, you know, we, we, we, we lost 25,000 jobs and they say we've gained 250.

23:47There are other ways to, you know, it's very hard to say, Hey, listen, we're creating jobs and then

23:52jobless claims is rising higher and higher. Cause we have multiple data line sets that we track here.

23:57So if there's one that's really odd from each other, like we do bar charts with, you know, the, um,

24:03ADP number versus the BLS number. And, you know, there's some, there's some wide variances,

24:09but there isn't like six months of ADP losing jobs and six months of BLS, you know, uh, rising jobs.

24:16So there, there are ways for everyone to kind of keep sure, make sure everything is kind of still

24:20legit. But I understand a lot of people are concerned that, you know, he's going to put in

24:24someone and they're just going to fudge the numbers up. And then that, that doesn't do anybody any

24:29good. Me, I would just, I would fund the BLS more. I'd get more people. I'd find ways to get,

24:33uh, more efficient survey data, uh, out there. And, you know, I mean, we are like the, the,

24:40the cream of the crop in terms of economic data out there. Like this isn't like China and other

24:45places where nothing is not, nobody believes anything. And we have multiple data line sets

24:50to track here. So it's not just one. All right. Well, we'll be keeping our eye out, uh, because

24:56that data is very important for everybody who's listening actually. So Logan, thank you so much

25:00for being on. We will talk again very soon.

Category

Transcript

Be the first to comment

Recommended