Archive
They’re Starting to Come Around on Rent Inflation
For a couple of years, I have been relentlessly defending my forward inflation forecasts against a sizeable group of people who looked at various high-frequency rent indicators and concluded that rents were going to be imminently in deflation. (For most of the last year many of those same people thought tariffs would be a large and immediate effect increasing inflation. Fortunately for them, being wrong on both counts, at least the errors offset somewhat.)
This battle began in early 2023, shortly after the publication of new indices by the Federal Reserve Bank of Cleveland, supported by a paper entitled “Disentangling Rent Index Differences: Data, Methods, and Scope” by Adams, Lowenstein, and Verbrugge. Those authors parsed the BLS rent microdata to separate out the new tenants, and created a “New Tenant Repeat Rent” (NTRR) Index that supposedly served as a leading indicator of what all rents were going to do. Naturally, NTRR had peaked early and was heading down sharply, which reinforced the observation from things like Zillow, Apartment list, etc that new rents in the aftermath of the post-eviction-moratorium catch-up were declining.[1]
The San Francisco Fed also published a piece in mid-2023, entitled “Where is Shelter Inflation Headed,” by Kmetz, Louis, and Mondragon. Don’t get me wrong, I love it when people try to create better models of inflation processes. But this was another one that made just terrible forecasts, because (as in the former case) it was put together by econometricians who didn’t understand the actual underlying process and thought they could just torture the truth out of the data. They included this wonderful (and subsequently damning, because the Internet remembers everything) chart.
Accompanying that chart was the helpful clarifying statement, in case you didn’t get the import: “Our baseline forecast suggests that year-over-year shelter inflation will continue to slow through late 2024 and may even turn negative by mid-2024.”
In case you were curious, it didn’t turn negative; in mid-2024 it was a bit above 5%.
So back then is when I had to start defending a fairly simple premise: the behavior of landlords when they offer rents to new renters does not necessarily mirror what they offer to renewing renters. In fact, I could be even more strident – landlords could not offer lower rents to everyone, even if they offered them to new renters. That’s because a landlord needs to cover his costs or he won’t be a landlord for long. And in 2023, the costs for a landlord were still rising very rapidly – labor, energy, insurance, taxes, maintenance, and so on. My model – first presented in Enduring Investments’ Quarterly Inflation Outlook in August 2023 – suggested that rents were going to decelerate, but much more slowly than others were forecasting. I had them as low as 3% by mid-2024 before flattening out, and even that turned out to be too aggressive on the disinflation side.
By now, regular readers are familiar with this model and familiar with the fact that it still is calling for Rent of Primary Residence to hang around the current 3% level for quite a while yet. Want ‘em lower? Lower landlord costs.
But this article isn’t meant (only) to pat myself on the back. I also want to recognize when someone gets it right and the great inflation analysts at Barclays recently published an article entitled “Apples and oranges in the CPI basket: Why market rent gauges mislead on shelter,” by Millar, Sriram, Giannoni, and Johanson. It is marvelous article, and you have access to Barclays Live and care about this topic you should read it. While they don’t build a cost-plus model like I did, they got to many of the core reasons why looking at new-renter indices is bound to be misleading. My favorite charts from the piece are below (I also had these in my recent CPI report).
What my model does is tell you why that had to be the case: landlords can’t just lower rents on their whole renter base if their costs are increasing. The only exception to that would be if there had been significant overbuilding such that there was a surplus of apartments over the demand from renters. In some places, especially those currently experiencing a negative immigration shock, that may be the case (although those places happen to also be the ones experiencing large increases in insurance costs, so it’s not quite that easy). But nationwide, there is not a surfeit of apartments for rent. Ergo, no rent deflation. And it’s going to stay that way for a while.
One final note here, about the recent Trump announcement that the Administration desires less institutional ownership of single family homes and apartments. I say ‘desires,’ even though that isn’t how it was phrased, since there appears to be no obvious way that the Administration can force this. They are reportedly looking into whether antitrust regulations can be used to keep institutions from accumulating very large portfolios of shelter units, but this looks like (at best) a task for the legislature, not the executive. But let’s consider quickly what the effect would be if Trump got his way in this regard.[2] Institutions which own homes and apartments don’t hold them off the market. That would be terrible carry. They rent them, just as landlords do. If you forced institutions to divest single-family homes, it would simply move supply from the rental market to the owned-home market. That would probably drive home prices a little lower, relative to the prior baseline, but increase rent growth at the margin. This doesn’t seem productive!
[1] I talked about NTRR in a July 2023 episode of my podcast: Ep.74: Inflation Folk Remedies
[2] Honestly, I don’t think he really means to do this. Some amount of what the President says – especially the impossible things – are intended for consumption by voters. I could be wrong on this. Mr. Trump does have a way of making things happen that didn’t seem possible initially, but in this case there’s probably not much he can do and anyway it wouldn’t have a big impact anyway.
Mamdani’s Effect on the CPI
Surprising no one, and yet shocking many, avowed socialist Zohran Mamdani won the election yesterday to become Mayor of the largest city in the United States.[1]
Probably the main reason for Mamdani’s victory is that he pursued the tried-and-true method of giving out free stuff, and a whole generation of Americans who have systematically been poorly educated in history and economics said “that sounds awesome.” So, now we will see whether socialism will work for the first time ever.
This is an inflation blog, so I want to review briefly the effects of price controls on inflation – and indirectly, on inflation instruments. It’s interesting because we actually have some direct and recent experience with what were effectively price controls: the Biden Administration’s ‘eviction moratorium’ during COVID, that prevented landlords from tossing out renters who weren’t paying their rent. Really, it’s a pretty amazing thing that says a lot about Americans that the vast majority of renters continued to pay rent anyway.[2] An ancillary effect, though, was that landlords had no leverage to raise rents and therefore, rents stopped going up. Unsurprisingly (and here is where the lesson should have been learned), when the eviction moratorium was lifted rents re-accelerated. In the chart below, note how in 2021 effective rents declined while asking rents went up – but the red line eventually rebounded and exceeded the prior trend.
I actually haven’t looked at that chart in a little while. It’s fascinating to me that ‘asking rents’ (which come from the Census department) have maintained their divergence from ‘effective rents’ (sourced from Reis Inc). I wonder if some of that is the effect of the LA wildfires. In any case, not today’s article. The point is that the effective price controls on rents did have an effect on measured rents, but it didn’t change the economics and eventually prices caught up.
Back in 2022, I produced an excellent podcast episode entitled Ep. 37: Bad Idea of the Year – Wage and Price Controls. In it, I discussed some of the trial balloons that had been floated by the Administration and some of the really bad economics that was being used to support the idea. This is a part of the transcript (from Turboscribe.ai), and I still love the analogy:
“But the basics of how it works are very simple to visualize. Price is a teeter-totter, okay? It’s a seesaw. On one side of the seesaw sits all of the buyers. On the other side sits all of the sellers. If there are lots more buyers jumping onto one side, then the teeter-totter drops on that side, and the fulcrum, in order to make everything balance, the fulcrum has to move. And if you move the fulcrum, then you can get that to balance even with more buyers and fewer sellers.
It just means that the fulcrum, which is price, has to move in one direction. If then people, those buyers drop off, then the fulcrum moves back the other direction. If more sellers jump onto the teeter-totter, the fulcrum moves the other direction as well.
So it’s a simple way to visualize it…and yes, there are all kinds of complexities in the real world. There’s behavioral, there’s stickiness that happens, but that’s the fundamental theory of price, is what I’ve just given you, is that price is the fulcrum that balances the buyers and sellers.
So what price controls say is that, well, we don’t like where this balanced. We have too many buyers, not enough sellers, and the fulcrum has moved way over to one side and we don’t think it should be there. So we’re going to take the fulcrum and we’re going to move it to where we like it. And guess what happens? There’s no balance. All of a sudden, if you move the fulcrum away, then all of a sudden, the side with all the buyers goes down and goes thunk on the ground. There’s no balance.
“How do you then balance it? If you say that the fulcrum has to be in this location, how do you balance the teeter-totter? Well, you have to take buyers away. And you take buyers away by making a shortage. And so those buyers can’t buy anything. And then voila. So if you force the price, then the quantity has to change. And if you let both things happen, then it will magically go and balance. If it’s truly a free market and there’s good information and all that stuff.
“So does this solve the problem to push the fulcrum to one side and say, oh, there’s no inflation and to make it balanced, we shove everybody off the teeter-totter by creating a shortage? It doesn’t solve the problem. And furthermore, the people that you’ve pushed off the teeter-totter who can’t get access to the thing anymore are pretty upset. They should be upset because before they had a way to get what they wanted and what they were willing to pay for. And now they can’t because you’ve shoved them off the teeter-totter. You’ve created a shortage.”
That was a public service announcement, just to remind you why price controls don’t work. That doesn’t mean they aren’t really good politics, especially if you can leave the removal of the controls to the next guy who ‘causes’ the inflation when they come off. And it’s the politics, not the economics, that leads to this dumb idea being tried over and over despite a roughly 0% record of success.[3]
Because can price controls affect price indices? You betcha. If you make it illegal to move prices, then at least official prices will not move. So let’s consider the potential impact of Mamdani freezing rents and grocery prices, for example.
New York City is about 7% of the CPI sample. Technically, it’s New York-Newark-Jersey City but we know most of that is NYC. In the New York consumption basket, Rent of Primary Residence is about 11%, 28% is Owners’ Equivalent Rent, and 8% is Food at Home. So, if rents and grocery prices were frozen, about 19% of the NY CPI would go to zero month/month right away (at least officially – the best tomatoes will be sold on the black market for a premium of course and the best catch of the day will be sold in NJ…[4]) And since OER is based on a survey of primary rents, eventually 47% or so of the NY CPI basket will go to zero price change. I’m ignoring the quality adjustments in the housing stock, which have the effect of increasing OER inflation slightly.[5]
The effect of this on the national CPI: if 47% of the NY basket goes from, say, 4% inflation to 0%, and NY is 7% of the national CPI, then the really-rough effect on the US CPI would be 47% x -4% x 7% = -13bps per year. Obviously that’s extremely rough, but I’m just aiming for an order of magnitude calculation. 13bps is small, but noticeable. Probably not tradeable.
But here is something that’s interesting and potentially tradeable. New York City is about 30% of the Case-Shiller 10-City Home Price Index. Let’s suppose that home prices in New York over the next year drop, say, 10%.[6] That move would cause the nationwide Case-Shiller (10-city) index to drop 3%, or to rise 3% less than it otherwise would. Here’s what is interesting. The chart below shows the February 2027 NYC Metro Case-Shiller futures contract, which trades on the CME (and settles to the index for December 2026, which is released in February 2027).
There has been exactly zero price effect of the Mamdani victory. To be sure, open interest in the NYC contract – in all of the Case-Shiller contracts, for that matter – is extremely low but there is an active market-maker and the current price as I write this is 344.40 bid/351.60 offer. The last print of the S&P Cotality Case-Shiller New York Home Price NSA Index, for August 2025, was 334.08. On the bid side, then, the market is paying 3.1% higher prices than the current index. That seems sporty to me. Why would home prices rise if rents are frozen? Why would they rise if people are leaving the city?
As always, my musings here are not trade recommendations; do your own research. Disclosure: I do not currently have a position either long or short in any housing futures contract, nor does any account or fund that I or Enduring Investments manages, nor do I currently have plans to initiate any position.
[1] New York, at least for now.
[2] At the time, we worried about what would happen with the CPI since a renter paying zero rent is not skipped but the rent goes into the calculation as a zero. So you could in theory have had 10% of the basket going to zero, which would have destroyed the inflation market.
[3] If you listen to the episode: I also love my thermometer analogy.
[4] Also, though rents will stop rising the quality of the apartments will deteriorate since landlords will skimp on maintenance. Mamdani has a plan for that, though – he has said the city will order maintenance to be done and if it isn’t, the city will seize the property. Just in case there was any question who really owns any property that you can’t pick up and transport elsewhere.
[5] N.b. – the increase in the CPI nationally from the owned-housing quality adjustment almost exactly cancels the decrease from quality/hedonic adjustments in other parts of the CPI. Yet another reason that the whining about hedonic adjustment being used to ‘manipulate CPI lower’ makes no sense.
[6] You can easily make a case for a much steeper drop if the city increases property taxes to make up for declining income tax collections, not to mention if the exodus from the city looks anything like the 9% of the population who claim they’d move if Mamdani won, or if the finance industry continues to relocate to Dallas and Miami.
An Update/Reminder on Rent Inflation
A subscriber to our Quarterly Inflation Outlook (you can subscribe here) wrote to me recently and asked about a research piece put out by a major sell-side investment house that discussed how private rental indices (such as Zillow) and the Fed’s NTRR (“New Tenant Rent Index”, as defined in a paper by the Cleveland Fed’s Randall Verbrugge a couple of years ago called “Disentangling Rent Index Differences: Data, Methods, and Scope”) were indicating that a decline in rent inflation was on the way. I felt like it was time for an update on this topic, since it has been a little while since the exact same arguments made the rounds a few years ago.
I even had a podcast (Ep. 74: Inflation Folk Remedies) in July 2023 in which I discussed (among other things) the NTRR issue. So the deceleration of Zillow and the other private rent indices, the NTRR which was forecasting sharply negative rent growth (before revisions!), the supply of new rental units – all of those are arguments from 2023!
Here are rents. The black line is the actual CPI for Primary Rents, y/y. In July of 2023, it was at 8%. You may notice that it never went negative in 2023 or 2024, and isn’t showing any signs of going negative in 2025.
Before I go any further, here is sufficient reason to ignore the NTRR, in addition to the other arguments I’ll make in a bit. Here is the chart of the NTRR from the 2023 paper.
And here is the updated NTRR from Bloomberg today. You will note that the 0% print in 2023Q1 from the 2023 paper has been revised up to around 4.5%. That’s even higher than the upper edge of the error range in the prior chart.
So forgive me if I don’t panic at the -2.2% current reading of the NTRR. Here’s the problem: the conviction among economists in 2023 (not just the Fed economists, but it was a general consensus at the time that rents were about to collapse) that the “stock” of rent inflation would eventually respond to the “flow” of new rents is just not how rents work. The new rents are not indicative of new conditions while the stock isn’t…those are two totally different populations.
There are people who turn over rents and move with some frequency, or who are moving now for one reason or another, and there is a stock of open units that landlords want to fill. But just because a landlord offers a low rent to fill his one open unit has nothing to do with his desire to cut rents on all of the units that aren’t turning over.
What is amazing is that the only reason this ever looked like it worked was because when both rates are very low, the noise outweighs the signal. So there’s no data for economists to really test the hypothesis on a period that matters because it’s similar to the current period of generally rising prices. But if economists just spoke to landlords, they would understand. That’s what I did, and the reason that in 2023 I switched my model from a top-down to a bottom-up (which is the dotted line in the first chart above…and that was not revised significantly). If costs are growing for landlords, they aren’t going to be cutting rents for their tenants even if they want to cut them for new tenants to fill a unit.
It should not be a surprise that the ‘faster’ NTRR has large error bars and large revisions. Essentially, the idea behind those indices is that they take the same rent data the BLS generates and squeeze out several different indices, some of which are “faster.” But basic information theory says you can’t get 3 bits of data from a pile that holds 1 bit, for free. What happens is those new indices are faster…but they have huge error bars that are huger the shorter the forecast length. Duh. Which means you can’t reject any null hypothesis about the near-term path. In the original paper they mention this and they show the data on the variance but they didn’t really explain it well. The short way to describe the problem is that you can’t get three pounds of crap out of a one pound bag. Period.
Now…having said all that I do entertain the possibility that rents could slow meaningfully further than here, even more than the mild softening that my model has. But my reasons for that are different:
- Rents in NYC will likely decline sharply if Mamdani wins, partly because jobs will absolutely flee the city but mainly because of his not-very-veiled-threat to seize property if they don’t. The smart landlords will dump their property at any price and get out, but some will try to ride out his term as Mayor. That’s unlikely to work but they’ll try. And NYC is a big part of the rent indices (by the way, one hedge for this is to sell the Shiller NYC property index, which futures trade (thinly, but they trade) on the CME. Combined with naturally slowing rent growth from some of the really hot but now getting overbuilt areas – like Miami – and you could get the overall indices to look better than the median would.
- (Offsetting this but probably nearer-term, LA rents will be buoyant for a while and maybe more sharply once the wildfires are further in the rear-view mirror so the claims of “profiteering” can be ignored. They bounced right after the fires destroyed a huge number of units but predictably people screamed at landlords so that stopped. But supply and demand, you know. There are fewer rental units in Los Angeles, and rents are going to go up faster as a result).
- If mass deportations really do turn into mass deportations, then what we are already seeing with Lodging Away from Home could become broader pressure on rents. The hotels were where the newest and biggest wave of illegal migrants were housed in the big cities. Elsewhere, they live in apartments and sometimes own homes when they have been here for a while. I can’t imagine the government will be able to deport more than say 1mm over the next year or two. That would be 2000-4000 per work day, and while the illegal immigrants generally walked in they generally have to be flown out. However, 1mm is still a big number and if enough other illegals ‘self-deported’ so that you’re talking about a million households then you’d have to consider a good chance of significant housing disinflation as the stock of rental units – currently just barely out of shortage – becomes a glut from the demand side.
But note that neither 1 nor 2 is currently something that you’d be able to detect with NTRR or Zillow or other rental indices. Maybe at the margin we could see deportations affecting rents in some of the ‘sanctuary cities’ where a lot of the deportations are concentrated, but I doubt it. Too soon.
In any event, my forecast for rents is not super-aggressive and I recognize there are mostly downside risks associated with those enumerated reasons. But right now? In the data? There is nothing that looks like it spells housing deflation.
Framing Home Price Inflation
The ever-increasing cost of homes obviously causes a lot of people a lot of angst. Chief among those groups, naturally, are the people who are planning to buy a home but do not yet have one; and, since higher home prices are very highly correlated with higher rents, renters too are alarmed that the rent is too damn high! (if that reference eludes you, educate yourself at this link: https://www.youtube.com/watch?v=OUx_32ABtw4 )
Right behind the people who have to actually buy homes and rent apartments, though, are the economists who seem to be perennially alarmed that home prices are “in a bubble” again. Certainly, if you look at nominal home prices (represented here by the S&P Case Shiller U.S. National Home Price Index, normalized like all of the charts in this article so that December 31, 1989 = 100 and the latest figure is for the end of September 2024) then you can see the cause for concern. Home prices are up 75% since the peak of the home price bubble of the late 2000s! If a house at $241,000 was in a bubble in 2006 (and subsequently declined in price to 175k), then surely it’s in a bubble if it’s now at $425k?
You can see in this chart the rapid acceleration in 2021-2022, and that should be a clue about one of the things that is going on with home prices. The overall price level is a lot higher than it was in 2006; the dollar simply doesn’t go as far as it did back then. Indeed I’ve chronicled how, thanks to the supercharged increase in the money supply, consumer prices are up 23% since just before COVID. Obviously, then, we have to adjust the dollar price of a home for the change in the measuring stick (the dollar) itself. Here are real home prices.
This still looks like a bubble, if real home prices are 13% higher than the bubble peak. After all, homes are unproductive real assets. Over a long period of time, home prices have risen less than 0.5% per year after inflation. In this way a home is like a lump of gold. Ten years from now, the lump of gold is still a lump of gold and so you would expect the real return to be roughly zero (you have the same amount of stuff at the end that you started with). In the case of home prices, there is deterioration of the housing stock over time but also new construction and homes have historically gotten larger and more comfortable, so some small drift higher in real prices makes sense. But home prices since 1989 are up 70% in real terms, when they should be up roughly 25 * 0.5% = 12.5%. And since the 2006 peak, we’d expect 9% (18 years x 0.5%) would represent a similar peak. We’ve risen more than that!
So, definitely a bubble, right? That deflation everyone keeps promising us is imminent, along with the collapse of banks and all the other stuff? Not so fast; there is one other important thing to consider and that is household formation. Or, rather, household formation compared to housing-unit formation.
We start by imagining what could plausibly push real home prices above or below a long-run flattish trend, that would represent a legitimate effect and not a bubble. What immediately comes to mind for me is the fact that for at least the last few years we have seen a massive increase in the US in the number of heads over which we need to put roofs. Something in the ballpark of 10 million new residents need roofs, and we surely have not constructed that many new roofs. For a long time, I’ve been highlighting this as one really good reason to not expect rental or home price deflation: the demand relative to the supply is out of whack. However, it turns out that we don’t actually need to rely on the ‘unofficial’ increase in the population to conclude that the “bubble” isn’t so bubbly.
The chart below, covering 2004 to the present, shows the real home price (the second chart above) on the y-axis. On the x-axis, I have a ratio of the number of households in the United States (source: US Census Bureau) divided by the total number of housing units in the country (source: US Census Bureau). As the ratio moves higher, it means there are more households for every housing unit or alternatively, fewer vacant units. I only have the housing unit data back to 2004, as that’s what was on Bloomberg.
There is a pretty clear relationship here between real home prices, and the occupation ratio. I have highlighted two areas. One, in red, is the January 2006 through June 2007 period – sort of the teeth of the housing bubble. Those points are well above the line, suggesting that prices were high relative to the occupation ratio. In fact, January 2007 is the point that is the highest above the regression line. On the other hand, we have the most-recent point in green. This is right on the regression line. Yes, real home prices have gone up a lot. But that’s mostly because the construction rate of new housing units has not kept up with household formation.
As an aside, the three points at (0.91, 130) or so are from mid-2020, when there was a surge of household formation but home prices (and rents) were being constrained by the lockdowns. In retrospect, it was a great time to buy a house!
Note that the charts above do not include undocumented residents in the US, except inasmuch as the Census Bureau is including them. Since the total increase in households since January 2021 is only about 6mm…and for the prior 4 years, the increase was 5mm…I am fairly confident that the recent surge in illegal immigration is not reflected on this chart. Ergo, you could make the case that home prices are too low in real terms. If every 5 illegals form one household, the ratio would rise from 0.912 to 0.926, and we would be off the chart to the right-hand side.
Now, this does not mean that real home prices will not decline. In fact, I am very confident that at some point they will, as building catches up with household formation. That does not mean that home prices will fall in nominal terms, however; I suspect that what is more likely to happen is that over a number of years, home prices will drift sideways to slightly higher while overall consumer prices continue to rise. But, if 10 million illegal immigrants are deported, the building of new units will catch up a lot more quickly and nominal prices and rents could decline in that case.
If that happened, rents really would be ‘too damn high’. And that is one very big reason that mass deportation is not inflationary. It also is not very likely; I have the over/under at 1mm deportations.
“Why Aren’t Home Prices Falling?”
From time to time, I like to point out errors that we make because we think in nominal space, or because we had 25 years of inflation being so low that we didn’t have to think about it very much. I do think that at some level, we should consider pointing the finger at economics education, which teaches static equilibria until you get into fairly advanced (graduate level) classes – and even then, generally in nominal terms.
There’s a very good videocast that I like to check in with occasionally, by Altos Research, which runs through recent data on home buying trends along with useful commentary. It tends to be more thoughtful and to not fall victim to the wild swings of emotion that seem to affect a lot of housing market observers. I think it’s important for me to say that I like this channel, since I’m about to criticize an episode they recently put out.
It was called ‘Why Aren’t Home Prices Falling?’ and you can find the quick 15-minute video here: https://www.youtube.com/watch?v=J-0bkqeFZEE. You can get a good feel for the videocast, and the useful analysis they bring, from this episode.
But the question ‘why aren’t home prices falling?’ is an odd one. Median CPI is still running at 4.2% y/y. Sticky CPI is +4.1%. Apartment rents are +5.0% and never declined y/y, even when there was a rent moratorium. Asset prices in general are quite a bit higher over the last few years also, so whether you’re looking at homes from the standpoint of an investment or a consumption item, it’s hard to see why one would naturally default to ‘home prices should be falling.’
The thought process is that ‘home prices went up so much, no one can afford them! Therefore, prices should fall.’ This thought process does not originate with Altos; they are just trying to answer the question being asked. In my view, though, they aren’t answering the right question. Really, when you think about it, the whole framing of the question evokes Yogi Berra saying that ‘no one goes to that club any more because it’s too crowded.’ Home prices going up a lot is a pretty serious piece of evidence that supply and demand has previously cleared at a price that (it is assumed) is too high for people to afford. That should sound odd.
The thought process goes further by noting that the volume of transactions has really declined markedly over the last couple of years, thanks to high interest rates keeping supply off the market as homeowners with current low interest rates locked in recognize that buying a new home would involve an effective refinancing to more expensive money. But if that restriction in supply is the main reason that home prices didn’t decline, then why have home prices in Australia and the UK also generally been rising, except for a dip around the same time that we had a dip in the US? Australian mortgages are normally floating-rate, and in the UK a 5-year fixed rate is the standard. But the low y/y change in Australia (according to the Dallas Fed’s index of Australian home prices – don’t ask me why they track Australian home prices) in 2023 was -4.3% (now +7.7%), the low in the UK was -2.5% (now +2.2%), and the low in the US was -3.4% (now +2.9%, using Existing Home Sales Median y/y). All of those markets saw very large rises, small and brief declines, and are now rising again.
These are very different property markets, very different mortgage markets, very different governments, taxation regimes, populations, and yet they have strikingly similar patterns of home price changes in a market that classically is all about ‘location, location, location.’ This should lead the thoughtful analyst to think that there’s something else going on.
The something else – not to beat a dead horse again – is the change in the quantity of money, which has followed a very similar pattern in every major economy in the years after 2019. And this is where conventional Economics education falls short. Here is a chart of the y/y changes in US M2, alongside the y/y change in Existing Home Median sales prices.
Not all of the price changes you are seeing in homes is a ‘real’ price change. Much of what you are seeing is a change not in the value of a home, but in the value of the currency unit relative to durable physical assets. But in Econ 101, they’d tell you that you should look at changes in supply and demand, and that will predict changes in the price and quantity at which the market clears. In that narrow frame, you might look at the large increase in home prices and attribute it to changes in demand due to declining interest rates, although you’d be confused when the massive increase in interest rates caused only a modest and temporary drop in nominal home prices. (In late 2022, the Case-Shiller futures for end-of-2023 were pricing in a 19% decline in nominal prices with inflation at a positive 3-5% per year, implying an unprecedented collapse in real prices).[1]
Obviously, that frame doesn’t make sense when the underlying price level is rapidly changing, and the underlying quantity of money is rapidly changing. This is often more obvious when we make it extreme. Suppose the money supply went up 400%, and prices quintupled as well, and interest rates went to 100%. Would you expect home prices to decline in nominal terms? That would be absurd – the price level going up by a factor of 5 means that the value of the measuring stick is what is changing. And remember, it is entirely consistent to have the volume of transactions decline sharply while the nominal price increases. Homebuilders care about the volume of transactions; homebuyers care about the price. You may be absolutely bearish on homebuilders, while still expecting home prices to increase, especially if the price level is increasing.
That’s exactly what we have been experiencing. And, with the money supply growing again and median prices still rising at 4% per year, it does not seem to me that there is any natural reason to expect home prices to decline. So the short answer to the question ‘Why Aren’t Home Prices Falling?’ is ‘There’s no reason they should.’
[1] Markets are where risk clears, not where investors ‘expect’ prices to be, and there were wonderful gains to be made even well into 2023 by helping the nervous real estate longs clear their risk. https://inflationguy.blog/2023/08/29/home-price-futures-curve-still-looks-weird/
Rising Mortgage Origination Hints at M2 Turn
One of the successes the Federal Reserve can tout from the last couple of years (and the list of them is pretty short, to be fair) is that after the unprecedented policy actions during COVID caused never-before-seen rates of money supply expansion, subsequent policy avoided normalizing that explosion.
Year over year growth in M2 reached 26.9%. But in 2022, as the Fed started hiking rates and shrinking its balance sheet, the rate of growth slowed until M2 reached its absolute peak in July 2022 and began to slowly decline. As of today’s H6 release, year-over-year M2 has been negative for 13 months in a row.
To be sure, after a massive explosion the level of M2 has not declined all that far as the chart below shows. I also documented this fact back in November in “Where Inflation Stands in the Cycle,” which was really a good piece. You should read it.
So the success of the Fed here can be summarized by saying, ‘at least they didn’t keep blowing up the money supply.’ Since the rise in prices is clearly and closely related to the explosion in the quantity of money we have seen (anyone who still resists this obvious truism after the mountain of recent evidence is added to the prior mountain of evidence), this was a sine qua non for getting inflation back down. It isn’t sufficient, unless it’s continued for a very long time, but it’s necessary. As I illustrated in that article linked to above (which, really, you should read), there are several ways that inflation could evolve from here as the shock to the system gradually unwinds. I’ve talked before about how velocity in the policy crisis behaved as a spring or a capacitor, absorbing a lot of ‘monetary energy’ that is doomed to be released back into the system. Velocity is still rebounding (in Q4, if forecasts for Thursday’s Advance GDP report are accurate, it will rise something like 4% annualized), but if money growth remains negative then that’s really the least-painful way this can resolve. In the last chart from that prior article (have I mentioned it’s worth reading?), slack money growth with decent growth and rebounding velocity is reflected in a movement mostly to the left, with the price level not rising much. Good outcome.
However, that outcome is predicated on the notion that the money supply remains slack. If M2 starts to rise again, then the curve drifts upward and the potential set of outcomes almost certainly involve higher prices. Naturally, I’m mentioning this because of developments that make me concerned on this score.
One thing that I seriously missed in 2022 was the fact that the increase in interest rates helped bring down money supply growth. That’s not at all intuitive, because in general changing the price of a loan tends not to change the demand for a loan by very much – especially when higher inflation is making the spot real interest rate paid by the borrower lower and lower. In other words, I assert with some decent evidence that consumer and industrial loan demand is somewhat inelastic for modest changes in interest rates. Ergo, my belief was that merely raising interest rates would not necessarily cause money growth to decelerate. As it happened, I was saved from my own mistake by the fact that the Fed was also shrinking the balance sheet, which (despite the fact that reserve balances aren’t binding on banks in the current environment, so they are essentially unconstrained in lending) I thought might help money growth to decelerate. Not that I thought we’d keep getting 20% growth, but I didn’t think we would have naturally seen money growth fall below, say, 5%. Fortunately, because the Fed was also shrinking the balance sheet my forecasts were not drastically inaccurate despite being wrongly inspired, and so I forecast 5.1% median inflation for 2023 and we got 5.06%. It’s nice when the ball actually bounces your way.
As it happens, though, for the most part higher interest rates seem to have not affected loan growth very much. C&I loan growth remained strong throughout 2022 and didn’t start to level off until the Fed was just about through tightening, and consumer loans as I expected really only started to level off when the Unemployment Rate started to rise…credit cards, not at all. And that’s because, as I said, most borrowers are not borrowing because they made a NPV calculation that said borrowing makes sense; they’re borrowing because they need to and 1% or 2% or 3% doesn’t really change that calculus very much.
But you know where it did change the calculus a lot? In mortgages. And that’s because a buyer might be reluctant to pay 1% more on a mortgage, but what the buyer also needs is someone who is willing to abandon their awesome loan. As has been noted elsewhere by lots of people, home sales absolutely cratered not because people weren’t wanting to buy but because there weren’t enough people who wanted to sell. So mortgage origination volumes also dried up, as a direct consequence of higher rates. The one large market where interest rates did have a big impact, although not for the reason you’d think, was in mortgages!
You know I wouldn’t say this unless I had a neat chart to show you. Here is the Mortgage Bankers’ Association Purchase index, tracking the volume of new loans for purchasing a home (in black), set against y/y money supply growth, in blue.
Let’s tie this up with a bow:
- Higher rates didn’t affect every kind of loan, but had a big impact on turnover, and thus origination, in one very large loan market: mortgages.
- Lower mortgage origination turns out to have been uncannily correlated with money supply growth. This may or may not be causal, but it at least means that mortgage origination merits consideration as a leading indicator of money supply growth.
- As interest rates have leveled off and even declined some, the housing market is gradually adjusting. We are seeing higher home prices, and mortgage origination has been showing signs of recovering as the chart shows (mortgage origination numbers are released before sales numbers, so expect a rise in home sales coming).
- It is going to be difficult for the Fed to keep the money supply shrinking, if origination of new mortgages rises even a little bit. This doesn’t mean M2 is going to skyrocket, just that it is going to stop shrinking (in fact, it has risen each of the last two months).
- If M2 rises at even a sober 5% pace, combined with money velocity that still has some normalization left, it will be extremely difficult for the Fed to hit its inflation target on a sustainable basis for some time.
And what should you do about it, just in case? For starters, read “Inflation Sherpa.”
Home Price Futures Curve Still Looks Weird
As we all know, shelter inflation is very important to the overall rise in the cost of living. Recently, concern about the acceleration in shelter inflation that happened between 2021 and 2023 has been dampened somewhat as the CPI for primary rents and for Owners’ Equivalent Rent have both clearly peaked.
We know that the rent deceleration will slowly unfurl over the next year, and the Fed has recently admitted that this ability to project rent deceleration gives them some comfort on that score. Now, I recently talked in one of my podcasts about why the new models forecasting rent deflation in our relatively-near future should be taken with at least a grain, and probably a shaker, of salt…but that being said, both of our models suggest rent inflation may decline to 3% by the second half of 2024, but is unlikely to drop further than that. Our newer, unique model is driven by modeling landlord costs, and it looks very promising.
Our older model has fared worse, but with good reason. It overestimated rent inflation in late 2020 into 2021, because the eviction moratorium put pressure on rents but did not affect home prices, and then underestimated inflation thereafter partly because rents had to catch up when the moratorium was lifted.
Although the first model above looks to predict much more tightly (although that’s partly because it’s a new model so we are just now generating out-of-sample data to compare with forecasts), the second model is of interest today. That model uses several different measures of home prices and related series, and blends them with different lags to generate the forecast. In other words, this model relies on the behavior of a substitute to rented housing, and that is owned housing. It’s interesting here because it is fairly typical of the way rent inflation has historically been estimated: as (mostly) a lagged function of changes in home prices. If home prices go up, then rents tend to rise because the price of a substitute is rising; if home prices fall, then rents tend to decline because the price of a substitute is falling. Microeconomics 101.
The reason I mention this is because the predictions that rents will be in outright deflation next year are partly driven by the fact that home prices peaked in nominal terms last July, and so year/year home price inflation has declined from about 21% at the peak to slight deflation, in nominal terms, recently (using the S&P Case-Shiller Home Price Index). In real terms, the Case-Shiller HPI dropped about 9% from peak to trough, and around 6% in nominal terms. So, the thought is that the absolute level of rents need to not only level off but actually decline in order to be consistent with what is happening in the housing market. And, of course, people firmly believe that not only do high interest rates cause declines in housing activity but also (despite the lack of evidence) in home prices. I’ve been pointing out for a while that that’s not historically true – home prices in the 1970s never declined in nominal terms y/y and mortgage rates were high and variable.
Well, today the S&P Case-Shiller Home Price Index was released and once again surprised to the upside. In nominal terms, home prices are almost back to the highs although they remain a bit below the highs in real terms.
Now, I showed the SPCS 10-city, rather than the whole index, for a reason. The broader index looks the same, but you can’t trade futures on it. What is really fascinating to me is not just that home prices have rebounded faster than I expected – there really is a shortage of houses out there – but that the futures market is pricing in declines after September and that home prices will not reach a new peak until at least 2027.[1]
Look, I definitely believe that a recession is in store and we may already be feeling some of it although weakness in China could help hold it at bay for a bit longer. But higher interest rates have actually slowed down some of the home building that was addressing the housing shortage; moreover, in an inflationary environment such as the one we are in right now home prices can fall in real terms without falling in nominal terms. That misunderstanding…that ‘bubbly’ home prices would have to be resolved with a steep decline in nominal home prices…is why at one point the Feb 2024 CME Case-Shiller Home Price Futures contract traded as low as 268. That price implied a 19% nominal decline in home prices from the high, on top of inflation running at 3-5% per year, in a housing-shortage environment! As the Case-Shiller numbers have persistently run far ahead of that worse-than-the-global-financial-crisis scenario, the futures have slowly pinched higher. But it is amazing to me that, even has nominal home prices are about to reach new highs, that the market is pricing in a second decline in prices before a weak multi-year recovery!
Futures markets show where risk clears, not where investors think the price will be in the future. So what this is really saying is that people who want to hedge home prices outnumber those who want to buy homes cheaply. And that’s plausible to me. But it still seems amazing! And it also means that the following wind the Fed believes they have from disinflating rents…may not be as strong a following wind as they currently expect.
[1] Disclosure: I run a long-only strategy that passively uses this (fairly illiquid) contract, and so I am always net long. But, for what it’s worth, it actually works against my long-run interest to have the longer contracts trade up to make my future rolls more expensive, so hopefully you’ll all think home prices should be going lower.
No Need to Rob Peter to Pay Paul
So, I suppose the good news is that policymakers have stopped pretending that prices will go back down to the pre-pandemic levels. My friend Andy Fately (@fx_poet) in his daily note today called to my attention these dark remarks from Bank of England Chief “Economist” Huw Pill:
“If the cost of what you’re buying has gone up compared to what you’re selling, you’re going to be worse off…So somehow in the UK, someone needs to accept that they’re worse off and stop trying to maintain their real spending power by bidding up prices, whether higher wages or passing the energy costs through on to customers…And what we’re facing now is that reluctance to accept that, yes, we’re all worse off, and we all have to take our share.”
I think it’s worth stopping to re-read those words again. There are two implications that immediately leap out to me.
The first is that this is scary-full-Socialist. “We all have to take our share” is so anti-capitalist, anti-freedom, anti-individualist that it reeks of something that came from the pages of Atlas Shrugged. No, thank you, I don’t care to take my share of your screw-up. I would like to defend my money, and my real spending power, and my real lifestyle. If that comes at the cost of your lifestyle, Mr. Pill, then I’m sorry.
But the second point is that…it doesn’t come at the cost of someone else’s lifestyle. This is why I put “economist” in quotation marks above. There is still a lot of confusion out there between the price level and inflation, and what a change in the price level means, but if you’re an economist there shouldn’t be.
You see statements like this everywhere…”food prices are up 18%. If people are spending 18% more on food, it means they’re spending less elsewhere.” “Rents are up 17%. If people are spending 17% more on rent, it means they’re spending less elsewhere.” “Pet food is up 21%. If people are spending 21% more on pet food, it means they’re spending less elsewhere.” “New vehicle prices are up 22%. If people are spending 22% more on new vehicles, it means they’re spending less elsewhere.” “Price of appliances are up 19%. If people are spending 19% more on new appliances, it means they’re spending less elsewhere.”
You get my point. All of those, incidentally, are actual aggregate price changes since the end of 2019.
This is where an actual economist should step in and say “if the amount of money in circulation is up 37%, why does spending 18% more on good or service A mean that we have to spend less on good or service B?” In fact, this is only true if the growth in the aggregate amount of money is distributed highly unevenly. In ‘normal’ times, that might be a defensible assumption but during the pandemic money was distributed remarkably evenly.
Okay…the amount of money in circulation is a ‘stock’ number and the prices of stuff changing over time is a ‘flow’ number, which is why money velocity also matters. M*V is up about 24% since the end of 2019. So a 20% rise in prices shouldn’t be surprising, and since there’s lots more money out there a 20% rise in the price of one good does not imply you need to spend less on another good. That’s only true in a non-inflationary environment. The world has changed. You need to learn to think in real terms, especially if you are a “Chief Economist.”
(N.b. to be sure, this is somewhat definitional since we define V as PQ/M, but the overarching point is that with 40% more money in the system, it should be not the slightest bit surprising to see prices up 20%. And, if velocity really does act like a spring storing potential energy, then we should eventually expect to see prices up more like 30-40%.)
Here’s a little bonus thought.
Rents are +17%, which is roughly in line with a general rise in the prices of goods and services. Home prices are up about 36% (using Shiller 20-City Home Price Index), which is roughly in line with the raw increase in M2.
Proposition: since the price of unproductive real assets is essentially an exchange rate of dollars:asset – which means that an increase in a real asset’s price is the inverse of the dollar’s decrease – then the price of a real asset should reflect the stock of money since price is dictated by the relative scarcity of the quantity of dollars versus the real asset. But the price of a consumer good or service should reflect the flow of money, so something more like the MV/Q concept.
Implication:
Discuss.
Homes Have Gotten Cheaper by Running in Place
As the Fed started lifting interest rates aggressively in early 2022, pundits almost universally declared the end of the housing market. Taking the rather lazy approach of projecting what happened in 2008-2010 and just changing the year, utter disaster was forecast for home construction, home sales, and home prices. The more clever analyses mused about how the higher interest cost of a mortgage lowered the amount of home that can be bought by a given payment, and suggested that home buyers would naturally back up their bids by that much and sellers would be obliged to hit those bids.
The Case-Shiller Home Price futures, which are (thinly) traded on the CME,[1] went from pricing in additional upward movement in home prices to pricing in a collapse worse than the post-financial-crisis debacle. For example, the February 2024 futures dropped 22% between May and November 2022. Keep in mind that these futures track nominal prices, so at the worst levels the futures market was pricing in something like a 25% drop in real prices.
That was never going to happen, especially in a housing market that was much, much tighter than in 2007. In the summer of 2007 there were approximately 3.4 million existing homes on the market; in the summer of 2022 the figure was about 1.2 million. And, as it turns out, homeowners did not hit any bid that was shown, which would have been irrational in an inflationary environment. Nominal home prices are sticky on the downside anyway, because buyers don’t like to sell below other recent prices which serve as an ‘anchor’ for their expectations. All of which is to say that 2007 really was an amazing outlier in a lot of ways: price, activity, builder activity, financial buyer activity, mortgage structuring, and home inventory. The current situation is much different.
Naturally, we all know that now as we have noticed that home prices have not in fact collapsed. But they have declined in real terms, because the overall price level has advanced while home prices have been flat. Given the level of inflation, this has actually changed the level of home valuation fairly substantially in a short period of time and I thought it worth pointing out.
Consider the following chart (Source: BLS, ADP, National Association of Realtors, author’s calculations). It plots the Existing Home Sales Median Price divided by the Median Annual Wage. I’ve used the Atlanta Fed’s Median Wage figures and converted them to annual wages so that the series matches, for the most recent point, the median annual wage reported by ADP.) By doing this, we can see roughly how many years’ wages it would take to purchase the median home outright. Note that one of the series is seasonally-adjusted and one is not, which causes the scalloping effect you see. I could correct for this, but figure this is close enough to make the main point.
And the main point is that as home prices have stagnated and wages have been rising rapidly to keep pace with inflation, the cost of a home relative to the wages people are receiving has dropped pretty sharply.
Although this measure doesn’t tell the whole story, you can see how there was a reasonable concern that home prices may have been getting ahead of themselves somewhat (although with extremely low inventory, that’s not necessarily unsustainable in the medium-term). However, since last summer homes have gotten much cheaper, by just staying in one place.
Don’t get locked in on the nominal price. That’s called money illusion, and in an inflationary environment it leads to mistakes.
[1] In full disclosure, we use the housing futures for one of our strategies.
Do Rents Really Actually Lead Home Prices?
The inflation thesis at this point has both a top-down and a bottom-up rationale (as all good theses do). The top-down rationale is that the extraordinary rise in the quantity of money over the last few years has yet to be fully reflected in the price level; ergo, inflation should continue for a while – even if money supply growth stops cold – because the price level has a lot of ‘catching up’ to do.
The bottom-up rationale depends a lot on what happens in the housing market. The first place that prices shot up was in the more flexible components of inflation, especially in goods. “Sticky” inflation followed, only turning north in 2021 and then accelerating in earnest especially as the eviction moratorium eventually ended and rents began to catch up. As the chart below (source: Atlanta Fed) illustrates, core “flexible” CPI (in white, right hand scale) is decelerating and is down to about 7% y/y…but core “sticky” CPI (red, left hand scale) is at 5.6% and shows no signs of even peaking.
An important part of the “sticky” basket is the weighting assigned to rents. Rents show up as both Rent of Primary Residence (you rent a place to live) and Owners’ Equivalent Rent of Residences (your opportunity cost is that you don’t have to pay for an apartment, so this is an imputed cost). Both rents move together, mostly because the Bureau of Labor Statistics reasons quite naturally that the best measure of the imputed rent a homeowner would pay is the market for rentals that he/she actually could pay. These two pieces of CPI are the biggest and the baddest, and they don’t even exercise. I always say that if you can forecast rents accurately, you will not be terribly wrong on overall inflation. Rents are the 800lb gorilla. Where they sit has a big influence on overall inflation.
Traditionally, observers of the inflation market have forecast rent based on a simple lag of home prices. There are reasons to suspect that’s not the whole story, but it has worked for a very long time. Here is a chart of the last 20 years or so, with the Case-Shiller index (lagged 18 months) in green and the Existing Home Sales Median Price y/y (lagged 15 months) in blue against Owners’ Equivalent Rent in red.
Even though inflation as a whole has been low and stable, home prices themselves have varied enough thanks to the housing implosion in the mid-2000s that you can see a reasonable outline of why inflation people tend to like this simple model. It’s at least suggestive.
Recently, that has been called into question by a researcher at the Mercatus Center at George Mason University. Kevin Erdmann wrote a paper published this year entitled “Rising Home Prices are Mostly from Rising Rents.” When the paper came out I tweeted it with the note “I need to read the whole paper.” If Erdmann is right, then the entire market is doing it wrong and (a) home price inflation should not be slowing down right now, since rents are not, and (b) the way the market models rents is just plain useless. So, this was definitely worth looking at from my perspective!
Well, I’ve read the paper. I am sorry to report that in my view, the author makes very strong claims but supports his argument with very weak statistics. That being said, I still think this is a paper worth reading – some might come to a different conclusion than I have.
It isn’t like I think the author is completely out of his gourd. It is absolutely reasonable to expect home prices and rents to be related since they are both ways to acquire shelter services. It isn’t as if Erdmann is saying that they aren’t related, and some of his cross-sectional data and findings are interesting. The problem is that he starts with a mental model of how things work, and then proceeds to show information which, given his assumptions, seem to support what he is saying. The mental model isn’t absurd: a home can be thought of as a way to purchase a whole stream of shelter services in one lump. When home prices rise, it could mean that buyers are evaluating this stream of services as being worth more than they previously were because they are observing rising rents, or because they were priced out of the rental market and chose to buy an asset with a shelter services component instead.
But it could also be the case that home buyers are reflecting rising expectations of long-term rent inflation, in which case spot rents needn’t change at all. It might be the case that home buyers are making totally stochastic decisions, and it just happens that when lots of people buy homes it pushes up home prices which then displaces people into the rental market.
All of these stories would result in time series that are highly correlated. And Erdmann has a number of illustrations and data points showing that there is a correlation. For example, he pointed out that in 2021, “the metropolitan areas with the highest rents also had the highest prices.” However, Erdmann’s real claim isn’t that home prices and rents are closely related, but that rents lead home prices. The point about the connection of rents and prices in various metropolitan areas is not evidence supporting his claim that rents cause prices, but it doesn’t refute it either. The problem is, he takes such data as support of his claim, when it isn’t. This turns out to be his modus operandi – start with a mental model of how it works, show data that demonstrates the two things are connected, and then assert causality.
In the paper, there is not a single test of causality. With time series, we can test whether one series statistically leads another in various ways; for example, with the Granger Causality Test (which doesn’t actually test causality but merely the lead-lag relationship). If the point of the paper is that (contrary to the usual assumption) movements in rents cause movements in home prices – which is a big claim – then at the very least I’d have expected to see a Granger test.
There is some evidence that statistical inference is not the author’s strong suit. He shows several clouds of data points where any reasonable person can see there is no clear trend, and then proceeds to run a regression line through them. The fact that we can calculate a regression slope – we can always calculate a regression slope – does not mean that it is statistically significant. And even if it is statistically significant, it may not be economically significant. Unfortunately, there are no such tests of significance in the paper and I suspect for several of the charts it would be impossible to reject the null hypothesis that there is no relationship at all between the variables despite a provocatively-drawn regression line.
He also has a figure (Figure 9 in the paper) which shows changes in prices and rents for a number of metro areas over time. Clearly, there is a positive relationship – but no one disputes that. The question is, does the relationship get better when you lag one of the variables? No such analysis is done.
In general, all the author “proves” is that there is a relationship between rents and home prices, which I think we already knew. The rest of it is storytelling, trying to persuade us that the causality makes sense his way. I don’t mean to suggest that the paper is a complete bust! The author does have some good ideas that I will borrow. He makes the point that discounting home prices by general inflation doesn’t really make sense because we don’t care about the general price level when we buy a home; we care about the price level of shelter. This is a simple point, but fairly profound in a way. It risks being somewhat circular if we aren’t careful, but it’s a good point.
And the funny thing is, despite the fact that I think the evidence is much stronger that the evidence for causality runs the other way, I agree with some of his policy conclusions. His main conclusion is that “…if rising rents are the more important factor [rather than temporary demand factors or monetary stimulus], then policies aimed at stimulating more construction may be more apt and may help increase real incomes for Americans in neighborhoods where rents have been rising.” I completely agree that, given the severe housing shortage that we seem to have in this country, that making it easier for builders to create homes and apartments would be good industrial policy.
But you don’t need to believe that rents lead home prices to think that is a good idea!
