Last weekend I wrote a long, chart-laden post that summarized the evidence about housing in the US. I received some criticism that seemed reasonable, so I figured I'd address them here.
Anne Paulson suggested that I shouldn't compare housing growth to population growth because housing is an adult thing. If a family has more children, it will still occupy a single housing unit. Here is adult population growth vs. housing unit growth:¹
I charted this over a long period in order to provide some context. After World War II there was a huge shortage of housing as soldiers returned home and got married. It was a major political issue that produced housing developments like Levittown; the growth of suburbs and interstate highways; and a huge increase in housing projects for low-income families.
The result was strong growth in housing units, followed by another strong growth in the '70s and '80s as the baby boomer generation grew up and moved out. More recently, growth slowed down in the aughts and flattened out to nearly nothing in the 2010s. In 2021 the number of adults barely budged, ending at 252.6 million. The number of housing units ended up at 142 million, for a ratio of 56.2% (compared to 56.6% in 2000).
You can draw your own conclusions from this. On the one hand, housing growth has slowed down considerably. On the other hand, this was largely because population growth also slowed down considerably. On a countrywide basis, we have the same number of housing units available per adult as we did 20 years ago.
A second comment came from Kevin Erdmann, who criticized my use of housing inflation vs. overall inflation. As it turns out, "housing" includes furniture and other things that go into the total cost of living. "Shelter" excludes those, but includes hotels and dormitories. The best measure is rent, which the BLS calculates directly for renters and as owners' equivalent rent for homeowners. I weighted the two together and produced this chart:
This shows a larger increase in rent CPI than my previous chart. In particular, rent has grown 11% more than overall CPI since 2000 (184 ÷ 166 = +11%). This suggests a tightening of the housing market—the rental market in particular—but note that it's a measure of inflation, not prices directly. I also included a chart that directly measures home prices, including the effect of interest rates, and it shows a decline in home price. Another chart shows average gross rents, and it suggests very little change at all.
In any case, these are easy charts to correct, so why did it take me so long to produce them? That's a different, and more complex story, and you might just want to stop here and skip the explanation. But if you want the read the whole thing, keep going.
Kevin Erdmann also had a longer and very different criticism of my piece. It took me quite a while to plow through it, and that, among other things, explains why this post is showing up on Sunday instead of last Monday.
To begin with, Erdmann has written a couple of books making the case that there was never a housing boom. There are quite a few economists who now agree with him, so this isn't a wildly contrarian take. And if you'll take a look at the top chart, which compares housing units to families, you'll see one bit of evidence that he's right: The line from 2000-2010 is pretty smooth, with only a small upward rise. This may not suggest a housing shortage, but it doesn't suggest a housing boom either.
However, Erdmann believes that we do have a housing shortage now and have had one for quite a while. What happened in the early aughts was not a housing boom, but a credit boom. Banks lent freely, but many cities refused to allow enough home and apartment construction to match the level of credit available. As a result, the price of housing of all types—both low-tier and high-tier—was bid up.
Now, obviously anyone who thinks we already had a housing shortage, and then failed to have a housing boom, is likely to disagree with me about whether we're currently suffering from a housing shortage. However, Erdmann's belief is, in the end, not so different from mine: Overall, there's no housing shortage, but there is a housing shortage in hot urban areas that people want to move to:
This is a new era in American housing....The norm until recent decades was for metro areas to grow when they are economically ascendant and to stop growing when they struggle. During the early and mid-20th century, the Detroit metro population grew by nearly 3% annually. Back then there weren't individual cities with incomes far above the norm, like there are today. Over the early 20th century, people moved to opportunity and regional incomes were converging as a result. Today they can't, so some metro area incomes are skyrocketing away from the norm. This is new!
As the chart shows, the highest rents and home prices are all in California—and unless I miss my guess, the blue circles representing 2007 are the same four cities. In other words, this is the standard complaint: House prices in California are rising too much for anyone to afford moving there. But that's where people want to go! So they keep trying, and this increases house prices (and rent) even more.
More interestingly, Erdmann makes the case that "credit suppression" is making hot urban areas especially unaffordable for low-income workers. This is because banks, which were lending to anyone before 2008, have learned their lesson and are now less likely to offer loans to working class families. Before 2008, prices for all types of neighborhoods went up and down at about the same rate. After 2008, low-tier housing collapsed because banks refused to lend into poor areas. For example:
I can't guarantee that I fully understand Erdmann's argument, but I guess I'd distill it into two points:
- California has a housing shortage. The rest of the country not so much.
- Thanks to the subprime crisis, working class families have been significantly shut out of the credit market. Richer families have fully recovered from the housing bust.
I haven't read Erdmann's books, so I don't have the full story on why he thinks we have a housing shortage. However, you can read a condensed version here.
POSTSCRIPT: I'm going to stop here because I'm just confused by too many of Erdmann's arguments. For example, he says that if rent inflation is high, but people are spending the same amount every year on rent, then they're spending less in real terms on rent. That's true. But I didn't say that people are spending the same amount on rent. I said they're spending the same percentage of their income on rent, which is a growing number. Maybe Erdmann has an explanation for this?
¹The number of families is here, Table FM-1. The number of housing units from 1950-2000 is here. The number of family units from 2000-2021 is here.
That's not too surprising. IIRC the percentage of US households that are homeowners only went up by a few percent even in the peak "housing bubble years" of the Aughts, and most subprime mortgages in that period were "cash-out refinancing" mortgages rather than new home purchases (IE people refinancing to get money out of their home values for consumption).
California is doing some incremental steps to increase housing supply, but they probably should take another look at the "found new satellite cities" idea. They have a fair amount of greenfield land for development that could be zoned to allow for any type of housing construction (including very high density stuff), and connected by roads and regular transit light rail to existing urban areas. People already live in some cities in California and commute decent distances to work in others, so it wouldn't be that strange either.
Thanks for engaging, Kevin! All of this is reasonable. A couple points:
1) Yes, you're right, my language may have been a little sloppy. Rent inflation has been high relative to general inflation and growth of real housing consumption has been lower than general real income and consumption growth. Housing per capita hasn't been declining in outright terms. That would be a very low bar. And, I would reiterate, as I wrote in my blog post, relatively low real growth combined with high inflation is the mark of market whose primary differentiator is inelastic supply.
2) I think it is giving California too much credit to simply assert that it is a popular place to live and to wash your hands of the fallout. The expensive cities (NYC, LA, SF, and Boston) have been extreme outliers in terms of supply. (As you discussed, post-2008 markets are distorted by changes in lending, but) pre-2008 those metro areas allowed FAR fewer new homes than any other major metro - including rust belt cities that have been struggling economically. That is what makes them different. I say that LA might be a more popular place to live than Kansas City. It's certainly plausible. But, LA would need to be willing to grow at Kansas City's (tepid) growth rate to really claim that is the driver. LA might be more popular than Kansas City. That's a reasonable assertion. On the other hand, LA permitting far less housing production than Kansas City is an empirical fact. Can we demand that LA approve more homes at a higher rate than Kansas City, or even Detroit, before we conclude that LA is just so darn special that all the poor folk need to move out to Phoenix and Dallas to make room for the newcomers or the next generation of children of the better off Angelinos? Is there really a more plain and reasonable solution to those folks' displacement than just allowing a moderate, say, 2% population growth each year?
Dunno. Is there a solution for the water-supply needs of Los Angeles and San Diego that doesn’t massively impact the natural environment with unpredictable consequences?
That is certainly a concern to manage for, but it isn't the reason that LA and San Francisco housing is expensive, not to mention Boston and NYC.
Demanding that Los Angeles, a city with high density, geographic barriers to expansion, and water supply issues, provide the same production as relatively open Midwest cities is, indeed, unreasonable, especially if you're comparing apples to apples.
The fundamental problem is that increasing housing in LA is much like increasing freeway lanes. No matter what you do it will get eaten up and things will get just as bad as before. I'm sure no urban area in America has produced more housing over the past 70 years than Los Angeles and yet Los Angeles still is amazingly unaffordable. There's no indication that increasing housing at any rate will make things any different.
Is there a better solution? Yes. Yes there is. Make the Midwest not be such a hellhole. Get rid of laws making it difficult for LGBTQ people to live decent lives. Get rid of racist attitudes. Be a supportive place for lots of different cultures and identities. And that has the happy side effect of ameliorating the disfunction of our federal government, where our janky constitution is barely functioning in response to the massive shift of population from the Iowas to the Californias.
Lolz, those "laws" already exist idiot. That is irrelevant, but your too stupid to understand it.
The Midwest is not a "hellhole" unless maybe you mean the climate (and frankly I'd rather deal with some winter cold than the summer heat in places like Florida or Texas). There are even nice progressive areas out that way, albeit not at Berkley CA level of progressive. The principle issue, other than the climate, is a dearth of middle class jobs.
There are other aspects to the Midwest that make it more on the hellhole side - too many areas with poor internet access, poor public transportation and lack of diversity - no one wants to be the first to move in, easier to move to enclaves where your "tribe" already lives.
I grew up in Michigan (near Ann Arbor) and also lived four years in Akron OH. I've known people in or near Indianapolis, Chicago and Minneapolis. I really don't recognize the picture you paint to the region, though I suppose it applies to places way out in the boondocks-- which is true in all 50 states. But if you're anywhere near a significant city, it isn't.
Demanding that Los Angeles, a city with high density, geographic barriers to expansion, and water supply issues, provide the same production as relatively open Midwest cities is, indeed, unreasonable
No it's not. We have the technology to put more housing per given unit of land. Indeed, this technology has been around for many decades!
Let them drink sea water!
Thanks for all the work on housing, it helps provide depth and balance.
I can't find a source for the "Median Home Rent and Price" chart (or quoted text) apparently from Erdmann. It doesn't current appear on his blog or anywhere else on the web when I search.
Could you cite this specific chart please? I'd like to find the details.
I'm not sure, without digging, which blog post it is from, but the data I used is MSA level data from Zillow (household income, rent and mortgage affordability, price/income ratio), which lately they haven't been publishing, but which you can get from the wayback machine ( https://web.archive.org/web/20190807144913/https://www.zillow.com/research/data/ ).
At the link, they are under "More Metrics"
+1 for this data from Zillow. It's extremely useful and I wish they would go back to publishing it.
The big cities mentioned get all the press, but there are lots of other popular places that have the same problems. Think college towns, resort towns, government cities. Where do short-term rentals fit in to the picture? In several resort areas I know of the rental shortage is extreme for the working class due to very few year-round rentals. Wages are higher, but not high enough even for people like policemen and school teachers to live in the actual towns--they all have long commutes. People lower down the income ladder are living in cut-up basements and garages.
Indeed. Boulder, CO, for example, was once a pretty sleepy college/resort town. It's completely unaffordable now, even for the faculty of CU. Same with places like Portland, OR, Boise, ID, Austin, TX, etc. About the only affordable cities these days are in the deep South, where governors and legislatures are making it all but impossible for anyone but white, gun-waving, right-wing fundamentalists to feel safe.
Nope. Debt is debt. Let the system die, most of those "pecker/nigger" small towns would starve to death as food production collapses. Famine would wipe out the herd so to speak. Gun toting???? Please, irrelevant.
It's not just big cities or destination/college towns. It's mid-sized cities, too.
Charlotte, Austin, Des Moines, Provo, etc. A simple look at the 2000 and 2020 census data on fastest-growing MSAs/CSAs and comparing it to the housing unit growth rate in same MSAs/CSAs shows that with only one or two exceptions (Provo being one of them), there is an acute housing shortage all over the country *in the places where people want to live and are moving to.* It isn't just California. I'm really getting sick of KD's assertions on this front.
The first two of those I'd actually classify as big-ish...
The regional chart is interesting. I wonder what would happen if you took tech out of SF and entertainment out of LA - where on the chart would those cities be then? And that the hell is happening in San Jose?
Google.* Google is happening in San Jose.
*Google standing in for a bunch of high-flying tech companies.
Am I missing something, but hasn't the total number of family units gone up with more singles, with and without children? Some are elderly widows and widowers. That would create a housing shortage by itself.
Seems reasonable to assume that there has been an increase in the number of single adult households (widowers, unmarried young adults) since the 1990's. This would mean that the housing shortage is more acute than the change in adult population would imply.
Widow(er)s do not create additional housing demand-- presumably the surviving spouse was already living somewhere before the death-- and the deceased spouse has no demand for housing.
“Overall, there's no housing shortage, but there is a housing shortage in hot urban areas that people want to move to.”
That’s like saying “overall there is no job shortage for [college grads, high school grads, whomever], but there is a job shortage in hot urban areas where all the jobs for that demographic happen to be.” Even when it’s super obvious that high school grads can only be employed at middle class wages in places that still have manufacturing or resource extraction, college grads can only be employed at middle class wages in a shrinking number of metro areas, etc. And if the localized housing shortages *are also* in the same place(s) as the localized job surpluses (or vice versa)… it results in lots of people stuck where they are with no chance for improving their lives, frustrating the hell out of them.
Again, let capitalism die, this would fade from history. It's pretty clear there isn't enough real growth anymore to support actual credit growth in the real economy leading to imbalance. It's a large reason why 2008 so important. The weakness in the western powers economic system began East Asia's push for the Red sea conspiracy and rebuilding the curtain. A debt liquidation is coming one way or another. Desert cities like Vegas and LA will get slammed and depopulation to cooler climates. Small rural Alabama towns will die out and disband. Florida bankrupt. The sports entertainment industry, heavily reduced as media moguls crash.
This board doesn't get it. Your living a illusion of capitalism that is more Bernie Lomax than anything else.
Post-capitalism will still be some form of capitalism. The most interesting area is the crossover/handover between capitalism and marxism (socialism), IMO.
Marxism????? Socialism was far earlier than that self hating fraud
I'm still waiting for the look at the housing starts chart from FRED. It's just one data point, but it's really obvious.
https://fred.stlouisfed.org/graph/?g=NdQX
It's kind of ironic that Kevin was the person (I believe) who pointed me to this chart about 10 years ago, showing that housing construction cratered after 2008 and didn't recover for at least a decade.
But more importantly, and this aligns with Erdmann's points, what that chart shows is that MAYBE there was a slight housing boom in the early aughts, but there was DEFINITELY a huge crash in housing construction following 2008 (if not beginning in 2006). It has MAYBE returned to the long term trendline as of this year. Maybe. Which, assuming it's moving towards an equilibrium point, should be a gentle curve rather than a straight line...
According to the US Federal Housing Finance Agency, New York and not California has the highest prices:
https://fred.stlouisfed.org/graph/?g=ONlS
and many other places are pretty high. But there have been differences between areas for a long time and Coastal California has been expensive for a long time. Most places are following a similar trajectory - bubble peaking in 2006, decline and then rise again to what is a new record for the area (I think this is true in real prices although I have not corrected all the curves in the graph). That is prices have been steeply rising everywhere since around 2011 and then went really crazy in 2021.
Kevin is still failing to explain the rise apparently everywhere since 2011 and especially the recent kick up, although that may be pandemic related. Are we going to believe Kevin or the invisible hand of the market which is saying that if prices are rising there must be a shortage with respect to demand? Kevin seems to think he can define shortage without taking account of how many people want to buy houses (where is that number?) or what they are willing to pay.
Again you may have to hit Max in the link to see the latest data.
I think it is helpful to think of housing supply and demand as purely a rent issue, and that informs prices. The lack of adequate supply is driving up rents and thus prices.
I think maybe that gets at your point, but I'm not exactly clear on what you're saying.
Homes are forms of capital and land that provide shelter as a service over time. Talking about housing affordability in terms of price is sort of like talking about food affordability in terms of the price of farm land. I think it can create confusion.
It's not just hot urban areas. There is a housing crunch in rural areas where housing is being converted from long term to short term rental. Ten years ago, there were lots of rentals one could afford on a $15 an hour minimum wage, that is, under $1000 a month. The median monthly rental is still around $1000 a month, but good luck finding a place to rent for that little. That statistic includes a lot of people who have long term rentals. The transition from affordable to unaffordable does not happen overnight.
One reason is that we are in a nice enough area near a national park, owners would rather rent by the week or even shorter term. This has driven up the asking rate for rents and the price of houses overall. Another reason has been increased baby boom retirement. We have a lot of people moving here from California or the I-5 corridor in Seattle who have sold their largely paid off houses and are now bidding up housing prices here. A third factor has accelerated this. COVID has encouraged people to move from urban areas out to rural areas. Once again, there is housing price leverage driving up local prices.
We know people living on the margin. Ten years ago, even five years ago, they could afford a place to live. They no longer can. One of them is living in a small camper and struggling month to month to find a place to park it for residential purposes. There's Hipcamp and the like for the more affluent, so slot rental prices are rising. The big local news is that a trailer park a bit west of here has tripled its rents, and the current tenants cannot afford them and have nowhere to go.
We know others in better shape financially who have left the area because they can't afford to buy a house here. This includes at least two in the skilled building trades, the kind of people who could deal with a fixer-upper. The cost of housing has made it even harder to recruit medical people, generally well paid sorts, to the local hospital. Rural hospitals and clinics have always been a hard sell, but the high cost of housing has eliminated one of the big selling points.
We aren't Vail, but we are facing the same problem. Worse, Vail could build its way out of it with subsidized worker housing. We can't. We don't have that kind of money out here. We turned housing from a necessity into a commodity investment, and we've eliminated local zoning rules on residential property. The finance sector makes its money by destroying marketplaces and leaves everyone else to deal with its mess.
Re: “Overall, there's no housing shortage, but there is a housing shortage in hot urban areas that people want to move to.”
Those areas are "hot" because that is where the jobs are. In theory, wages would rise until those employers moved elsewhere with cheaper housing and wages, but most companies hire people at a broad range of wages. Your high end, highly educated staff is not going to bleed from the high housing prices the way your low end, relatively less educated staff is going to. When exactly does an employer give up? When the toilets aren't getting cleaned? When they can no longer hire drivers? And then there's the issue of their customers. How far can they move from them?
If I understand the charts right, doesn't the rent chart say it all?
If housing, which is often the single most expensive item in a person's budget, plus, the one thing in the budget which is not fungible and also which one cannot do without, goes up to the tune of 11% more than inflation and we all know wages (especially low class wages) are not even keeping pace with CPI, I mean, that's it, right?
When someone says, we have a housing shortage or crisis, that's almost a slogan, for the longer thing, which is "housing is expensive, if we built more, then the law of supply and demand would render prices less expensive, so we have not built enough."
So when Kevin then wonders, "have we (built) enough?" Well, pretty soon its chart atopia..
I mean, to take a step back, its almost like as a country we don't even have a housing policy, the "policy" was - we have such a big ass country, just build some new, cheaper housing farther out. Now, we have reached the limit of that (as Europe already has done).
...the one thing in the budget which is not fungible
In the fundamental sense of needing a roof over one's head, absolutely not fungible.
But as to quantity and quality, very much so.
You can have more people contributing to rent/mortgage. You can move down the housing chain into smaller, less desirable units.
In modern societies, issues about housing are usually about quality of life, not basic need for shelter.
Pingback: Here’s a long, boring look at the housing market – Kevin Drum
I visited Erdmann's blog. My take is that, to some extent, he and Kevin are talking about different things. My cursory (haven't read his book) interpretation of Erdmann's thesis (it's unimpeachable from what I can see) is that the housing market that used to exist in America—that accommodated and even powered widespread, shared prosperity—no longer exists. Or is broken. We see the stark evidence of this in the abysmal rate of new housing inventory coming onto the market in our most productive cities.
Kevin focuses a lot on national aggregates, and, while this isn't completely wrong, implicit in his approach is the reality that the country has benefited from the national housing "safety valve" of the Sunbelt*. In recent decades metros like Orlando, Dallas, Phoenix, Atlanta, Charlotte, Las Vegas, Tampa, Houston and so on have accounted for a massively oversized share of new housing construction. While less NIMBYism isn't the only factor that powers the growth of such places, it pretty clearly has to play a major role. And that's because housing starts in metros with objectively higher wages—the sorts of places that used to power much of the country's growth because people wanted to move there for the opportunity—haven't kept up. (But the high housing prices in such cities—have you checked rents in the suburbs of San Francisco lately?—indicate there's ample demand for shelter there).
There's lot of other granular detail on his blog, including findings with respect to mortgage lending (it would appear stricter rules now make it much more difficult to supply new housing for the working class).
*And thankfully we at least have that safety valve. But not everybody can move, and people who already live in safety valve metros can face enhanced displacement pressures when lots of newcomers arriving mainly because they can't afford their former cities. And, needless to say, forcing an "artificially" higher share of population growth into less productive metro areas costs the national economy a great deal, and this in turn results in an America that is poorer, and thus less well-resourced to deal with national priorities.