Fast-growing Texas metros offer blueprint to handle U.S. housing shortage

Q. You write that America has been building too few homes for decades. How big is the shortfall?

We estimate that America in the 21st century has underbuilt by 6 to 8 million housing units. We created a pretty simple model predicting how many [housing units] we would need per year based on population growth. We assumed a long-term, very gentle downtrend in the number of people per household—more adult singles, fewer kids—and then made some assumptions about normal [housing unit] replacement.

If you fit that model from 1960 through 2000, it fits really, really well. It shows we produced enough housing for about 40 years. We then fell off the trend in a very discontinuous way with the financial crisis in 2008–09, after which we were way, way below what we needed.

Underbuilding translated to about a 20 percent [housing price] increase relative to incomes in the 21st century. Before that, prices and incomes mostly moved together over time. That's across the whole country. Home price appreciation varies enormously across states and metropolitan areas. There are places that had very little increase in home prices relative to incomes; some places had a lot more than 20 percent.

Q. When young people today say owning a home seems unattainable, is that a real thing?

It's very real. It's 20 percent real on average. And it's way more than that in a lot of the places where the young people wish to live. They're likely to gravitate toward relatively high-opportunity places, like growing metropolitan areas with rich job markets. Those places typically have had bigger-than-20-percent increases in home prices relative to incomes.

Q. What else are you saying about the demand side?

During the pandemic, there was clearly an enormous increase in the demand to live in some places, offset by low growth or even shrinkage in the demand in other places.

I think that we saw much greater than average increases in home price-to-income and rent-to-income in places that also experienced giant influxes of people—big upward inflections in the net inbound migration rates from elsewhere in the United States. People moved to places such as Dallas-Fort Worth, Austin, Houston, San Antonio, Raleigh, Nashville and Phoenix and several places in Florida.

You see an upward shock in the number of people moving in and also better-than-average or higher-than-average home price increases. I would call that telltale evidence of positive demand in those places; the demand shock was super powerful.

Places that were better than average typically at building new homes were overwhelmed. I think that's the basic story of Texas. The thing that's harder to say is whether there was truly a nationwide, positive demand shock for housing.

There's some anecdotal evidence that people did experience something of an upward demand shock for space maybe because working in a tiny, cramped bedroom was a little annoying to some people. That was anecdotal. It's not super easy to find that in the evidence.

I believe there was a kind of mini-speculative bubble in some places that has partly unwound now. Places that had the most evidence of that are the places that typically have seen retrenchment in prices in the past two years since interest rates started going up.

Q. How did we get into such a housing deficit?

We have two factors that kind of got superimposed on each other. One, there was a longer trend starting, some would say, in the ’80s toward localities and states just turning the screws, making it harder to build housing.

Notwithstanding, we got something of a building boom in the mid-2000s. Then we got a housing bust and the financial crisis [2007–09]. Out of the financial crisis, we got a whole witch's brew of problems. This was the second factor.

After the financial crisis, the banking sector was capital constrained as was, in general, the whole financial system. The federal government imposed a number of tests, a number of restraints on the mortgage market that some thought were going to result in a really dark period for mortgage origination. More specifically, the federal government determined that a lot of the problems that families had were actually the results of predatory, arguably, illegal decisions on the part of lenders, loan originators and the investment banks that packaged the securities.

The federal government basically said, “Well, we're not going to do that again. We're going to make it much, much harder for you to fool someone into getting a mortgage they can't afford,” which can pretty easily translate into, “We're just going to make it harder for you to originate new mortgages.”

All those things were factors, and I think they were kind of gone by the mid-2010s. Homebuilding rates didn't suddenly take a turn up. They just gradually got better and better over the course of the 2010s.

Q. You identified places that got it right in terms of keeping up with housing demand. Which communities are you referring to, and what did they do?

There is a tendency to say, “Well, if the home prices are high, or if new building is low, they must have bad policies in that state or that locality.”

That's not a very good argument because low building can reflect low demand, and high prices could reflect high demand without there being underlying policy differences with regard to housing supply.

We relied on predictive models essentially asking, “What would you expect across 250 metropolitan areas of America?” If places were far off those predictions, it raised the suspicion that they built way fewer homes than the model said they should have. It raised the suspicion that they might be places with exceptionally bad housing policies that make it really hard to build housing.

Likewise, if localities outperformed, we inferred that they were good, and then we tried to confirm that by actual observation of policies and whatever evidence that we could find. We mostly agree with the conventional wisdom: Big Texas metropolitan areas are way better than average at allowing housing to be built. Big California metros along the coast are exceptionally bad.

On the whole, we confirmed the conventional wisdom on which places do better and which places do worse. But there were surprises. For example, people observe that in New York City, apartments in Manhattan are really expensive. But we concluded that actually, New York was kind of a middling performer, so not a bad one.  When I lived in New York in the early ’90s, the home-price-to-income ratio was about the same as San Francisco and L.A.

Now, there's a huge difference. Home prices in San Francisco and L.A. are something like 50 percent higher relative to incomes than in New York City. They [New York City] actually had a whole lot of reform when Mike Bloomberg was mayor for three terms, reforms that were not matched on the West Coast.

I think there's also a sense that in the mountain and desert states, some places that grew pretty fast, like Denver and Phoenix, must be relatively housing-growth friendly. Here was another surprise. We concluded that they're actually not. They're below-average performers. It's just that we've had an aging population. [Those] people like natural amenities like golfing in the winter and hiking. There was a giant increase in the demand to live there—Denver, Phoenix and also smaller mountain towns like Boise. They didn't have especially good housing policies, but they still had pretty good growth in housing supply.

Q. What are the places that are doing things right? And what are they doing?

While there is no single magic bullet, we generally observe that these places allow construction of new apartment buildings and townhomes in their core areas.

Houston, for example, has probably the best environment of any big city in America for building townhomes because they did a minimum lot size reform starting in 1998. And then they expanded it in 2013 with very quantifiable results; there are a lot of studies of Houston’s success. Austin had cranes everywhere in the core city over the past 10 years or so.

What is less visible to those who live in the core city are the huge differences across metropolitan areas, what’s happening on the periphery, how much outward expansion there was and the form it took.

In Texas, we do suburban growth like it's going out of style, like virtually no place else in this country. Our fast-growing suburbs in Dallas–Fort Worth or Austin or Houston aren't Nirvana. It's not like they did every imaginable thing right. But, in general, they wanted to grow. They planned for growth. They built highways to grow. And they had reasonably supportive environments for most kinds of housing.

Texas metros have been kind of a growth paradise by the standards of American cities, despite their flaws. Places in Texas, almost across the board, while they have far-from-perfect policies, they have better policies in almost every imaginable policy category if you want more housing supply growth than, say, the West Coast.

Q. What about the costs of urban sprawl and other adverse environmental impacts?

We are basically saying that, generally speaking, it would be a good thing for the American people if in-demand metropolitan areas, which are experiencing growing populations, try to plan for intelligent, sustainable outward expansion. That means, for example, allowing a wide variety of housing types including townhomes and starter homes on relatively small lots as well as apartments.

It also means a high degree of mixing of uses and adequately providing for green space and, of course, adequately providing for infrastructure. We are generally saying, if thoughtfully done, a certain amount of outward expansion is, in fact, part of the answer. It's not the only answer.

It’s a challenge. Some Texas suburbs want to grow so badly that they're in effect cutting deals with developers that take so much tax revenue off the rolls for the long term that they're having a hard time even providing adequately for policing. We have a growth mindset in Texas, but we don't have all aspects of the execution fully down. The road infrastructure, the water— I mean, we've generally succeeded in Texas at building out these things ahead of growth.

Q. Urban planners frequently suggest that increased housing density, typical in most European cities, is the answer to housing supply and affordability. Why aren’t you convinced that density—more apartments and condos—is the answer?

It’s inconceivable that we will replicate major European cities on a large scale. Yes, individual neighborhoods, but not at really large scale to have it like central London or central Paris or something other than what we have in New York, which is its own distinct flavor.

Among other things, the population of the country isn't going to grow very fast. It’s hardly growing at all right now. In the future, it may actually shrink, some analysts think. So there just isn't going to be the demand or probably the appetite in terms of rules to try to promote a gigantic increase in density.

I really think it's important to note that density is an outcome, not a policy. In the real-world political economy circumstances in the United States, it is not possible for a federal, state or local government to say, “Let there be high density.” The Chinese Communist Party could do that, if they wish, by sweeping away several square miles in Beijing and ordering developers who are state owned to build what they [the party leaders] want. But that doesn't happen in this country, and I hope that it never does.

In Brooklyn, in Manhattan and in parts of San Francisco, there's a demand for distinct pockets of density, in particular walkable, cool places. When people get older, they want a little more space. It's just kind of the human way, it seems.

We don't say density is the answer. We say allowing a wide range of housing types in a lot more places—core city, suburban, traditional, downtown—is desirable. We are on the record in favor of policies that would allow greater density to emerge in places where there is a demand for it.

I think we'll see that happen. All over Texas in the big metropolitan areas, we've had the emergence of many relatively dense, walkable, mixed-use alternative downtowns. We've actually been pretty good at alternative downtowns in Plano and Frisco and Allen and McKinney and suburban locations all over the Austin, Houston and San Antonio areas.

Q. What are the lessons learned from your report in terms of addressing the housing shortfall in the future?

We have a long list of federal, state and local policy recommendations in the report. At the local level, we discuss specific policies that would allow for smaller lot sizes, permitting and parking reform, and building apartments in commercially zoned areas for mixed use developments.

There's also a decent amount of upside to be had, I would argue, from just allowing more innovation. People think of manufactured housing as an unsightly trailer home. Actually, if you look at images of manufactured housing in American cities, you'll see there's no shortage of communities where it has been allowed, where you have a smallish but very pleasant-looking home, which was more or less entirely built in a factory and plopped down off a truck.

You have the technologies that are nascent, like 3D printing and so on, that ought to be given a shot to see what they can do.

We have a bit of an optimistic case. Some people think there is no reversing this [deteriorating housing affordability], but that really need not be the case.

We make a quantitative case that says that if you take the kind of policy package as it exists in 25 leading sunbelt metropolitan areas that we identify in the report—Dallas–Fort Worth, Austin, Atlanta, Charlotte, Raleigh, Nashville and some others—we would have several million more homes than we do, something on the order of 5 or 6 million, and we would have considerably lower home prices.

We're now in a period of very low population growth and housing reform. If we don't totally mess it up, the net combination of slow demand growth and a somewhat better supply environment ought to actually mean that home prices go up less than people's incomes [go up] over the next 10 or 20 years.

This is an edited and abridged version of a conversation available on the Southwest Economy Podcast.