Almost half of American counties have shrinking population

It's a problem

I’m spending a lot of time in Chicago this month, and it’s been a good opportunity to learn more about the area. I’ve been thinking in particular about the problems induced by the stagnant (and now falling) population in the city and most of the nearby suburbs. Unfortunately, it can be somewhat challenging to get people to think clearly about these problems because they’re so deeply intertwined with Chicago as an entity of partisan and ideological struggles. So it’s worth looking at the map of county-level population decline across the entire U.S. to get a sense of its scope.

Note that almost all of West Virginia, Louisiana, and Mississippi are losing people. In Ohio, which has taken a right political turn over the last 15 years and enacted a lot of conservative policy ideas, their governor is rightly worried about population decline to the point where he’s considering changing the state tourism office’s mission to “include not only attracting visitors to Ohio, but also workers, residents, and students.”

It’s a perfectly unobjectionable idea, but I think state and local leaders who are worried about population decline really ought to look at the whole map and consider reading Matthew Yglesias’ seminal 2020 book “One Billion Americans.”

The point about looking at the whole map is that if your community is growing much more slowly than the national average, that suggests either a problem of underlying policy fundamentals or else a failure of marketing. But if your community is growing just slightly slower than average, that doesn’t necessarily indicate anything at all. Population shrinking causes various problems, and it used to be sufficiently rare that shrinkage corresponded very clearly with being way below average. But the issue in the contemporary United States is that the average growth rate is very low. Fully 47% of counties lost population between July 2021 and July 2022.

That means the only way to avoid the negative zone is to be above average. But this isn’t Lake Wobegon — we need to raise the average rate of growth so a larger share of places can have positive growth.

When homes aren’t worth replacing

One big problem with population decline is that it makes houses too cheap. I know, I know, I’m normally talking about how it’s bad when regulatory barriers make houses too expensive. But it’s also bad for houses to be too cheap.

Why? A concept I learned about years ago from urban economist Issi Romem and that I wish was in wider circulation is the idea of the market price of housing vs. the replacement cost of a house. The market price is obvious: if you put your house up for sale tomorrow, how much could you get for it? The replacement cost is what it would cost if a tornado demolished your house tomorrow and you had to rebuild it from scratch.

You would normally expect the market price of a house to be somewhat higher than its replacement cost. Some of that is because when someone builds a new house and tries to sell it, he wants to do so at a profit and he also needs to cover the opportunity cost of having his capital tied up during construction. The other piece, though, is that when you buy a house, you’re also buying the land the house is located on. Some parcels of land are very cheap but others are expensive. An average-sized suburban home with a modest replacement value can have a high market price if it’s located in Silicon Valley or Santa Monica or the suburbs west of Boston or in Bethesda, Maryland.

But in a less regulated market, the land prices would matter less. Here’s a 2,700-square-foot two-story home sitting on a 5,600-square-foot lot not far from the Clarendon Metro Station, but just far enough that the area is strict single-family zoning. It’s a nice house, but the $1.7 million price is mostly due to the expensive land. Absent that zoning rule, you might redevelop a parcel like that as a six-story building featuring ten 2,000-square-foot units (two on each floor) plus a 4,000-square-foot penthouse. The total value of the whole parcel would be much larger, but the individual units might be 50% of the price of the old house at 75% of the size. That redevelopment is possible because $1.7 million is so much higher than the replacement value of the house — regulation ensures that the price of housing is dominated by the price of land.

In a scenario of population decline, though, the market value of a house falls below the replacement value of the structures. You can see this by looking at statistics, but it’s pretty easy to eyeball where it’s taking place because the signature is a neighborhood with more vacant lots or vacant buildings than ongoing construction projects. Huge swathes of Detroit are, infamously, like this. But it’s also true of Cleveland and St. Louis and Milwaukee and Baltimore and other cities. Abandoned or vacant buildings are a source of blight — negative amenity value to the neighborhood. And each city with a significant amount of vacant property has its own policy apparatus for dealing with it and its own local discourse about the merits of that apparatus and whether the problem should be handled some other way. In years of reading about this, though, I’ve never heard of a city that has a magic formula to deal with vacant lots and blight. What I have seen is the vacant building problem vanish in my own neighborhood in D.C. due to robust market demand for housing. I’ve also seen in Chicago that a city can be capacious enough to support a lot of construction activity in the West Loop, even while significant swathes of the South Side look a lot like Cleveland.

Mirror universe real estate economics

In addition to the negative amenity value of blight, it’s worth thinking more broadly about the economic mirror universe that exists when market prices fall below replacement prices.