Question
Why are mid-size American cities so much cheaper than European equivalents
...so I keep running into this comparison and it bothers me because people use it to argue the US has a housing "supply" problem when really we have a *demand* problem, or maybe a preference problem, and those are different things.
Take Buffalo versus Lyon. Both around 250k in the city proper (though metro areas differ). Lyon apartments run €10-12k per square meter. Buffalo is maybe $2-3k equivalent. People reflexively blame zoning, and sure, Buffalo's zoning is stupid and restrictive. But Lyon has its own planning constraints. The real gap is that Lyon is extracting genuine economic value from being in a wealthy regional economy (Paris is 2.5 hours away, Zurich closer). Buffalo's hinterland is... Ohio and Western PA. The jobs that would make people want to pay Lyon prices just aren't there, or they were and got hollowed out 20-40 years ago. You can't zoning-reform your way out of that. Hamburg's similar—it's extracting value from being the main port for Central Europe, from having banks, from being genuinely useful to the German economy. Cleveland isn't, and hasn't been since the 70s.
What's weird is we talk about this backwards. We act like housing is expensive *therefore* we can't build enough. But actually housing got expensive in places where demand exploded (SF, NYC, Boston) and we couldn't build fast enough. Buffalo and Cleveland got cheap because demand *fled*. The building isn't the bottleneck there, it's getting people to want to move in. Loosen zoning in Buffalo and you don't get Parisian prices, you get... still-cheap housing but maybe better buildings. Which would be fine! But it's not a supply problem dressed up as a zoning problem.
The uncomfortable question is whether mid-size US cities can actually regenerate demand without becoming interesting enough to compete with coastal metros *and* without the geography/proximity advantages of Hamburg or Lyon. I don't know the answer. But I'm skeptical that it's a zoning thing.
4 comments
Log in to comment.
You're right that the framing matters here, and you've identified something real: there's a difference between "we can't build enough to meet demand" and "we built plenty but nobody wants to be here." But I think you're underselling how much zoning actually shapes which problem you end up with.
The thing is, demand isn't a fixed quantity that either exists or doesn't. It's partially revealed preference, sure, but it's also constrained by what's actually *possible* to build. I watched this play out tangentially in a company relocation decision—we were looking at secondary cities and the CFO would pull up maps, see that yes, technically housing existed and was cheap, but then you'd actually visit and realize the housing stock was either 1970s suburban McMansions or genuine urban neighborhoods that zoning had basically frozen in place thirty years ago. The demand wasn't "we don't want to move to Buffalo," it was "we don't want to live in the only housing that's legal to build there." And yeah, that's partly about jobs, but it's also self-reinforcing. You can't attract the knowledge workers you need if the only rentals in the neighborhood are zoned single-family houses in suburbs that require a car.
Buffalo and Lyon are different animals economically, agreed. But the fact that Hamburg maintains price pressure despite not being as wealthy as Swiss cities suggests that having *legal room to build mixed neighborhoods where people can actually live* matters independently of whether you're extracting value from finance or ports. Your supply problem and demand problem might not be as separate as they look—at least at the margin where a city could actually move the needle.
You're describing a real distinction but I'd push back on the demand versus supply framing, or at least complicate it. In logistics we see this constantly: the bottleneck moves. You fix one constraint and hit another. But the constraints are real at each stage.
Buffalo's cheap because nobody wanted to move there for decades. That's true. But that doesn't mean zoning doesn't matter now—it just means zoning is not the *only* thing that matters, and it's probably not the binding constraint today. If you actually got a critical mass of remote workers or a university expansion or something that made Buffalo marginally more attractive, the zoning would then become the thing that stops conversion from happening fast enough. I've watched this in smaller metros that caught some momentum. You need both the pull (jobs, reason to be there) and the capacity to absorb quickly when the pull arrives. Most mid-size cities are set up so that when demand twitches, the physical friction takes it out again.
The harder problem is you're right that most mid-size cities don't have a natural extraction point like Hamburg's port status. That's structural and doesn't get fixed by land-use policy. But that's almost orthogonal to the zoning question. They're different problems requiring different solutions. Zoning reform isn't a panacea for demand destruction, but I've also never seen a city regenerate without being able to build quickly once something starts working. It's not sufficient but it does seem necessary.
I'd push back on the premise a little—I think you're setting up a false binary between "supply problem" and "demand problem" when the real pattern is uglier: places get cheaper, that cheapness attracts different kinds of demand (remote workers, cost-conscious startups, people priced out elsewhere), but the local institutions and culture lag by 5-10 years, so you get this weird liminal phase where housing is affordable but the reasons to move there are still fuzzy.
My prediction is you see 3-4 more mid-size cities crack this in the next 8-10 years, but not through zoning reform or some grand revitalization plan. It'll be accidental—a few tech companies hire remote and pay market rates, some subset of that cohort actually moves to wherever (because why pay SF rent if you're working async), that creates a small density of people with disposable income, which attracts restaurants and cultural stuff, which makes it actually *interesting* rather than just cheap. Austin already did this in slow-mo. Pittsburgh's probably closest now. But it requires external capital flowing in first, not zoning changes creating the conditions for it. The zoning stuff matters for the *next* phase, after people already want to be there.
What I'm skeptical of is whether that happens in Buffalo or Cleveland specifically. They've been "about to regenerate" for 30 years. At some point the gravity well is just too deep—the talent that would fix the schools and local government left, so the schools stay bad, so talented people don't move back. It's not a housing problem, it's a network effect problem, and I don't know what actually breaks that cycle.
I think you're conflating two separable things though, and that's where the supply/demand framing gets muddy. You're right that Buffalo's demand problem is real and probably unfixable by zoning alone. But zoning *also* suppresses whatever residual demand does exist, which is a different failure mode entirely.
The clearer way to think about it: assume Buffalo genuinely can't compete with Lyon on fundamentals. Fine. But does Buffalo's current zoning even allow it to capture the subset of people who *would* move there if housing were cheap enough and the neighborhoods didn't look like 1970s suburban sprawl? Because right now you can't build dense, mixed-use neighborhoods in most of Buffalo's buildable areas—not because nobody wants to, but because the code forbids it. You end up with people who might've accepted Buffalo-on-the-cheap instead choosing Columbus or Pittsburgh, which at least have *some* density in the core. So you're losing demand you could theoretically capture, on top of losing the demand you never had a shot at.
The regeneration question you're asking is real and hard. But I'd separate "can zoning reform fix Buffalo's fundamental economic position" (probably no) from "are Buffalo's zoning rules currently blocking whatever marginal demand recovery is possible" (almost certainly yes). They're both true. The first doesn't make the second irrelevant.