Daniel sent us this one — he's been thinking about the supply side of city property markets, and he's asking a question that sits right at the intersection of planning and politics. Basically, if you let the free market run wild, you get imbalance. All housing goes to investors, renters are squeezed out. All residential, no commercial. All luxury, nothing affordable. So the question is, how have different governments actually intervened to keep a balanced portfolio of land use? Not just zoning in theory, but the real mechanisms.
This is one of those questions where the answer is basically, everybody intervenes, they just pretend they don't, or they intervene in completely different ways and call it different things. And the outcomes are wildly uneven.
Where do we start? The obvious poster child has to be Singapore.
Singapore is the most legible example because they just own it. The government controls about ninety percent of the land. They acquire it through compulsory purchase, they lease it out, they decide what gets built where. And the Housing Development Board, the HDB, isn't just a housing agency — it's essentially the central planning authority for the built environment of an entire country.
They build mixed-use developments as the default, right? It's not a special program.
The standard HDB town is designed around what they call the precinct concept. You have a cluster of residential blocks, and at the center of every precinct there's a commercial hub — a hawker center, shops, a supermarket, medical clinic, sometimes a library. All within walking distance. The residential and commercial are planned together from the start. It's not like someone builds the housing and then hopes a grocery store shows up.
The government is essentially the developer, the landlord, and the planner all at once. And they're enforcing the mix by fiat.
Right, but here's what's interesting — they also use price signals within that framework. The commercial spaces in HDB estates are sold on thirty-year leases, and the buyers bid for them. So there's a market element. But the government controls the supply. They decide how many shop units exist per thousand residents. They're not leaving it to chance.
Which solves the problem Daniel's getting at — if you leave it purely to the market, you get what's most profitable, not what's most needed. A developer will build another bubble tea shop before they build a medical clinic.
The Singaporean approach has a name for this philosophy — they call it "allocative efficiency with social outcomes." The market allocates within boundaries the state sets. It's not socialism in the old sense, it's more like...
Sounds like a playlist for people who think Adam Smith was too edgy.
It kind of is. And the results are hard to argue with. Something like eighty percent of Singaporeans live in HDB flats. Homeownership rate is around ninety percent. Commercial spaces are distributed across the island so you're never more than a few minutes from daily needs. It's not perfect — there are critiques about conformity, about the heavy hand of the state — but as a mechanism for balanced land use, it's remarkably effective.
Singapore is the extreme case. Total state control over land supply, market mechanisms within that. What's the opposite end? Who's tried to get balance without owning all the land?
Japan is the fascinating counterexample. Japan intervenes very lightly in land use compared to almost any other developed country. Their zoning is famously permissive. There are twelve basic zone types, and they're structured as a pyramid of allowed uses rather than a list of prohibited ones.
Explain that — the pyramid thing.
In most countries, zoning says "this area is residential, you cannot build a shop here." In Japan, each zone says "this zone allows everything up to this level of intensity." The lowest zone is exclusive low-rise residential. The next zone up allows small shops and offices. The next allows larger commercial. And so on, all the way up to industrial. The key is that everything allowed in the lower zones is also allowed in the higher zones.
You can't have a factory next to a house, but you can have a small shop next to a house even in a mostly residential area.
And the result is that Japanese cities organically develop this fine-grained mix of uses. You walk down a street in Tokyo and you'll see a house, then a tiny factory, then a convenience store, then an apartment building, then a restaurant. It's all legal by default.
Which is the polar opposite of Euclidean zoning where you have the residential zone, the commercial zone, the industrial zone, and never the three shall meet.
Named after the village of Euclid, Ohio, by the way, not the mathematician. Though I've always found that disappointing.
I assumed it was the mathematician for years.
But yes, the US model of single-use zoning came out of a 1926 Supreme Court case, Village of Euclid versus Ambler Realty, which upheld the right of municipalities to segregate land uses. And the reasoning was explicitly about protecting single-family homes from what the court called the "nuisance" of apartments and commercial uses.
Protecting homes from apartments. The polite way of saying "keeping certain people out.
That was absolutely part of it. The history of American zoning is deeply entangled with racial and class exclusion. But even setting that aside, the practical effect was to bake imbalance into the system. If you legally require housing to be separate from jobs and shopping, you guarantee car dependency and you guarantee that market forces alone won't produce a balanced mix.
Japan gets balance through permissiveness, and Singapore gets it through control. Those are the two clean examples. What about the messy middle?
Most countries are in the messy middle, and they use a grab bag of tools. Let me run through a few. The first and most common is inclusionary zoning. That's where the government says, if you want to build a residential development of a certain size, you must include a percentage of affordable units. Or you can pay a fee instead. It's essentially a tax on market-rate development to cross-subsidize below-market units.
Which places do this well?
Montgomery County, Maryland, just outside DC, has been running one of the oldest inclusionary zoning programs in the US since the 1970s. They require twelve and a half to fifteen percent of units in new developments to be moderately priced. And they've produced something like fifteen thousand affordable units over the decades. It's not nothing. But it's also a drop in the bucket relative to total housing need.
The critique is that it's essentially a tax on new housing, which makes new housing more expensive for everyone else, right?
That's the economists' critique. The counterargument is that in high-demand markets, the land value is so high that developers can absorb the cost and still profit. The truth is probably somewhere in between — it works better in hot markets and can stall development in weaker ones.
What's the next tool?
This is where the government says, we'll let you build more units than zoning normally allows if you include certain public benefits. Affordable housing, public open space, community facilities, whatever the city wants. It's a negotiation rather than a mandate.
Bribery, but make it urban planning.
Incentive-based land use regulation, thank you very much. And it's incredibly common. Vancouver has used density bonuses extensively to fund community amenity contributions. Developers get extra floors, the city gets daycare centers and public art and affordable housing. But it's been criticized for being opaque — every project gets negotiated separately, and the public doesn't always know what was traded for what.
It's flexible but not transparent.
And that brings us to the third tool, which is the most direct — land banking. The government simply buys land, holds it, and releases it for development according to public priorities rather than market pressures.
This is what Israel does, isn't it? The Israel Land Authority?
Israel is actually an interesting hybrid. About ninety-three percent of the land is owned by the state or by the Jewish National Fund, and it's leased out on long-term leases, typically forty-nine or ninety-eight years. The Israel Land Authority controls the release of land for development. In theory, this gives the government enormous power to shape land use. In practice, it's been messy.
The ILA has often acted more like a revenue-maximizing corporation than a planning agency. It auctions land to the highest bidder, which drives up costs. And the planning system is separate from the land ownership system — so you have local planning committees approving zoning and the ILA releasing land, and they don't always coordinate well.
You have state ownership of land but market-driven release of it. Which doesn't necessarily produce balance.
It produces revenue for the state, but not necessarily the right mix of uses. There's been a lot of criticism, especially around housing affordability, that the ILA treats land as a financial asset rather than a public resource.
Which brings us to the fourth tool, I'm guessing — something that directly controls use rather than ownership.
Use it or lose it provisions. Vacant land taxes, land value taxes, penalties for holding undeveloped land. The idea is to discourage speculation and hoarding. If you own a prime development site in the middle of a city and you're just sitting on it waiting for the value to go up, the government steps in and says, develop it or pay us.
Who does this effectively?
Taiwan has a land value increment tax that captures the increase in land value when property changes hands or when zoning changes increase value. It's designed to prevent speculation and to return unearned value increases to the public. Australia has a version of this in the Australian Capital Territory, where all land is leasehold and there's a betterment tax on value increases from rezoning.
Does it work?
It doesn't solve everything. The Australian system has been credited with keeping Canberra's development relatively orderly, but it hasn't prevented housing affordability problems. No single tool does. Which is really the core insight here — balanced land use isn't achieved by one mechanism, it's achieved by layering multiple interventions that push in the same direction.
Let's pull back and talk about what "balance" actually means. Daniel's prompt framed it as ensuring that different interests don't monopolize the available stock. But who defines what the right balance is?
That's the political question underneath the technical question. In Singapore, the government defines it. In Japan, the market largely defines it within a permissive framework. In most Western countries, it's negotiated through the planning process, which means it's defined by whoever shows up to planning meetings.
Which is usually existing homeowners who don't want anything to change.
The dreaded incumbent bias. The people who already have housing, who already have commercial services, who already have parks — they have every incentive to block new development that might change the character of their neighborhood or reduce their property values. And they show up. Renters, future residents, businesses that might want to locate there — they don't have a seat at the table.
This is the fundamental asymmetry in land use planning. The costs of development are concentrated and visible — construction noise, traffic, shadows. The benefits are diffuse and future — more housing, lower prices, more services. The concentrated interests always organize better.
It's worse than that. Because even when cities try to plan for balance, the timeline problem makes it nearly impossible. A city might plan for a certain ratio of residential to commercial. But the residential gets built first because it's more profitable. Now you have thousands of new residents and no services. The commercial development was supposed to follow, but the market has shifted, or the developer pulled out, or the financing dried up.
You get bedroom communities that were supposed to be mixed-use.
And the reverse happens too — cities approve massive office developments in what they hope will become a live-work district, but the residential never materializes, and you get a dead zone after five PM.
Which leads to the phenomenon of the "planned community" that's really just a residential subdivision with a single sad strip mall at the entrance.
The architectural equivalent of a participation trophy. "We zoned for commercial." Yes, and the commercial that showed up is a vape shop and a struggling dry cleaner.
What's the actual mechanism that forces the timing to work? Because Singapore can do it — they build the housing and the hawker center and the school all at the same time.
Singapore can do it because they're the developer. They control the entire pipeline. In a market system, you're relying on private developers to coordinate, and they have no incentive to coordinate with each other. The developer building the apartments doesn't benefit from the grocery store that opens next door — not directly, not in a way that affects their pro forma.
Unless they're the same developer building both.
Which brings us to master-planned communities and the rise of the mega-developer. This is the private-sector version of Singapore. A single entity buys a huge tract of land, gets it rezoned, and builds out a complete community — residential, commercial, parks, schools, the whole thing. Because they're capturing the value of the entire development, they have an incentive to get the mix right.
The canonical example is...
There are many, but I'd point to something like Reston, Virginia, which was one of the first planned communities in the US, built in the 1960s. The developer, Robert E. Simon, explicitly designed it around the idea that people should be able to live, work, and play in the same community. It had a town center, mixed housing types, preserved open space. It was genuinely visionary for its time.
Today Reston is generally considered successful. It's got a real urban core, it's walkable, it's got a mix of uses. But it's also expensive, and the original affordable housing component has eroded over time. And it took decades to fully build out. The master-planned approach works if you have a patient developer with a long time horizon, which is not how most development finance works.
Most developers want to build, sell, and get out. They're not in the business of managing a community for thirty years.
And that's the fundamental tension. The market for land development is optimized for short-term returns, but balanced communities require long-term stewardship.
Let's talk about the European approach. I know Germany does something different.
Germany has a fascinating system. They use what's called "binding land-use plans" at the local level, but they also have a strong tradition of municipal land policy. Many German cities actively buy land, hold it, and release it according to public priorities. It's not as centralized as Singapore, but it's much more interventionist than the US.
They have this concept of "socially just land use," right?
The Munich model, they call it — "Sozialgerechte Bodennutzung." The city of Munich developed this in the 1990s. When land is rezoned for development, the city requires that at least thirty percent of new housing be affordable, and the developer has to provide the social infrastructure — schools, daycare centers, parks — at their own cost. In exchange, the developer gets the certainty of a clear planning process.
It's inclusionary zoning plus infrastructure requirements, bundled into a social contract.
It's been remarkably effective. Munich has built tens of thousands of affordable units through this mechanism. It's not cheap — Munich is still one of the most expensive cities in Germany — but without this policy, it would be dramatically worse.
The key is that the city has the legal authority to impose these conditions because they control the zoning. The rezoning is the leverage point.
The landowner comes to the city asking for a zoning change that will multiply the value of their land. The city says, fine, but a third of the value increase comes back to the public. That's the deal.
Which is essentially a betterment tax, but negotiated project by project rather than applied as a standardized rate.
Yes, and that project-by-project negotiation is both a strength and a weakness. It allows flexibility, but it also creates room for political dealmaking and inconsistency. The German system works partly because there's a strong administrative culture of consistency. In a place with weaker institutions, the same mechanism could become a corruption vector.
We've covered Singapore's total control, Japan's permissive zoning, American inclusionary zoning and density bonuses, land banking in Israel, use-it-or-lose-it taxes in Taiwan and Australia, the Munich model in Germany, and master-planned communities as a private-sector alternative. What haven't we talked about?
Transfer of development rights. This is a clever mechanism that a lot of people haven't heard of. The idea is that you separate the right to develop from the land itself. If the city wants to preserve a historic district or farmland, they can say, you can't develop here, but you can sell your development rights to someone who wants to build more densely somewhere else.
The owner of the preserved land gets compensated, and the developer in the growth area gets to build taller.
New York City has used this extensively to preserve theaters in the Theater District and to protect historic buildings. The owner of a landmarked low-rise building sells their unused air rights to a developer building a skyscraper a few blocks away. The landmark is preserved, the developer gets extra height, and the landmark owner gets paid.
Which solves the political problem of preservation — instead of just saying "you can't develop your land" and destroying the owner's property value, you give them a financial exit.
It's a market-based solution to a regulatory problem. But it's complicated to administer, and it only works in hot markets where there's demand for the transferred rights. You can't do this in a declining city where nobody wants to build anything.
Let's synthesize. Daniel's question was essentially, how do governments ensure a balanced portfolio of land use without the market either underproducing necessary uses or overproducing the most profitable ones. And the answer seems to be that there's no single clean solution. It's a portfolio of interventions, and the mix depends on the political culture, the strength of institutions, and the local market conditions.
I think that's right. But I'd add one thing that cuts across all of these mechanisms — the importance of flexibility over time. A city's needs change. The balance that made sense in 1990 doesn't make sense in 2026. The most successful systems are the ones that can adapt.
That's where zoning often fails, right? Because once you etch a land use map into law, it becomes incredibly hard to change. Every change requires a political fight.
The Japanese system handles this better than most because the zoning categories are broad and permissive. If demand shifts from office to residential, developers can respond without needing a rezoning. In a Euclidean system, you're stuck with what you zoned for until you can muster the political will to change it, which might take a decade.
There's a deeper point here about the relationship between planning and markets. The standard libertarian view is that planning is distortion and markets produce optimal outcomes. But land is different from other goods. It's finite, it's immobile, and its value is largely created by public investment — roads, transit, schools, parks.
The Henry George point. Land values are socially created, so society has a claim on them.
If the government builds a new subway line, the land near the stations suddenly becomes much more valuable. The landowner didn't create that value — the public did. So there's a strong argument that some of that value should be captured for public purposes, including balanced land use.
That's exactly what the betterment taxes and inclusionary zoning mechanisms are trying to do, however imperfectly. They're saying, the public created this value through infrastructure investment and zoning changes, so the public should get something back.
Which brings us to the question of who actually pays. Because in theory, these mechanisms tax landowners or developers. In practice, a lot of the cost gets passed through to buyers and renters.
The incidence question. It's hard to know. In a very hot market where supply is constrained, developers have pricing power and can pass costs through. In a cooler market, they absorb more of the cost. The empirical evidence is mixed, and it varies by city and by mechanism.
The honest answer is, we're not entirely sure who bears the cost of these interventions.
The honest answer is often "it depends" and "the evidence is mixed." That's not satisfying, but it's better than pretending we know for certain.
Spoken like a true retired pediatrician who's read too many papers.
There's no such thing as too many papers.
There absolutely is, and you've read all of them.
I'm going to take that as a compliment.
It wasn't not a compliment. So let me try to pull this together into something useful. If I'm a policymaker and I want balanced land use, what's the checklist?
First, decide how much control you're willing to exercise. Are you Singapore, willing to own the land and direct development? Are you Japan, willing to set broad permissive rules and trust the market? Or are you somewhere in the middle?
Most places are in the middle.
Most places are in the middle. So then you layer tools. Inclusionary zoning for affordable housing within market-rate developments. Density bonuses to incentivize public amenities. Some form of land banking or strategic land acquisition so you're not entirely at the mercy of private landowners. Use-it-or-lose-it provisions to discourage speculation. And ideally, a betterment tax or infrastructure levy to capture some of the value that public investment creates.
You build in flexibility so the system can adapt as needs change.
You accept that no system is going to produce perfect balance. You're managing tensions, not solving them permanently. The ratio of housing to commercial that works today might not work in twenty years. The demographic profile changes. The economy changes. Climate change reshapes where people want to live. You need mechanisms that can respond.
It's less about achieving balance and more about building a system that can rebalance itself.
That's a much better way to put it. The goal isn't a static equilibrium. It's a dynamic system that can adjust.
Which is why the Japanese approach is interesting even if you wouldn't adopt it wholesale. The permissiveness is the flexibility mechanism. When anything can be built anywhere within broad parameters, the market can rebalance quickly. The downside is you get less control over the outcome.
The Singaporean approach is the opposite — total control, but it requires a level of state capacity and political consensus that most countries don't have.
Probably don't want. There's a reason Singapore is often described as a benevolent dictatorship with a housing policy.
The "benevolent" part is doing a lot of work in that sentence.
I'm not sure I'd want to live in a country where the government decides exactly what mix of shops I get in my neighborhood. But I also don't want to live in a country where the market decides I get nothing but vape shops and payday lenders.
That's the tension. The libertarian impulse says let the market decide. The planner's impulse says the market will get it wrong. The truth is that markets are very good at some things — responding to demand signals, allocating capital efficiently — and very bad at others — providing public goods, accounting for externalities, planning for the long term.
Land use sits right at the intersection of all of those. It's a private good with massive public externalities. It's a market commodity that's also a fundamental human need. It's an investment asset and a place to live.
Which is why it's so politically charged. Every intervention in land use is simultaneously an economic policy, a social policy, and an environmental policy. You can't disentangle them.
That's why the mechanisms vary so much by country. The political culture determines what's acceptable. In Germany, requiring developers to build schools is normal. In parts of the US, it would be called socialism.
Though even in the US, the reality is more complicated. Texas, of all places, has some of the most permissive land use regulations in the country, and Houston famously has no zoning code at all. But Houston still has land use controls — they're just done through deed restrictions and private covenants rather than public zoning.
It's not that Houston is unregulated. It's that the regulation is privatized.
And the outcomes are mixed. Houston builds a lot of housing and is relatively affordable for a major city. But it's also sprawling, car-dependent, and has significant environmental problems. The private covenants don't produce balanced communities — they produce whatever the developers think will sell.
Which circles back to Daniel's original concern. If all housing is built and sold to investors, there's none left for renters. The market, left entirely to itself, doesn't necessarily produce the mix of tenures and uses that a functioning city needs.
That's the case for intervention. Not because planners are smarter than markets, but because the market for land has structural features — immobility, monopoly power, externalities, public investment creating private value — that mean unregulated outcomes won't be socially optimal.
The question isn't whether to intervene. It's how, and how much, and with what tools.
With what humility. The worst planning disasters come from overconfidence — the belief that we can design the perfect city from scratch and it'll work exactly as intended. The best planning acknowledges uncertainty and builds in flexibility.
The city as a garden, not a machine.
That's Jane Jacobs, essentially. She argued that cities are complex systems that emerge from the interactions of many actors, and that top-down planning that ignores that complexity is doomed to fail. But she wasn't anti-planning — she was anti-bad-planning. She wanted planning that worked with the grain of how cities actually function.
The grain of how cities function is mixed-use, fine-grained, adaptable, and human-scaled. Which is what the best interventions try to enable, whether through Singaporean master-planning or Japanese permissiveness.
The mechanisms are different, but the goal is the same. Create the conditions for diverse, balanced, livable places.
To answer the prompt directly — governments have handled this through a spectrum of interventions, from direct state control of land to light-touch zoning reform, with most countries somewhere in between. The specific tools include inclusionary zoning, density bonuses, land banking, betterment taxes, transfer of development rights, use-it-or-lose-it provisions, and master planning. None of them work perfectly alone. The successful approaches layer multiple tools and build in flexibility.
The unsuccessful approaches tend to share a common flaw — they're static. They lock in a particular vision of balance and can't adapt when conditions change. The cities that get this right are the ones that treat land use policy as an ongoing process, not a one-time plan.
Which is deeply unsatisfying if you're looking for a clean solution.
The clean solutions are the ones that fail most spectacularly. The messy, iterative, context-specific approaches are the ones that actually work.
The sloth approach to planning.
I was waiting for you to work that in.
Slow, deliberate, adaptable.
I don't think naps are part of the planning framework.
They should be. A well-rested planning commission is a wise planning commission.
I'm not sure the evidence supports that.
I'll find a paper. Give me time.
I'm sure you will. And now: Hilbert's daily fun fact.
Hilbert: The Outer Hebrides island name "Barra" derives from Saint Finbarr of Cork, who established a monastery there in the sixth century — though Finbarr himself may never have visited the island, and the name probably transferred from a now-vanished chapel on a smaller isle nearby.
The island is named after a saint who might never have set foot on it.
The patron saint of plausible deniability.
This has been My Weird Prompts. Our producer is Hilbert Flumingtop. You can find every episode at myweirdprompts dot com or wherever you get your podcasts. If you enjoyed this, leave us a review — it helps. We'll be back next week.
