Alain Bertaud
What this covers
Alain Bertaud and Tyler Cowen discuss how cities function as markets, with Bertaud arguing that urban dysfunction arises not from density or market forces but from regulatory distortions and property-rights dilution. The conversation ranges across planning philosophy, specific cities (Shenzhen, Jakarta, Brasilia, New York, Paris), and mechanisms of urban change. Bertaud contends that NIMBY emerges when regulation reduces what owners can do within their own lots while granting them power to block neighbors' development—a contradiction that paralyzes cities. He frames affordability as a supply problem, solvable by removing floor area ratio and minimum-size mandates rather than through rent control or inclusive zoning. Roads, by contrast, demand different logic: since urban road supply is fixed, efficient use requires congestion pricing by time and location, not expansion.
Bertaud draws historical comparisons—examining how Marx's labor theory of value crippled socialist cities' ability to use land prices for redevelopment, or how New York's grid established Property Rights as Symmetry of Information that prevented political manipulation. He analyzes Deng Xiao Ping's Shenzhen not as top-down design but as a functioning labor market that attracted migrants, and dismisses Paul Romer's vision of creating many charter city schemes as implausible—great cities exist where geography permits. He challenges planners to see cities as purposeful rather than haphazard, and argues that when housing serves as the primary retirement vehicle, homeowners resist density regardless of actual congestion costs. Throughout, he applies Markets as Gravity—inescapable forces to be understood and accommodated rather than denied.
Bertaud argues that cities are living markets whose vitality depends on letting prices, mobility, and clear property rights work, and that most urban dysfunction stems from regulation (zoning, rent control, NIMBY) and political distortions rather than density or markets themselves.
- NIMBY freezes cities by diluting property rights inside lots while creating negative property rights over neighbors' land
- Affordability is solved by increasing supply (removing floor-ratio and minimum-size rules) plus fast transit, not by rent control or inclusive zoning
- Roads, unlike housing, are fixed in supply within a city, so congestion must be solved by pricing demand rather than expanding supply
Cities live through constant change; NIMBYism and freezing development kills urban vitality by preventing adaptation to new people and needs.
- Cities are alive because they constantly change; NIMBYism freezes a city in place, which kills it like putting a human being in a straitjacket, so resisting change ultimately destroys urban vitality.
“Cities are alive because they change constantly. As soon as you freeze a city — and this is what NIMBY is all about, freezing cities — the city dies, like a human being that you tied in a straitjacket.”
- Setting building regulation at the street or neighborhood level hands control to existing residents who, by definition, oppose anyone wanting to move in; moving the scale up gives prospective entrants more say and produces better outcomes.
“If you keep it at the street level, the people who control it are the people who are living there, and they are against everybody who wants to move in, by definition.”
- Modern systems dilute property rights by reducing what you can do within your own lot while granting you a negative property right—the power to prevent neighbors from building what they want—and this dilution of property rights is a major cause of NIMBY.
“At the same time, you increase the property right outside your boundary, but it’s a negative property right. You are allowed to prevent people from building what they want. This dilution of property right, I think, is a big cause of NIMBY.”
- Urbanization is never 'haphazard' as the World Bank often describes it; every element is placed by people for very good reasons (sometimes distorted by regulation or discrimination), so understanding a city means asking why each feature is where it is rather than assuming randomness.
“One word I hate that the World Bank use all the time in a report is, urbanization is “haphazard.” There is no such thing. Urbanization is done by people, and they have a very, very good reason for doing it.”
- Institutions like the World Bank are like people: they age and deteriorate, and at some point should be buried and replaced with a new one rather than reformed—a 'charter World Bank'—even though the existing one still does useful work.
“institutions like the World Bank are like people. They age, and they deteriorate, and at a certain time, you have to bury them and then build a new one.”
- Cities are very much like individuals: they share common physiology but differ in culture and luck, so each city has a chance to develop or stagnate—much like a person's fortune of being born in a particular era.
“Cities are very much like individual. We all have the same physiology, but we have different culture, different ... But we have also different luck”
Housing supply constraints force the poor into long commutes and inferior conditions because cities block new construction while jobs grow.
- When high-income people move into a city, rents rise not because the newcomers are greedy but because the city simultaneously blocks new housing supply through arbitrary block-by-block zoning limits and mandated large units; you cannot increase jobs while freezing housing, so the problem is the city refusing to build, not the people coming in.
“yes, certainly it will raise the rent, but not because higher-income people are greedy or whatever. It’s because the city is blocking the supply at the same time. You cannot have an increase in jobs and a freeze in housing at the same time.”
- When people view their house as their primary retirement savings vehicle—an ATM machine—they fight any nearby construction that could limit their home's appreciation, so cities where housing is not the main savings vehicle have a better chance of becoming YIMBY.
“If people perceive that the only way of saving money for their retirement is in their house — their house is some kind of an ATM machine — then they don’t want any competition.”
- If building is allowed at much greater volume with variety in building size, the poor can outbid the rich in desirable locations by consuming less space, making cheaper, upwardly mobile living possible even in high-demand areas like the Bay Area or Paris.
“you could have a way where the poor outbid the rich by consuming less in a very desirable location. This is what you see sometimes in Paris, by the way.”
- The new urban proletariat are not people who are starving or lack clothing as during the Industrial Revolution, but people forced to commute three to five hours a day—like a fully employed Johannesburg woman commuting five hours—whose family life and the advantages of the city are destroyed by the combination of restrictive housing and deteriorating transit.
“The new proletariat are the people who are commuting back and forth three hours or four hours a day.”
- Given magical power over New York law, Bertaud would remove all floor-area-ratio restrictions, remove minimum floor-space requirements, and ensure no developer ever receives a property-tax holiday—because tax holidays are too convenient for politicians, never appear in budgets, and contribute to underfunded infrastructure.
“I will remove all floor ratio restrictions. I will remove the minimum floor space requirements. And I will make sure that no developer ever gets a holiday on property tax.”
Markets allocate resources like gravity—always present and misunderstanding them leads to failed policies like rent control and floor-area ratios.
- Because Marxist doctrine held that land has no value (only labor does), socialist cities could not use land prices to finance redevelopment; replacing a low building with a high one in a valuable center required a planner's decision and a cash outlay to compensate users, so centers stayed underdeveloped—reversing the market logic where land prices automatically finance new building and planners only slow change.
“according to Marx, land has no value; only labor has value. So if land has no value, and you are in the center of the city ... you are stuck with the existing building.”
- Markets behave like gravity—always present and to be understood, not denied; rent control is like a flying carpet that pretends to solve affordability while ignoring how markets work, and like an airplane designed without regard to gravity, it won't fly.
“Rent control is a flying carpet. As soon as you try to find a solution to affordability, and you look outside the market — how the market works — immediately you end up ...”
- Consumers can visually make tradeoffs among the three things they can see—dwelling floor space, lot size, and location—so regulation should not artificially limit these; floor-area ratios and minimum apartment sizes force people to consume more land and floor space than they want, while fire and sanitation rules (which users can't evaluate) are a separate matter.
“Users should be able to make tradeoffs between those three things, and they are able to make it visually.”
- Housing's essential function is mobility—people move between dwelling types and locations across life stages—but subsidized and rent-controlled housing ties the subsidy to a specific unit, so people never move, eliminating mobility and benefiting no one else; the market, with people moving in and out, is what provides housing.
“When you are in a subsidized house, you don’t move. If you are in rent control house, you will never move because your subsidy is entirely linked. You have no market, so you have no mobility. The idea of housing is mobility.”
Dense cities require transit maintenance as critical infrastructure and efficient road pricing rather than free parking or congestion avoidance.
- Infrastructure is important but exists as a tool to serve people, not as a purpose in itself; planners should look at people before looking at infrastructure.
“infrastructure is important, but infrastructure is there to serve people. Just look at the people before you look at infrastructure. Infrastructure is there as a tool, not as a purpose in itself.”
- Congestion pricing should be understood as a way to efficiently allocate scarce road space, not as a new tax or cash cow; treating it as revenue makes it incoherent, and the right model is Singapore's, which prices roads by location, time, and even duration of use.
“It’s not a tax. It’s a way of efficiently using road. If you look at it as a cash cow, it doesn’t work.”
- Unlike housing supply, the supply of urban road is finite once built and cannot be meaningfully expanded, so the only way to use it efficiently—including under self-driving cars—is to price demand by time and location, like a cheap motel adjusting rates every 10 minutes, otherwise roads clog and tasks like food delivery become impossible.
“You can expand the supply of housing as much as you want. You cannot expand the supply of existing road. So pricing, yes, is the only way to do it.”
- Dense cities like Manhattan, Paris, and London achieve mobility only through heavy transit use, so the lack of maintenance and decay of the New York subway is as serious a public failure as lead in Flint's water—transit must be maintained like sewer and water systems.
“the destruction, the lack of maintenance, the attention which has been given to the subway is equivalent of lead in the water in Flint or a thing like that.”
- Blaming Uber for Manhattan congestion is misguided: roughly 25 percent of New York's street space is used by cars parked full-time and free of charge, so if congestion were the real concern, cities should reclaim curbs for traffic, loading, pedestrians, or bicycles rather than free parking.
“I think something like 25 percent is used by cars which are parked full time, free of charge, in the street.”
- The question about advising a young Robert Moses is a difficult one.
“This is a difficult question.”
- Progress should not be assumed linear; the Enlightenment may have been a fluke, institutions that once constrained figures like Robert Moses could deteriorate faster than expected, and authoritarian regimes—and a 21st-century Robert Moses—are quite possible.
“We should not consider that progress is a linear thing. Some of my colleagues even think that the Enlightenment might have been just a fluke in the history of the world, and that within 20 years, we may all be under regimes which are authoritarian.”
People, not physical structures or plans, create successful cities; location, labor markets, and voluntary migration matter far more than state decree.
- A city is fundamentally its people; if residents are driven out (e.g., by Airbnb conversions), the city loses its essence, so even a laissez-faire economist may favor restrictions like Switzerland's red line preventing foreigners from outbidding farmers for land.
“Tourism might be important and interesting, but it’s not a sense of city. The cities are the people. If the people are gone from a city, it’s not very interesting.”
- Shenzhen succeeded not because the government decreed a city like Brasilia, but because Deng Xiaoping created the first labor market in China within its perimeter—where people could change jobs and earn skill-based wages—attracting bold, confident migrants from across the country, which is why many residents speak Mandarin rather than Cantonese.
“Nobody in his right mind will decide to live in Brasilia just by choice. It’s one of the worst performing cities. It’s not just my taste. It’s the worst performing city.”
- Jakarta's roughly 30-million-person metropolitan area is far more productive than pleasant cities like Bandung because its agglomeration of diverse talents working together creates efficiency that cannot be transplanted to a new capital in Kalimantan; at most some parts could shift south toward Bogor, but moving the city is infeasible and would drain resources better used to fix Jakarta with technology.
“The productivity of people working in Jakarta is much, much higher than a city like Bandung for instance, which is rather pleasant to live in.”
- You cannot create a city merely by building physical structures (e.g., Saudi Arabia's Neom); cities require the right location and the right people, not just concrete spending.
“You don’t create a city by just putting concrete.”
- Brasilia works only as a city of bureaucrats who were essentially moved there involuntarily and paid for by taxpayers; by objective measures—world-record pedestrian deaths and extreme income segregation with the poor 30 km away—it is one of the worst-performing cities, unlike Shenzhen which attracted voluntary migrants.
“Nobody in his right mind will decide to live in Brasilia just by choice. It’s one of the worst performing cities. It’s not just my taste. It’s the worst performing city.”
Informal economies and property rights work far better than planners assume; slums function as real markets with genuine price signals.
- The boundary between formal and informal economies is weaker than most planners assume: slums have functioning markets, and in Indonesia a water bill served as an effective proxy for a property title, discounted only about 10 percent versus formal title—so informal property rights can guarantee fair transactions much like formal ones.
“they found that a property guaranteed by a water bill was discounted maybe about 10 percent compared to a formal thing.”
- Delhi's unworkability stems not from density but from deficient infrastructure and the Delhi Development Authority's monopoly on land development—so density itself is not the cause of urban dysfunction.
“Not because of density. Because of deficient infrastructure and because of the monopoly of the Delhi development authority on land development.”
- China's Hukou system is not equivalent to a congestion tax because it does not prevent people from coming to cities; it merely penalizes the poor who come anyway, often leaving children in the countryside for free schooling, delaying their integration—and the urban Chinese economy depends on non-Hukou workers, so expelling them would collapse it.
“It’s just a tax on poverty because people come to the city to work, and they will pay whatever tax. So you are just paying for being poor. It doesn’t restrict people.”
- Haiti's deforestation was not natural but a systematic political effect: heavy taxes on kerosene under Duvalier made it affordable only to the wealthy, so the poor cut every tree and bush to cook, driving environmental collapse.
“kerosene was so expensive because of the high taxes, were cutting every tree, every bush in order just to cook. This deforestation, again, was manmade, but a systematic political thing.”
Political and cultural centralization, not planners, determines city-size distributions; decentralization policy alone cannot alter them.
- Cities need good locations and there are not 50 (or even many) excellent unused locations left in the world, so Paul Romer's vision of creating 50 charter cities is unrealistic; great cities like Singapore, Malacca, and Mumbai exist for good geographic reasons.
“He had decided that he could create 50 charter cities around the world. And my reaction ... is that there are not 50 very good locations for cities around the world. There are not many left.”
- Skewed city-size distributions (e.g., France jumping from 12-million Paris to one-million second cities) are caused by political and cultural centralization that hugely advantages the dominant city; planners cannot spread growth to smaller cities without changing the political system, as French decentralization in the 70s–80s failed to alter the pattern.
“If you have this pattern, it’s because you have a political system which gives such an advantage to the major city that this is where people want to go.”
- Le Corbusier is underrated as a writer, mixed as an architect, but grossly overrated as an urban planner, and his personal bitterness despite fame is reflected in his planning and architecture.
“As a writer, he’s underrated. As an architect, well, more or less. As an urban planner, of course, he’s grossly overrated.”
- The Middle East is most likely to institute a charter city because it must solve a persistent refugee problem, has large deserts with no alternative use, and the refugee population includes a mix of skill levels that would suit a new city—with Jordan or a post-change Syria as plausible locations.
“if you have a large piece of desert which has no alternative use, I think it will be a good way of starting a charter city, and the demand is really the refugees.”
- Urban cableways (as used in Cali, La Paz, Medellin, etc.) are likely overrated as transport because, as with ski lifts, cable cars cannot carry that many people.
“my gut feeling is that it’s overrated. Because I used to do a lot of skiing, and I know how much a cable car can carry, and it’s not that much.”
- Bertaud's career advantage came partly from graduating at the time of decolonization, when newly independent countries were desperate for even minimally skilled Western graduates—allowing a 25-year-old to become inspecteur de l'urbanisme in Nigeria—an opportunity unavailable to comparable young graduates today.
“we graduated at the time of decolonization, and a lot of countries were absolutely desperate for people where just the minimum skill you have when you just get out of university.”
Technology once invented cannot be uninvented; banning surveillance merely shifts it to private actors, so regulation must govern its use instead.
- Once a technology like facial recognition is invented it cannot be dis-invented; banning government cameras just shifts surveillance to private actors whose data police will access, so the only way to preserve freedom is for government to use it under extremely strict rules rather than pretend it can be eliminated.
“I don’t think you can dis-invent a technology. You could have a law which prevents the government from having cameras ... but then you will have cameras run by department stores, and you will see that the police have access to it.”
- Hitchhiking declined because of the rise of driver liability for passengers—once you can be sued for an accident or anything else that befalls a rider, you stop picking people up; the same litigious dynamic now threatens mountain climbing and other risky freedoms, destroying personal initiative.
“As soon as you are liable for your passenger ... So as soon as you do that, you completely discourage this freeloading.”
Regulatory scale and property-rights structure handed to existing residents create NIMBY incentives by excluding prospective newcomers from voice.
- Apparent core depopulation in cities like Paris and Manhattan often reflects gentrification—people now occupying far larger apartments than before—rather than actual loss of population, so falling density does not mean failing labor markets.
“you had a gentrification of central Paris, so people who used to live in 10 square meters or 12 square meters — there’s still some who do — are living now in 120 square meters, so you have a loss of densities. It doesn’t mean a loss of population in this sense.”
- Central Park's exact size, like an avenue's exact width, matters far less than the fact that New York's grid established clear public/private property boundaries long in advance, giving the market complete symmetry of information so buyers knew street locations before development, which prevents manipulation by those who would move streets for political payoff.
“That allowed the market to work with a complete symmetry of information. Even when, say, 85th Street was just a field, everybody knew that it was 86th Street.”
Self-driving cars will dramatically reduce road real estate needs through tighter spacing, making urban transport fundamentally a land-use question.
- Self-driving cars will consume far less road real estate because they can travel within half a meter of each other rather than needing the roughly 85 square meters per car required at 25 mph due to two-second human reaction times—urban transport is fundamentally a real estate problem.
“if you go on 5th Avenue at 25 miles an hour ... you need about ... 85 square meter of road because of the distance between two cars. Because you need a two-second reaction time.”
- Google's Sidewalk Labs project is acceptable as a real estate operation but was oversold by claiming a parcel the size of Washington Square could solve housing, poverty, transport, and pollution; its 'smart' features are mostly gadgets whose value can only be judged if replicated at city scale, so the overselling will likely backfire.
“I have a problem when they say that on a piece of land which is basically the size of Washington Square, they are going to solve a housing problem, poverty, transport, pollution.”
New urban infrastructure and technological change create unexpected secondary uses; retail collapse will spawn service economies rather than new housing.
- As online shopping advances and big-box retail empties out, the freed-up urban space will be used mostly for more restaurants, bars, and barber shops.
“A lot more restaurants and bars, and maybe barber shops.”
- E-scooters will last (perhaps in a different form) and are superior to the Segway primarily because they are foldable and light enough to carry and stow, whereas a Segway is too heavy to bring inside.
“I could see bringing my e-scooter here, folding it, and putting it under my chair. I cannot put a Segway under my chair. Segway is just too heavy.”