Is the new market urbanism overrated?

That was the question I had reading Joel Kotkin’s new and interesting The Human City: Urbanism for the Rest of Us.  Kotkin doesn’t himself come out and say that, but it is hard to avoid seeing how his arguments point in that direction.  He has two powerful arrows in his bow:

1. Birth rates in cities are too low, so highly urbanized countries such as Singapore and South Korea will have difficulty sustaining themselves.  Making cities nice, while it brings human benefits, does not solve this problem and in some ways makes it worse.

2. Lots of high-density, vertical building doesn’t really make cities cheaper.  In fact it sucks more talent in, and more business activity, and in the longer run makes cities more expensive.  Just look at Seoul and Singapore, which have built plenty but are nonetheless considered some of the most expensive cities to live in.  After all, isn’t that the increasing returns to scale story?

If I read Kotkin correctly (and this post is my interpretation of him, not a summary), he is not criticizing the policy choices of Seoul and Singapore, which have elevated those countries, or in nerdier terms you could say they have brought significant infra-marginal benefits.  He is simply pointing out that liberal building does not solve the problems it is supposed to solve, most of all the margins looking forward.

Perhaps to address those problems we need to look outside the realm of the city.  America, by the way, is uniquely well-positioned to do this.  Singapore, short of cutting a deal with southern Malaysia, has nowhere to go, so to speak.

Here is an excellent essay by Kotkin on Singapore.

I say Singapore should inspire more social science.  Pararg Khanna, who lives in Singapore, also has a new book out on cities and the value of interconnectivity: Connectography: Mapping the Future of Global Civiliation.  I haven’t read it yet but here is his TED talk on the same.


You can do a lot of social science on American cities that differ in terms of how much their potential commuting zones are hemmed in by deep water or mountains.

For example, Dallas can draw commuters from almost 360 degrees, while San Francisco's suburbs are greatly restricted by geography. This "dirt gap" has major influence on the supply of housing, and therefore its costs, and therefore on fertility, and therefore on politics:

Most "red states" have inland metropolises, while most "blue states" have their major cities up against oceans or Great Lakes or severe terrain.

The supply of housing in San Francisco's suburbs is not restricted by "geography." It is restricted by an alliance of homeowners and community activists who show up at every community meeting quivering, shouting and spitting with hatred and sanctimony at anyone who wants to build housing. What housing gets built happens only after about ten years of obstructionism and a thorough shake down of the developers. Berkeley is not nearly as dense as even the Sunset, and Palo Alto is not nearly as dense as Berkeley. It is not geography that explains that difference, and developers would gladly build them all out if government did not meddle.

What's west of the Sunset neighborhood anyway?

The Farallon Islands.

You're being obtuse. My claim is it would be profitable to build extant suburbs more densely, and the Sunset is an example of a denser residential neighborhood that already exists in the Bay Area. That density would add to the supply of housing. Therefore, the supply of housing is not constrained by geography, as you claimed.

Quite the opposite, you're picking a single city and ignoring the implications of what you are saying. Even if San Francisco was razed and turned into Sim City Arcologies, you'd have a gigantic urban core and not sprawling suburbs, and still high property values. Or just look at New York City: is New York City more conservative because it allowed higher density?

Love all of the either/or arguing in this thread. Why can't it be both policy and geography? San Francisco (like Singapore) is obviously more geographically constrained than most American cities, occupying the tip of a peninsula. The ocean, bay, and surrounding mountain ranges (prone to slides when built on) constrain most of the region. There is no good place to develop new suburbs relatively close in to San Francisco, so the cheap housing is out near Stockton now, in the Central Valley.

But yes, San Francisco could be built much denser. 50 - 60 years ago there were schemes to flatten San Bruno mountain and fill in most of SF Bay south and east of the SF airport. Rejecting this was a policy decision (and I would not have liked to have seen the effect of liquefaction in the 1989 Loma Prieta quake on all of that fill). All of the surrounding suburbs, not just San Francisco, are fighting increased density. Here in the Boston suburbs, rampant NIMBYism and anti-multi-family housing attitudes help to keep the housing market tight and drive up prices, but since Boston lacks geographical constraints on three sides, prices are unlikely to ever catch up to San Francisco (aside from any differences in desirability).

I haven't read Kotkin's book but taking off from what Tyler said, it seems obvious that to whatever extent denser housing keeps prices from increasing as much as they otherwise would, it also is going to encourage more people to move in than would be able to at the higher price. In other words, in highly desirable cities, there will always be a "shortage" of housing relative to other cities, leading to high prices. Which doesn't mean we shouldn't build more densely.

Conceivably you can't import high density without importing a whole host of other socio-cultural phenomenon that lead to other left-leaning politics. Also, you cannot build enough housing to reduce housing costs, but merely mitigate the increase. And even if you could build enough housing to reduce costs, you cannot scale transportation infrastructure fast enough, which still leads to high-price real estate around ever smaller transportation hubs.

You can't just change a single variable and assume everything will turn out just the same. The curious task of economics and all that.

No, Lewis, YOU'RE being obtuse. The fact that SF's housing supply is restricted by things besides geography does not mean it isn't also restricted by geography!

I'm old enough to remember the emergence of the modern environmental movement, from the January 1969 Santa Barbara oil spill onward. As a resident of the inland San Fernando Valley, I was always vaguely aware of environmentalism as particularly appealing to wealthier coastal Californians in places like Malibu in need of an ideology for why they are Saving the Planet by keeping the inland masses like me out of their paradises.

Thus today, the population of Malibu, which has 27 miles of coastline, is about 13,000, with zero highrises.

"Conceivably you can’t import high density without importing a whole host of other socio-cultural phenomenon that lead to other left-leaning politics."

I think what is well attested to all across the United States since, say, the turning point year of 1969 is: physical limitations on suburban expansion tend to conjure into existence nominally leftwing ideologies, such as environmentalism, justifying why the current residents deserve the lion's share of the benefit from any future development and are justified in shutting down future development if they won't be made better off.

I'm not saying it is good or bad, just that it is a pattern, one that is tied into the division of America into Red States and Blue States.

"But yes, San Francisco could be built much denser."

My cousins are San Francisco real estate developers. It took my cousin G. a decade to get the permissions to build an 8 story apartment building on Van Ness, and he only got it done that fast because he had already had decades of experience learning how to play the game in San Francisco.

All I'm saying is that there is a fairly predictable pattern, based in part in geography, about which city's politics will be pro-development and which will be anti-development.

In addition to the issues mentioned by Steve Sailer, the development of land is also restricted by the Williamson Act. (There are multiple Federal Acts that are even more intrusive). The entire Bay Area is pock-marked by these set asides. The net effect is to limit city borders and push home-owners out to the Ex-Burbs. Very little of that land needs to, or should be, preserved. No one visits or uses it, no one cares. The real estate constraints are driven by the vested interests.

I gotta call this one for Lewis, Steve. San Francisco county has lower density than four of five of NYC's boroughs. At SF land values, there's no reason it shouldn't be as built up as Manhattan, which it itself isn't exactly known for ease of construction.

To be fair, the recent history of East Asian and Persian Gulf megalopolises indicate that rapidly building a city skyward does little to lower the cost of housing. Demand seems to scale super-linearly with density. Double the square footage on Hong Kong Island and all of a sudden four times as many workers are clamoring to move to a white hot job and property market. Undeveloped land is what keeps housing cheap. Ever taller skyscrapers seem to only benefit billionaire property developers.

"Singapore, short of cutting a deal with southern Malaysia": just my thought. Singapore should try to buy both ownership of, and sovereignty over, a small chunk of Malaysia.

Singapore is probably better off reclaiming more land from the sea or negotiating with Indonesia over improving transport links with them.

In any event - I am not sure why the idea that 4 times more valuable apartments is considered a problem - isn't this a good thing to have expanded supply of housing four times but the value has increased? From a Singapore or Hong Kong point of view they have more assets. And by definition people are able to afford these assets, that's even better, now we have expanded the amount of rich people by a similar amount. It's a bit strange to see Tyler coming from the expensive houses are bad approach. Expensive houses can only occur in rich areas. Ideally every house would be worth a lot then the world would be rich all over instead of just in some very restricted locations like the center of major cities.

Steve's got this one. All the NIMBYism would not exist if they had more land to build on. They could double the population/living space without increasing density.

San Fran COULD build denser, but obviously does not want to. Why would you want to be like NY if you don't have to (density wise)?

"the recent history of East Asian and Persian Gulf megalopolises indicate that rapidly building a city skyward does little to lower the cost of housing."

No it does not. You have not isolated the variable.

I'm a land developer in Phoenix and where I see so many developments getting bogged down in places like Berkeley and San Francisco (I also saw it 30 years ago in Vail and Aspen) is with affordable housing. No one disputes the problem but the municipalities want to solve it on the back of new development. It's a regional problem and as such needs to be solved by taxing the entire population base ...but politicians don't have the backbone to do it and as a result nothing gets done.

Would you mind explaining a bit more why this is so?

Geography + earthquakes, but in California sometimes the car bedroom (A.K.A. minimum parking space) is larger than the people bedroom. If parking costs are bundled to housing costs, housing cost increase by a non-trivial %. Parking minimums reduce congestion not by making traffic more fluid but by making population density go down as Doug mentioned.

San Francisco housing parking minimums should be put under revision since 30% of households do not own a car. If carless households are paying the parking costs in the monthly rent, they're being ripped off .

The solution is pretty obvious: just build a second New York City!
Logistically, this will be tough: NYC has great shipping channels, semi-tolerable weather, and endless supply of drinking water, and the geology to support subway systems and skyscrapers. Where else in the U.S. offers all of those things that isn't already built up?
But if you can find a place, it's just a matter of creating a special economic zone and gathering the funding. Grant visas + pathways to citizenship for the best five million immigrants you can muster (conditional on productive behavior for, say, ten years in our NNYC zone), raise a two-trillion stimulus to build Tokyo-style infrastructure (without the issue of path dependence, you'd get a huge bang for your buck), and watch all the hipsters, banksters, and techies rush in to profit off the thing one way or another.
But again, the main question is where to put it. The best idea I have is right on top of where Baltimore would then use to be, but if done by diktat should still be a huge profit for all if done humanely.

I put in paragraph breaks, I swear?

I guess you have to double them?

Miami's already pretty close to what you described.

The whole region becomes very sensitive to tides and rain if sea levels rise another foot or two. Not sure I'd pony up my share of the trillion dollar development cost.

I would suggest building a major new Airport (Dubai style with 5 runways) in rural New Jersey then putting in fast transport links to NYC. Then you can reclaim the existing airports at La Guardia and JFK and make them new high density cities. Ideally with a mix of housing and apartments.

Geoff and ChrisA. Both good solutions.

This has been tried (Brasilia, Islamabad) with mixed results, but the US should do better. Also, "endless supply of drinking water" is an...iffy...characteristic of NYC, depending on what parameters you consider safe for drinking.

I'd nominate somewhere like Savannah or Mobile over Baltimore. If you had to go greenfield, anywhere along the lower Mississippi really. You're going to have to get around the "efficient market hypothesis" of all the desirable areas being taken, and the biggest historical constraint we've removed over the last 100 years is probably heat and humidity by developing air conditioning and making electricity cheap enough to run it full blast everywhere.

Don't ruin Savannah.

"Also, “endless supply of drinking water” is an…iffy…characteristic of NYC, depending on what parameters you consider safe for drinking."

NYC's municipal drinking water is the best in the country:

I'd imagine that somewhere in New Jersey/Delaware/Maryland would make the most sense. Build up the Delaware Bay, Delmarva Peninsula, eastern Chesapeake?

Otherwise, why not just continue to develop Houston?

Shipping is out of New Jersey now and the water is imported from the Catskill Mountains.

I am not a crackpot—just infantilized by the otherwise ubiquitous preview and/or edit buttons.

"Lots of high-density, vertical building doesn’t really make cities cheaper. In fact it sucks more talent in, and more business activity, and in the longer run makes cities more expensive. Just look at Seoul and Singapore, which have built plenty but are nonetheless considered some of the most expensive cities to live in."
First, what counts as "built plenty"? If you have any supply-and-demand chart with the demand curve shifting quickly rightward, but, by regulation, you only allow supply to rise half as quickly as it would without regulation, then you will still see "plenty" of new supply and swiftly-rising prices. America makes and imports "plenty" of cars but if you put import quotas and caps on production prices will rise. Imagine what a skeptic of "market carism" would say in that world. Everyone in San Francisco currently believes they are in a construction boom and that the city has been taken over by developers. They have <1% growth in units. The mayor's big housing plan is to build about 1%.
Second, how do you know that the claim is true? Kotkin does not do any econometrics. You can't find the slopes of supply and demand curves by casually looking at price indices across cities like Kotkin---who has no training in statistics nor interest in it---is fond of doing.
Third, kind of strange how vertical building doesn't lower prices, and yet Singapore's developers and property-owners have been worried about oversupply since 2011. Just google "singapore oversupply." Here is a top hit:

You need justification to argue that housing in tall buildings makes prices rise but when you spread it out it makes them fall. You need it because the reason people build tall buildings, which have higher marginal costs, is that the price of housing is high.

It is very easy to convince book-selling thought leaders like Kotkin, who hates tall buildings, that tall buildings are counterproductive. It also gets your books attention: listen everyone, you don't have to accept those lousy multifamily units after all; the snobby planners who believe in supply were wrong and this maverick is here to tell you that you want to do is the right thing to do, too. Those eggheads! Also, cigarettes are good for you.

Exactly, you've nailed two of the most important observations. Kotkin (or perhaps Tyler's interpretation of Kotkin) confuses cause and effect with respect to high rents and high rise construction. And Kotkin has a weird phobia against high density cities.

Make cities cheaper should not be the (only) objective. With high density cities productivity usually rises and so should incomes. And what should really matter for people utility is income-cost of accommodation.

'Productivity' of what, exactly?

Incomes may be higher in most cities, but that income is mostly transferred to landlords via much higher rents. The landlords won the economic war. While economists and policy advocates fight over "capital" and "labor," it turns out land was/is the most important economic factor for affluence and societal well-being.

Until we tax rent incomes (start with 15-25% of gross rent income, less maintenance costs) and reduce regressive payroll and sales taxes by the same amount, expect the landlords (and banks who finance them) to continue to aggregate much of the wealth in the economic system.

I lived in San Francisco for a few years. The $120,000 a year income is great, but after paying ~35-40% of income on various federal, state and local taxes, plus another $35,000 for rent, the net, net income isn't as great as a 6-figure income might sound. The landlords (and banks) won the economic war.

Landlords who were there as the property climbed in price, but not necessarily today's landlord.

If you are correct, I assume all your money is in REITs.

"If you are correct, I assume all your money is in REITs."

If he's correct, I assume he holds REITs in proportion to their share of the market portfolio. Thinking "A is a good business" doesn't mean you put all your money in A.

Rent incomes are taxed. If I recall, the return on buying and renting out a property in NYC or similar is quite low. You have to rely on appreciation to make up the difference.

This is a common perception. Most small scale real-estate investors prefer secondary markets because it's a lot easier to make rentals cash-flow there. Appreciation is seen as more risky than cash-flow.

If you put it in equity terms, Silicon Valley housing has a higher P/E ratio than Cleveland housing.

Rental income is taxed.

Also, I don't want to interrupt your envious Marxist rant, but increasing taxes on income will not make your rent go down.

NET rental income is taxed AFTER tax subsidies for interest write-offs and PHONY depreciation. Real estate investing/speculation is one of the best tax loopholes in the entire 77,000 pages of the tax code and regs. Get rid of those two tax subsidies and property prices and home affordability will improve for millions of families.

A deduction, which allows you to keep to more of what you earn, is not a "subsidy". Not having to give the government something is not the same as the government giving that something to you.

Coyote explains the Great Stagnation for TC:

There's tons of warm, arable land and cheap housing in the south and midwest. But there are some obvious roadblocks to taking advantage of it for sustainable growth. The bedrock employers (usually manufacturing) in those little towns have let wages stagnate and downsized. Three quarters of the kids are getting college education at the University down the highway. Those kids have debt and aren't enticed to move back to have kids and work for terrible wages. So they stay in the city with a slightly better wage and just try to get by. Then the more successful ones pump out a kid or two and pay for daycare because Grandma is far away and still has to work anyway.

I don't know of any solutions to that that are palatable to a free market economist.

"The bedrock employers (usually manufacturing) in those little towns have let wages stagnate and downsized"

Those manufacturers are constrained by global prices. Many of them (textiles, light bulbs, small consumer electronics) are no longer constrained, because they shuttered the factory.

If you build a city bigger, it makes other cities cheaper. If you build all cities bigger, they all probably get cheaper. If you build just one city bigger, at some point it starts to get cheaper, as in any case only finitely many people are willing to pay more than $X/square foot. Or Y% of average wage, if building makes everyone more rich, as it well may be.

But whenever you build something that somebody buys, you add to global happiness, as it is worth more than the cost. Plus it usually rises value (read: benefits) of neighboring buildings too, and productivity. So it is not important whether the prises go up or down. Let them build it all.

My thoughts exactly

I don't understand. If the more we build the more expensive the cities get, shouldn't we build like crazy?

It couldn't get expensive in the absence of high positive effects of urbanization.

This makes me think we should build an entire ghost city in the middle of nowhere. If you build it, they will come!

New Ordos, anyone?

California City, CA isn't exactly a field of dreams:,_California

The US in particular is in some ways stuck in the middle twentieth century. Vertical cities are demographic fossils from an era where the telephone was the leading edge of communication. People at work or at play are no longer tied to a physical location except in basic industries like agriculture, extractive industries and manufacturing, all employing a fraction of what they once did. There's no need, except for personal preference, to locate a business or home within the high-rent urban complex.

It's because smart, young-ish, and high-value, people like those environments more than suburbs or the middle of nowhere. And believe it or not, the majority of businesses still need employees to have actual face-to-face contact with colleagues or clients with some regularity.


Actually I think this view misses some important features of relating to both consumption externalities (*cough*sex*cough*) and knowledge spillovers in work.

E.g. in RE development/investment many companies will locate their dealmakers in nyc because of the proximity to other dealmakers. Their asset managers/operations will typically be stranded in a place like Atlanta. But for many face to face tasks people need to co-locate.

You talk about no need except personal preference: That's like saying that there's no need for things to fail other than gravity.

I live in a cheap metro area, while working for a Bay company, paying Bay wages. I can travel the world when I want, and work from anywhere: I work every day with people all the way from Berlin to Hawaii. And yet, most of my coworkers prefer to live on the bay, because of the huge social advantages that provides, especially for the young and single.

When we think of discrimination, we think race, but in practice, people's preference to be surrounded by people like them goes way past race or wealth. Hipster tech people want to be surrounded by hipster tech people, not live in the bible belt, and vice versa. The costs of setting up a large enough settlement to make young people want to move are so high, we'll keep going to large, vertical cities for as long as humans are social.

Real soon now the Information Superhighway will mean that nobody will have to crowd anymore into Silicon Valley or Wall Street.

The economies of scale argument makes sense and I would add that most cities seem to have a two-tier market where some buildings are clearly designed to attract high-income tenants. When a city becomes desirable to high-income people, most of the new construction that happens there will be for buildings with relatively spacious units and amenities to justify high prices (and therefore higher rates of return). The obvious policy response is to try to preserve a certain ratio between luxury housing and housing for ordinary people when new buildings are being planned.

Singapore is an extreme case of the two-tier market because most citizens live in government-owned and subsidized public housing. This means the private housing market is almost invariably geared toward well-off locals and expatriates who often receive a housing allowance from their employers. There is no incentive to build housing for ordinary people because the government already has a de facto monopoly on that market segment.

We read lots of blog posts here about technology as the key to economic growth, or the absence of technology as the cause of slow economic growth. It's as if Cowen et al. are waiting for the next big thing to catapult the economy into the next growth spurt. Of course, technology is already here, it's just that it's underused. It's underused because it would require the implementation of a Hamiltonian vision for America that is ideologically anathema - even if the musical about Hamilton is wildly popular. From my childhood in the 1950s, I vividly recall the images of the future, a future in which space-age transportation would connect people from their homes to their work and to each other. Of course, we have the technology but refuse to use it, relying instead on a vast network of ever-expanding highways to nowhere, suburbs extending out into exurbs, with congestion choking the highways and polluted air choking our lungs and the enormous time wasted restricting our productivity. Cowen's answer: more of the same.

Sign me up for the teleporter, Ray. Wherever you're from that has them.

What does infra-marginal mean in this context?

Google says it means, "inside the margin", but inside what margin are the benefits of building in Singapore?

The last question itself is so convoluted that it barely makes sense.

There's a lot of tension between these two observations. If there aren't enough people, how do property values stay high (assuming that's what expensive means)?

There are plenty of people, just not enough babies.

Rural/suburbanites create babies. Babies grow up and move to cities. The city folk end up dying childless.

You can have a downtown demographic death trap AND a growing population in that city at the same time.

So then what's the problem?

Lots of mood affiliation in these responses. I said "overrated," not bad. It benefits society, but maybe still doesn't make it easy to move into major cities cheaply.

This whole post is just mood affiliation with suburbanites.

Green acres is the place to be
Farm living is the life for me
Land spreading out,
so far and wide
Keep Manhattan,
just give me that countryside.

As long as there is a large gap in regional incomes, there will not be enough housing supply in the high-productivity cities. Land rents can't fall until the marginal migrant is satisfied--people will keep moving faster than housing can be built until the quality-of-life gap is closed.

That said, I think we can agree that if *horizontal* urban growth boundaries are bad and artificially raise land rents, then it must also be the case that *vertical* urban growth boundaries are bad, artificially raise land rents, and mandate excessive sprawl. Kotkin is chomping at the bit to ease the first, but happy to leave the latter in place.

Market Urbanism is the tonic both for anti-vertical "Kotkinism" and for anti-horizontal "New Urbanist" greenbelting around the (sub)urban periphery.

On the one hand, Kotkin says real Americans want to live in suburbs (as opposed to effete hipsters).

On the other he says that if you build more stuff in cities more people will come, which certainly implies a latent demand for city living.

Joel Kotkin, meet Joel Kotkin.

Overrated depends preferences, but I'd go with oversold. Kotkin seems a useful corrective to density enthusiasts, by simply reporting what has been happening in the last couple of decades -- at least in the US. There has been no great urban renaissance, no big return to the cities, no dramatic shift in preferences for urban rather than suburban living. It's true that both people and employers are still moving from rural to metropolitan areas, BUT most of them are settling down in suburban and exurban regions of the metro area, not in the dense cores of central cities which have grown little in terms of population and employment. And, at the same time, the population of central cities has changed dramatically -- in some places, it has become almost exclusively a 'playground for the young, childless, and/or wealthy' (San Francisco). In other places it is that playground combined with large areas of poor, deteriorating, depopulating neighborhoods with middle income neighborhoods inexorably vanishing -- as has been graphically demonstrated by Daniel Hertz:

This graph, BTW, seems like a rebuttal to the idea that if only development regulations could be relaxed, then the pent-up demand for urban living could be satisfied for people who would like to live in dense neighborhoods but cannot afford it. In Chicago in recent decades, there has been quite a lot of gentrification of formerly middle class neighborhoods, but there has been even more deterioration of formerly middle class neighborhoods into high poverty areas. Had there been more demand among the ~7 million suburbanites to finally move to that place in the city they'd always wanted, there were plenty of neighborhoods where developers could easily have satisfied this demand. But that did not happen.

If there is truly no mass desire to move to cities, then legalizing Manhattan density as-of-right in every urban core should have very little physical effect, and should rapidly extinguish rising house prices in coastal cores...

The "happier alternate reality" that Daniel Hertz says existed not that long ago in Chicago was pretty miserable. I was there. White working-class families (now called "middle class" for political reasons), squeezed into tiny houses, were fleeing the city as fast as they could, not just because of black migration into the city but because much better housing was available at a low price in the suburbs. The wealthy were squeezed into a few blocks on the near north side, and did not dare to venture more than four blocks from the lake (this was the advice given to me when I moved there). Susan Sontag said the whole city looked like a giant slum and she was not all that far off.

Today, there is more poverty on the south and west sides, not because the people living there became poor, but because *poor people moved there*. The parts of the city that are not poor are doing great, and look like nothing a Chicagoan had ever seen in 1970; you had to go to Winnetka to see the kind of affluence that has sprouted up all over the north and northwest sides and in Hyde Park-Kenwood.

Singapore has an explicit two-tier housing market, one for resident citizens and the other for nonresident aliens or very rich citizens. The cost of housing in Singapore would no doubt be much lower if all restrictions on HDB flat sales were removed, although the median would rise given that HDB flats are subsidized.

Aren't #1 and #2 contradictory?

#1 says poulation is city not sustainable due to birth rate

#2 says city population will grow faster than supply due to attractiveness of city

Either population will grow fast making city "awesome" or it won't and city will become more affordable. It can't be both can it? Or neither?

My first reaction exactly.

Maybe I misunderstand, but can't it be both?

The city is basically a population sink, a trap for higher earners. They want to go there, but once there they can't get out and they die childless but well entertained.

It's unsustainable - we can't all do that. But the population will continue to go up, because it's an attractive way to go. Eventually, #2 will have to come to an end, and then the cities will fall again. You'll have filtered out all the suburban and rural people who care more about their ability to walk home from an evening's drinking than the affordability of children. In places like NYC, you're already needing to draw primarily on the international crowd because everyone in America who wants to live there already does.

"it’s an attractive way to go."

Manhattan is a sort of pleasant, slow suicide.

People move in when young and then leave when their kids increase appeal of public services. If people delay kids, more will live in cities. Vice-versa, I think, is not the causation direction.

There's some of this, though in part to delay having kids is to have fewer of them.

That's one of the major reasons why, if a country is observed to have an unsustainably high birth rate, then education of girls and promoting their access into the job market are supposed to be good. They will delay their first birth, and have fewer children.

I don't know about Hong Kong and Singapore, but one alleged caution about the current pattern of urban population growth is exactly the opposites: young affluent people live in the city until the couple and decide to procreate and then move to the suburbs. This foretells doom when post-millennial cohorts, which are somewhat smaller, begin to dominate among young professionals, who will not need so many urban housing units.

I don't think any of that's right, but I also don't think anyone is getting "trapped" in American cities.

It's a fertility trap, not a physical trap. Spend time in the cities, have fewer kids. The time you could have spent making babies is gone forever.

Also, some people do physically stay in the cities through their thirties and forties. It's not like that rate is zero.

I suspect that Kotkin might find himself drawing a bit on the myth of cities being fundamentally sterile, draining their hinterlands of people and so flourishing for a time, but at their expense. In contemporary Japan, for instance, Tokyo still shows some growth while rural areas are fast-depopulating.

(Of course, there's the question of _why_ the rural areas are depopulating. Why are young people leaving, after all?)

Japan may be so non-diverse that there just aren't many people not vulnerable to the city trap to be left behind. They have no equivalent to the Mormons.

Tyler - your interpretation of Kotkin seems to contradict the conclusions of Glaeser et al. Is there a shift in how economists are evaluating the costs of density restrictions embedded in zoning? Or is this just Kotkin's view?

What does the literature say?

Kotkin, whom I know slightly, would be the first to say that his work contrasts with that of Glaeser. Joel has sort of set himself up as the voice of counterurbanism.

If housing costs are high, I would advise building more housing.

If birth rates are low, and that is a problem, I would advise economic incentives to have children.

If a woman gets $250,000 to have a baby, how long until birth rates rise?

I'd also advice immigration in the second instance.

There are 4 million births in America every year. $250K*4 million is $1 trillion.

Creating a baby bonus large enough to meaningfully impact birth rates would be ruinously expensive.

That's the problem.

Assuming that a shortage of births is a problem. It doesn't seem to be.

You'd get more babies for sure, but mostly not from the parents you wanted to get them from.

The problem is not San Francisco, it's the nearby suburbs.

Most of the Bay Area suburbs are filled with low density sprawl and have heavy restrictions on new building/infill. They have the effect of creating a quasi-greenbelt of 30 miles around SF.

Alameda had a 40 year ban on multi-family housing construction. Consequently, the population density of that city is only 7K per square mile. Merely bumping that up to Berkeley levels of density would provide housing for around 30,000 more people.

Kind of funny that Joel Kotkin cites Singapore – which has more public housing than any country that's never had "people's republic" in its name – as an example of failed market urbanism.

Also – is housing in Seoul really that expensive compared to other first world megacities? Seems cheaper than NYC and London to me.

2. Never reason from a price change.

Higher supply leads to lower prices and therefore higher demand and higher prices!


One interesting example of cheap high rise housing in America is along the extreme north shore within the city of Chicago, after Lake Shore Drive runs out about 8 miles north of the Loop. There are a couple of miles of solid 30 story apartment buildings north of about the 5800 block. They were built for profit in the postwar era. When I looking in the late 1980s, it was pretty cheap to buy condos in these buildings (because expectations for the future weren't high), although the monthly maintenance fees were high because Lake Michigan was washing away their foundations. The area was not very prestigious but not particularly bad either.

You're thinking of Edgewater. The condos you mention are even on the Wikipedia page:,_Chicago
From what I understand from the other young people, Lincoln Square and Uptown are both much more "hip" communities, though Lincoln Square would be roughing it.

I quite like Uptown, though.
Also adjacent to the Devon Ave, which is where you want to go for a sari or naan. I like Naan and consider Saris quite beautiful, but my Wife isn't standing up at any more Indian weddings any time soon.

Edgewater is getting some gentrification but is still cheap. Lincoln Square roughing it? Not in the last can easily spend $1 million on a house there and there are a few around $1.5 million.

tour de france

Comments for this post are closed