Facts about cities

by on January 16, 2013 at 11:51 am in Books, Data Source, Economics, Uncategorized | Permalink

The average tract density of all these (U.S.) cities taken together declined in every decade since 1910, from 69.6 persons per hectare in 1910 to 14.6 persons per hectare in 2000, almost a fivefold decline.  Fitting an exponential curve to this average density in every decade from 1910 to 2000, we found that the average annual rate of decline for the entire period, assuming a constant rate, was 1.92 percent.  Declines in average tract density between any two consecutive censuses were registered in 124 of the 153 observations, or 81 percent of the time.

That is from the new and quite interesting Planet of Cities, by Shlomo Angel.  My takeaway is that the Avent-Yglesias push for greater urban density, which I sympathize with, is unlikely to happen on a significant scale.  If you are looking for hopeful signs, there is this:

…between 1990 and 2000, six cities in this group registered an increase in average tract density: New York, Washington, Los Angeles, St. Paul, Syracuse, and Nashville.  Hence, while average densities in U.S. cities have been in general decline for almost a century, they may now be reaching a plateau and even gradually increasing.

I definitely recommend this book to all those with an interest in urban issues.

Peter Schaeffer January 16, 2013 at 12:14 pm

Dense cities are like public transportation. Everyone wants someone else to use it, so they can have uncongested mobility…

Of course, the real problem is that dense cities (in the U.S.) have deeply dysfunctional schools which means that the family oriented middle-class invariably departs. Of course, even in countries where urban schools are good, dense cities aren’t exactly family friendly. It would appear that people with kids (or who want kids) prefer a suburban environment everywhere in the world.

Of course, there are exceptions to all of the above. However, the general rules do apply.

Gene Callahan January 16, 2013 at 4:55 pm

“Of course, the real problem is that dense cities (in the U.S.) have deeply dysfunctional schools which means that the family oriented middle-class invariably departs.”

My kids have gone to uniformly very good public schools in NYC.

“Of course, even in countries where urban schools are good, dense cities aren’t exactly family friendly.”

I grew up in the suburbs. My kids are growing up in Brooklyn.

Brooklyn is far, far more family friendly than the suburbs! The amount of driving I would have to do in the suburbs for my kids to take part in the activities they do would be outrageous.

“It would appear that people with kids (or who want kids) prefer a suburban environment everywhere in the world.”

There are so many kids in middle class families in Brooklyn now that there are “stroller wars” fought by childless people who resent the space the strollers take up.

The suburbs boomed after WWII because of policy choices, not due to any inherent superiority to cities.

Peter Schaeffer January 16, 2013 at 7:13 pm

“The suburbs boomed after WWII because of policy choices, not due to any inherent superiority to cities.”

In real life the suburbs were growing long before WWII (see my comments on the Main Line below). Post-war policy (freeways) accommodated what the American people wanted. Democracy (rather than policy elitism) was a stronger sentiment back then.

NYC schools are worse than NY state schools and below the national public school average. See the NAEP data on the subject. Enjoy your enclave.

NYC’s birthrate is the same as the national average. However, 50%+ children are (literally) born into welfare dependency.

Bob January 17, 2013 at 3:05 pm

The issue with urbanism density is easily seen by using game theory models: Very small changes in an initial population can lead to an end result where a large majority of the population are less happy than optimal, but they lack the tools to make the change happen themselves.

The classic example of this problem is that of a book club. Some people would always want to go to the club. Others never. Some will miss visits at random, and others will only go if they expect a critical mass of people. A couple of bad sessions, caused by random effects, can make people stop going because they expect less attendees than the number that gives them utility, As those people stop coming, more people fall below the threshold, and eventually the club is a failure.

So it’s pretty easy to find a situation where, with perfect communication of intent, a book club would thrive, and yet the same people would not be able to keep said club afloat because their decision of going to the club is hidden from everyone else.

The same kind of models explain, for instance, the huge levels of segregation in some American cities.

Peter Schaeffer January 16, 2013 at 12:20 pm

Anyone who reads Yglesias knows that there is an element of double crazy to all of this. Yglesias wants dense cities to make room for the countless millions (billions?) of people he wants to import from the third world. In other words, Yglesias wants to kill the American Dream to flood the nation with immigrants the American people don’t want.

That’s double crazy.

maguro January 16, 2013 at 2:00 pm

But think of how vibrant everything will be.

libert January 16, 2013 at 2:15 pm

I’m an American, and I want immigrants to come here. I also prefer living in cities than living in suburbs and would rather take public transportation than drive, when it’s possible.

Peter Schaeffer January 16, 2013 at 2:27 pm

libert,

You need to work on teaching immigrants how important it is to resist the siren song of cars, suburbs, lawns, etc. If too many succumb, we won’t have room for the rest.

Nyongesa January 16, 2013 at 8:56 pm

There’s plenty of physical room in America, it could easily absorb several hundred million more people, socially though, it would need to be slow enough to make integration work, BUT the lack space is in the hearts of the likes of you. The defenition of America is the importation of immigrants and their eventual blossoming. Ironically there has been a “Peter Schaeffer” at every moment in American history insisting that any more and the place will go to hell. We are in a Long term competition of civilizations against people with higher birth rates, and we need to be generating as many “new” Americans as we can absorb and transform.

Peter Schaeffer January 17, 2013 at 6:13 pm

N,

If you will read my post, I was pointing out the current wave of immigrants are not cooperating with Matty’s visions of Kowloon on the Hudson, Potomac, etc. They (sadly) seem to aspire to a rather conventional notion of the American Dream. They may not be able to afford a nice suburban home because of low skills and incomes. However, they don’t appear to share Matty’s fantasies of future America.

You need to study a subject a bit more before you comment on it. Americans were generaiy pro-immigration throughout the 19th century. Public opinion only turned sharply against Open Border because the quality of the immigrants declined drastically as improved transportation made it possible for low skill immigrants to enter the United States. See Goldin, Hatton, and Williamson on the subject.

The public rebelled and mass immigration was ended around the time of WWI. The subsequent decades were the most prosperous and cohesive in our nations history. Since 1970, we have foolishly allowed low skill mass immigration to resume.

The consequences have been dire. We have imported multi-generational poverty on a massive scale along with family instability (50%+ illegitimacy), educational failure, declining wages, soaring inequality, political polarization, welfare dependency, irredentism, linguistic divisions, etc. An ugly picture at best.

Three simple points on the folly of mass immigration.

From Milton Friedman “You cannot simultaneously have free immigration and a welfare state”. We have a massive welfare state. The borders must be closed.

The current wave of immigrants demand, get, and sadly need racial quotas. The idea that a nation can prosper by admitting immigrants who depend on racial quotas is obviously absurd.

It is very likely that America faced a resource (energy, greenhouse gases, etc.) constrained future (like most of the world). Adding more passengers to the already crowded lifeboat is absurd.

Matthew Yglesias January 16, 2013 at 8:00 pm

“Yglesias wants dense cities to make room for the countless millions (billions?) of people he wants to import from the third world.”

Exactly. Though billion is an overstatement.

Dismalist January 16, 2013 at 8:44 pm

Yes, we are all immigrants, voluntary and involuntary.

Engineer Dad January 16, 2013 at 10:21 pm

Why are liberals so enamoured with stupid people?

Peter Schaeffer January 17, 2013 at 6:15 pm

Killing the American Dream on the altar of Open Borders… Who knew?

prior_approval January 16, 2013 at 12:23 pm

Well, considering that each of these three – New York, Washington, Los Angeles – represents the current pinnacle of power in our globalized era, it isn’t exactly suprising.

I would have considered that especially obvious to someone living in one of those cities (anyone who thinks NoVa is not part of DC is naïve, and very unaware of the current geography of power in our world – something that I would never claim is true of the general director of the Mercatus Center).

Millian January 16, 2013 at 12:42 pm

Please explain all of the logical steps. If power => more density, then New York was not very powerful when it was becoming less dense. Does this make sense? There is no evidence that the author does not believe that NoVa is not part of the Washington metropolitan area. What is your Mercatus Center comment about?

MotorBoatingSOB January 16, 2013 at 2:58 pm

Why do you insist on burying the occasional nugget of insight you have under a mountain of garbled innuendo (this time about Mercatus/GMU)?

Gene Callahan January 16, 2013 at 4:58 pm

“Since NYC has not become less dense, it seems difficult to make sense of what you wrote.”

Are your reading skills really that bad? NYC has become more dense SINCE 1990.

Major January 16, 2013 at 5:42 pm

NYC has become more dense SINCE 1990.

But only because of immigration. People already living in New York have continued to leave it in droves for the suburbs and other parts of the country. NYC is a magnet for immigrants because it has large established immigrant communities, lots of public services, and lots of low-level service jobs that are attractive to immigrants.

Corey January 17, 2013 at 9:25 am

Of course, immigrants don’t count as people in your view of the world.

Nyongesa January 16, 2013 at 9:02 pm

Hey!, what about San Francisco as an Information power pinnacle, arguably the fourth leg of the power quadrant.

Peter Schaeffer January 17, 2013 at 6:16 pm

N,

Silicon Valley, not San Francisco.

But wait… SV is depressingly suburban.

Michael Tinkler January 16, 2013 at 12:41 pm

Anyone who uses Syracuse, NY, as a barometer for urbanism is going to be disappointed. Is that the best they can do?

Urso January 16, 2013 at 12:45 pm

Although I do agree with Yglesias to an extent, his arguments on this issue seem to me to be colored by his SWPLish personal preferences: “I like living in extremely dense city in a skyscraper, surrounded by hip boutiques and restaurants, so EVERYONE should want to live in an extremely dense city in a skyscraper, surrounded by hip boutiques and restaurants.”

ed January 16, 2013 at 2:07 pm

No, he explicitly admits that not everyone has the same preferences. But he argues that lots of people would prefer to live in a more urban environment, if prices weren’t so high. In fact the high prices themselves demonstrate the high demand for this kind of living.

And he argues the high price/low supply situation is significantly caused by all sorts of land-use regulations restricting building, restricting height, requiring parking minimums, requiring historic preservation, etc.

Major January 16, 2013 at 2:22 pm

But he argues that lots of people would prefer to live in a more urban environment, if prices weren’t so high. In fact the high prices themselves demonstrate the high demand for this kind of living.

No they don’t. Price alone doesn’t tell you anything about demand. Dense urban housing is more expensive because it costs more to supply. Land is more expensive, because there’s more demand per square foot and because in mixed-use development homebuyers are competing for land with commercial businesses. And construction costs are higher because taller buildings cost more to build per square foot than shorter buildings.

And he argues the high price/low supply situation is significantly caused by all sorts of land-use regulations restricting building, restricting height, requiring parking minimums, requiring historic preservation, etc.

Right, because those regulations are necessary to limit negative externalities such as loss of privacy (caused by building things too close together), loss of mobility (caused by inadequate parking) and loss of culture (caused by destruction of historically important buildings).

Jan January 16, 2013 at 3:33 pm

Your first argument seems circular. And on the second part of your first argument, I would suggest that it is cheaper to build 50 homes in a 20 story building than it is to build 50 single-family, detached houses. So what if it costs more to build on a square foot basis? On per-unit basis, it is cheaper.

Finch January 16, 2013 at 3:51 pm

It would be even cheaper to put 50 cardboard boxes on an empty lot! We should totally do that because on a per-unit basis, it is cheaper.

Major January 16, 2013 at 3:52 pm

Your first argument seems circular.

Please describe this alleged circularity.

And on the second part of your first argument, I would suggest that it is cheaper to build 50 homes in a 20 story building than it is to build 50 single-family, detached houses. So what if it costs more to build on a square foot basis? On per-unit basis, it is cheaper.

Because people value housing size, obviously. A tiny studio apartment and a spacious detached house are both “housing units,” but that doesn’t mean people value them equally. If people didn’t value more space, they wouldn’t be willing to pay more for it.

Jan January 16, 2013 at 4:10 pm

Even accounting for the size of the unit, I am sure it is cheaper to build up.

Finch January 16, 2013 at 4:18 pm

> Even accounting for the size of the unit, I am sure it is cheaper to build up.

I think that’s just false. Costs go way up when you have to start building more than a couple of stories.

This is the first thing that Google popped up, I have not evaluated the source, but it suggests high-rise costs are much higher: http://places.designobserver.com/media/pdf/Explaining_Res_803.pdf

Jan January 16, 2013 at 4:21 pm

ed: In fact the high prices themselves demonstrate the high demand for this kind of living.

Major: Dense urban housing is more expensive because it costs more to supply. Land is more expensive, because there’s more demand per square foot

How is “demand per square foot” different than “high demand” Circular?

Major January 16, 2013 at 4:32 pm

According to the figures I have seen, high-rise condos/apartments cost around three times as much to build per square foot as single-family homes. Low-rise buildings can use inexpensive building materials, inexpensive construction techniques and inexpensive labor. High-rise buildings generally require more expensive materials (lots of steel, reinforced concrete, glass, etc.), more expensive construction (cranes, pumps, hoists, pile-drivers, etc.), and more expensive, skilled labor. In addition, a significant fraction of the total area of high-rise buildings cannot be used for private living space but must be dedicated to shared access (lobbies, hallways, stairwells, elevator shafts). This not only increases construction costs, but also operating costs, for lighting, heating/cooling, cleaning and maintenance of the common areas.

Jan January 16, 2013 at 4:32 pm

The linked provided shows that the construction costs of of a mid-rise building to be 1.60 to 2.00 times the cost of a single-family dwelling. But that estimate ignores land cost. So, as you add more levels you to a building the land value makes up a smaller and smaller proportion of the total cost. Once you get to 10 stories, that is a pretty small share in most locales.

I still think building up costs less in general. Obviously, the specific location matters (e.g. a 10 story, 20 unit apartment building in Manhattan is not going to cost less that building 20 houses in Wyoming of the same square footage), but I think it holds true when comparing U.S. cities to mid-range suburbs.

Peter Schaeffer January 16, 2013 at 4:37 pm

“On per-unit basis, it is cheaper.”

They can only be cheaper (with higher construction costs) if they are drastically smaller (which they typically are).

Why not just force Americans to live in unheated or cooled barracks? The savings in dollars and energy would be vast.

Finch January 16, 2013 at 4:52 pm

> The linked provided shows that the construction costs of of a mid-rise building to be 1.60 to 2.00 times the cost of a single-
> family dwelling

I think it said 2.5x for >8 stories.

Major January 16, 2013 at 4:53 pm

How is “demand per square foot” different than “high demand” Circular?

I don’t know why you can’t understand the difference between demand per unit of supply and total demand. If ten times as many people prefer the suburbs to the city, then the suburbs are much more popular than the city. But if the suburbs are a hundred times larger than the city, there is ten times as much demand per square foot for land in the city as in the suburbs. That means land in the city is likely to be much more expensive per square foot than land in the suburbs, despite the fact that the overwhelming majority of people prefer the suburbs.

Jan January 16, 2013 at 5:04 pm

Thanks for explaining the concept of demand on a per-unit basis and how it relates to supply. I think your argument is still circular. The primary reason that many people have chosen to live in the suburbs is because the price is low. Don’t you understand that, all things being equal, the very fact that the per square foot cost of housing in the city is higher than in the suburbs implies that more people actually would prefer to live in the city? It is true that not everyone’s preferences are the same, but in general the price reflects that city housing is in more demand, on a per square foot basis, than the suburbs.

Major January 16, 2013 at 5:15 pm

Don’t you understand that, all things being equal, the very fact that the per square foot cost of housing in the city is higher than in the suburbs implies that more people actually would prefer to live in the city?

No, it doesn’t imply that. You still don’t understand that price comparisons alone tell you nothing about total demand. “A has a higher price than B; therefore there’s high demand for A” is false, whether A and B are different types of housing, or anything else.

jan January 16, 2013 at 5:42 pm

But it does imply something about the relationship between the total supply and the total demand. Isn’t that what prices reflect and what we care about? What is the relevance of total demand absent any other context?

Major January 16, 2013 at 6:02 pm

No, it (that is, “the very fact that the per square foot cost of housing in the city is higher than in the suburbs”) doesn’t imply anything about “the relationship between the total supply and the total demand” either. You just seem totally confused about the relationships between supply, demand, costs and prices.

Jan January 16, 2013 at 7:21 pm

If supply and demand are not related to price, I don’t comprehend your magic.

Peter Schaeffer January 17, 2013 at 6:21 pm

Jan,

The national data shows that detached houses are cheaper to build that high rises, even including land costs. Of course, that would not be true in downtown New York City. Apartments are only cheaper because they are (typically) so much smaller.

In general, land is around 1/3rd of the cost of a detached home (with high variability). That means that detached homes are (almost) always going to be cheaper than high rises given the 2/3x greater cost of high rise construction.

Gene Callahan January 16, 2013 at 5:04 pm

“Dense urban housing is more expensive because it costs more to supply.”

It would be even more expensive to build houses on the peaks of 20,000+ foot mountains. And yet we don’t see $50 million homes being built in such places.

That’s because people will only pay high costs for something when… they have a high demand for that thing. But this is just micro I, and you already knew it: “because there’s more demand per square foot”

Right. Demand is high. Your argument is not circular, it is self-refuting.

Major January 16, 2013 at 5:24 pm

It would be even more expensive to build houses on the peaks of 20,000+ foot mountains. And yet we don’t see $50 million homes being built in such places. That’s because people will only pay high costs for something when… they have a high demand for that thing.

Dense urban housing is obviously worth its price to people who are willing to pay that price. There’s just no credible evidence that such people comprise more than a small share of the total housing market. That’s why the claim that the high price of dense urban housing means that there is “high demand” for such housing is nonsense.

Benny Lava January 16, 2013 at 5:43 pm

I love it when we have to explain supply and demand to libertarians. They are like school children.

Gene Callahan January 16, 2013 at 7:57 pm

“There’s just no credible evidence that such people comprise more than a small share of the total housing market.”

Roughly 30% of the US population lives in cities. And, of course, if the cost were lower, even more people would do so. That’s hardly a “small share”!

Look, things only have high prices if there is strong demand for them. Really, it’s just Micro I, as I said.

Major January 16, 2013 at 8:17 pm

Roughly 30% of the US population lives in cities.

“City” does not mean “dense urbanism.” As Tyler explained, the density of U.S. cities has declined dramatically over the past 100 years. Even “central” or “principal” cities (the largest city in a metropolitan area) aren’t very dense any more — about 2,700 people per square mile, on average. They may have small, isolated pockets of high density — typically a few square blocks downtown — but for the most part American cities are low-density and car-oriented. Mile after mile of single-family houses, low-rise apartment complexes, office parks, strip malls, gas stations, big box stores, etc.

Finch January 16, 2013 at 2:24 pm

> In fact the high prices themselves demonstrate the high demand for this kind of living.

I’m not sure this is true. It shows that the people who live in these places very much want to do so and can afford a lot. It doesn’t mean the pool of interested buyers is deep. NYC seems like a great place to live if you earn 7-figures, but it’s a lousy place if you earn $200k, which is a lot in most of the country. So the prices reflect a high desire to concentrate among the very high earners, not some pent-up demand across the country to live in a dirty, crowded place with great restaurants they can’t afford.

Tourbillon watches are very expensive. That doesn’t mean there’s some great pent-up demand for them.

Brad January 16, 2013 at 3:49 pm

Do you not realize you are way off on your estimation of Americans (and New Yorker’s) income levels. I mean the median income for a family in Manhattan is only $50,299 per year. Lots of people live there with much less than seven figures.

Finch January 16, 2013 at 3:57 pm

Define NYC.

Besides, I didn’t say people don’t live there making $200k a year, I just said it’s a lousy place to live if you make that little (and aren’t 25 where there’s still some argument). The families making $50k a year aren’t driving the real-estate prices that were being discussed.

Finch January 16, 2013 at 4:04 pm

But you make a fair point. Most of my experience with New York City is professional and lower-Manhattan. One thing I notice when I’m there is the clothing people wear – you realize that the place is not occupied by Goldman analysts, but by a much less wealthy group. It’s not clear where all the lower-earning folk live, but surely some of them are local.

Gene Callahan January 16, 2013 at 5:08 pm

“but it’s a lousy place if you earn $200k”

Crazy wrong. If all the millions of NYC residents making less than 200K think it is a “lousy place to live,” what are we all doing here? There is no law stopping us from moving to the suburbs! The fact is, most of us love it here. YOU think it would be a lousy place to live on under $200K. We don’t.

Major January 16, 2013 at 5:32 pm

Between 2000 and 2010, New York City suffered a net loss of over one million domestic migrants. That is, a million more people moved out of NYC for other parts of the country than moved into NYC from other parts of the country.

As the numbers cited by Tyler illustrate, the overwhelming trend for the past 60+ years has been a shift towards lower urban densities, not higher ones. We’re not building any more New York Cities. New urban development is overwhelmingly low density and car-oriented. Not high-density and walking- or transit-oriented. You (sorry, I mean YOU) may prefer a dense urban lifestyle, but you’re an increasingly rare minority.

Jan January 16, 2013 at 7:27 pm

Actually, no. In general, urban growth has outpaced suburban growth in recent years, even in New York. And the rural population continues its steady decline.

Check these stats: http://www.theatlanticcities.com/neighborhoods/2012/06/urban-or-suburban-growth-us-metros/2419/

Major January 16, 2013 at 8:28 pm

Actually, no.

Actually, you’re wrong yet again. Your linked article is about population growth, not density (or migration), so it doesn’t even address the comment of mine you’re responding to. Did you miss the density numbers Tyler quoted in his post?

And your link doesn’t support your claim that “urban growth has outpaced suburban growth in recent years” either. You need to read your own sources more carefully.

Tom West January 18, 2013 at 1:28 am

New urban development is overwhelmingly low density and car-oriented. Not high-density and walking- or transit-oriented.

True, but boy, is life going to get interesting if, 10-30 years down the road, energy costs make private commuting too expensive for the majority of Americans.

Here’s hoping we find we find our magic energy source before the rest of the world reaches our level of wealth and the demand for oil quintuples.

Peter Schaeffer January 16, 2013 at 2:37 pm

Yglesias is not arguing that people should have a ‘choice’ to live in high density cities. He trying to use state power to force his preferences on everyone whether they like it not. Don’t believe me? Here is direct quote

“I’m absolutely on board with Diana Lind in her argument that cities should be aiming to dismantle their urban highways. Something that I think should be clarified, because it drives home how strong the case is, is that building freeways through cities would be a bad idea even if traveling in cars were the only means of transportation on the planet.

The reason is simply that the purpose of a highway is to make it easy to travel long distances in short periods of time. But the central fact about cities is that almost by definition they’re not far from downtown. When you build a freeway that leads from downtown, through residential areas, out to the suburbs what you’re doing is making it easier to get to stuff downtown without living in the city.”

The essence of Yglesias’s cosmopolitanism is his authoritarian mindset. To his credit, he wears it on his shirtsleeve.

Jan January 16, 2013 at 3:39 pm

But the state sure did force those highways onto urban citizens, didn’t it? Highways brought plenty of changes to cities, most of them negative, such as crime, blight and racial divisions.

JWatts January 16, 2013 at 4:15 pm

“But the state sure did force those highways onto urban citizens, didn’t it? ”

How often are highways built against the will of the majority of citizens?

“Highways brought plenty of changes to cities, most of them negative, such as crime, blight and racial divisions.”

Is there any evidence of any of that being true?

MD January 16, 2013 at 4:23 pm

Depends on who the citizens are. That’s partly why in LA the 10 is where it is but the proposed freeway to go through Hancock Park failed.

Peter Schaeffer January 16, 2013 at 4:48 pm

Jan,

Highways brought crime? By allowing suburban criminals to invade safe cities and prey upon the residents? I didn’t know that.

Blight? Pre-highway America had no slums? Actually, housing conditions were notably worse before the suburbs took off.

Racial divisions? Segregation was the law of the land before the freeways were built in some part of America. De facto segregation existed in other areas. Neither dates to the start of large scale highway construction. Desegregation does however (to a significant degree).

Urso January 16, 2013 at 3:25 pm

Sure, Yglesias says that; he pretty much has to. But when I read his stuff I can’t help but feel that he thinks everyone else either (1) would love living in a dense city, if only they gave it a chance; or (2) is just wrong.

It’s like the old joke about “I can’t believe Nixon won; no one I know voted for him!” Mayy Y and his partner in crime at slate, Farhad Manjoo, seem to have difficulty understanding that not everyone on earth is a well-educated, upper class, urbanely liberal male living on the coasts. Like, they’re vaguely aware that these people called “middle Americans” exist, but they’ve never actually seen one.

Peter Schaeffer January 16, 2013 at 4:51 pm

U,

There is a very basic way to understand everything Yglesias writes.

“You didn’t go to the Dalton school and I did”. Here is what I wrote in response to his original post. No need for updates sad to say.

“Critics regard the Urbanist agenda as deeply and fundamentally repressive, totalitarian and elitist… Posts like this prove it.

“The reason is simply that the purpose of a highway is to make it easy to travel long distances in short periods of time”.

Imagine that? Helping people get around. That’s the last thing government should do. Let’s ban trains and buses as well. Everyone will have to live within walking distance. Density will be almost infinite.

“When you build a freeway that leads from downtown, through residential areas, out to the suburbs what you’re doing is making it easier to get to stuff downtown without living in the city.”

Why in a Democracy would anyone want to do that? Maybe because vast suburbs make housing more affordable? Because people like living in them? Intolerable.

“But trying to preserve downtown at the cost of eliminating your residential neighborhood’s core advantage”.

Wow. Make housing radically less affordable (in dollars or commuting time or both) so that the 1% (or is it the 0.1%) can revel in their privileged status. Pathetic.”

The Anti-Gnostic January 16, 2013 at 8:59 pm

Yglesias’s problem is personal, not public. He wants the yuppie lifestyle enjoyed by lawyers and bankers on an Internet pundit’s salary. There are lots of connected people out there who write well and produce the same commentary on their own dime. I suggest he brush up his resume and talk to his friends about a job in finance or high-level government work. Then he’d have no problem affording a hip, safe area of town.

Skyscrapers aren’t a magic wand. There’s HVAC, they have to be supplied with utilities, you have to get rid of the sewage and trash and you need infrastructure to supply the groceries, the bars, the restaurants and the coffee shops. And you’ve got to find a way to deal with rats. Lots of rats.

I bet there are plenty of places in the world where the government eschews any sort of land use regulation for the country’s urban centers. Maybe Yglesias could find such a place and move there.

Peter Schaeffer January 17, 2013 at 6:34 pm

TAG,

Your read on MattY is at least partially correct. He wants the investment banker lifestyle on an artist’s incomes. Doesn’t work of course. However, he also craves Open Borders and wants to pack Americans into pigeon coops to make it possible. See his post in the comments section above.

You are wrong about the economics of high-rises. Construction and utility costs are high. See the other posts in this thread.

Ray Lopez January 16, 2013 at 12:46 pm

You can get some of this info off the web–for example these two links:

http://en.wikipedia.org/wiki/List_of_cities_proper_by_population_density
http://en.wikipedia.org/wiki/List_of_United_States_cities_by_population_density

Note Manila has 112k people per sq. km, by far the most crowded, and the Philippines have a couple of such cities; India also ranks high; Greece has a few on the list, and by comparison the most crowded place in the USA is parts of New York city with only about 22k people /sq.km. Downtown DC is about 10k/km2 as I recall from memory…not even close. I’m surprised though that greater Tokyo, with 30M people, and parts of China don’t make it.

Rich Berger January 16, 2013 at 12:51 pm

Just curious, Professor Cowen – “My takeaway is that the Avent-Yglesias push for greater urban density, which I sympathize with, is unlikely to happen on a significant scale.” Why do you sympathize with this push?

Major January 16, 2013 at 1:16 pm

Why do you sympathize with this push?

I would also like to see Tyler address this question. If Americans have been choosing, more or less consistently for 100 years, to shift to lower densities, that suggests lower densities are better. So why should we support efforts to reverse this?

Peter Schaeffer January 16, 2013 at 2:06 pm

Cosmopolitanism > Libertarianism

libert January 16, 2013 at 2:19 pm

But isn’t the point that the shift to lower densities is at least in part due to government regulations (zoning), spending (highways), and subsidies (mortgage interest deduction)?

Peter Schaeffer January 16, 2013 at 3:12 pm

L,

“But isn’t the point that the shift to lower densities is at least in part due to government regulations (zoning), spending (highways), and subsidies (mortgage interest deduction)?”

Zoning is a mechanism through which the externalities of development are constrained and overall efficiency is raised. In other words, zoning (helps) to avoid a tragedy of the commons where inefficient development enables (some) private development to exploit the ‘commons’ of the community as a whole.

Houston, Texas provides a model of this. Houston (still) has no municipal zoning. However, restrictive covenants on land development are widespread and strictly enforced. They invariably mandate low-density development (as in the Memorial). The were devised as a private sector mechanism for addressing externalities that are dealt with by government else.

Highways are paid for by gas taxes. Notably, the Interstate Highway system of the 1950s was financed by raising the gas tax (not by selling bonds).

The mortgage interest deduction subsidies housing, not low density housing. In fact, it really subsidizes urban development because the per-square foot cost of skyscrapers is much higher than suburban homes.

Alex A. January 16, 2013 at 4:38 pm

That’s a very glib way of phrasing it. Worse, it’s not the relevant tradeoff here. In realistic terms the choice we face is between sprawling car-oriented municipal roads or denser transit-oriented municipal bus/light rail/subway systems. At current margins there is no “truly” libertarian choice for transportation planning, so framing Tyler as somehow straying from libertarianism isn’t a relevant criticism. Purely on the merits, we know that denser cities are more productive per capita, pollute far less per capita, and are increasingly in demand contra arcane zoning regulations. Google “Ed Glaeser productivity density”.

—An astute commenter might note that a “truly” libertarian approach from the beginning would likely have been denser and less car-oriented, and that overzealous government highway agencies and zoning boards are responsible for our stifling and unpleasant highway+beltway morass of car dependency. Local regulators killed 20th century trolley companies by tightly regulating fares beneath inflation and mandating that they pave their routes for automobile traffic. Only after decades of the consequent balance sheet decay were they ripe for takeover and shutdown by the auto industry–that’s left out of the lore that “the big auto companies bought up and killed the streetcar trolley system”. NYC shut down or appropriated all its private el-train companies. Google “Market Urbanism” and read their stuff.

Major January 16, 2013 at 5:54 pm

Purely on the merits, we know that denser cities are more productive per capita, pollute far less per capita, and are increasingly in demand contra arcane zoning regulations.

No, we don’t know any of those things. Your productivity and pollution claims are highly dubious, and your “increasingly in demand” claim is contradicted by census data showing a continued shift from cities to suburbs.

Peter Schaeffer January 17, 2013 at 6:38 pm

“Purely on the merits, we know that denser cities are more productive per capita”

Actually we don’t. Higher prices more than offset greater nominal output. Real output per worker is lower. Check the actual numbers. Glaeser claims otherwise by simply ignoring the (very large) price deltas.

Funny note. Where does Glaeser live? In a suburb zoned for very low density. Why? He has a family. Call it ‘revealed preference’.

Michael B Sullivan January 16, 2013 at 7:01 pm

Matt Y. consistently argues the libertarian line on density: he wants developers to be able to build their buildings to arbitrary heights, not ones forced on them by the government; he wants developers to be able to choose not to build parking lots on their property; he wants people who own dubiously historical buildings to have the right to demolish them.

On this issue, Matt Y. is arguing the libertarian position, and Schaeffer and Major are arguing against the libertarian position.

Peter Schaeffer January 17, 2013 at 6:41 pm

MBS,

MattY is a libertarian only when it is convenient (which is somewhat rare). Otherwise, he is an authoritarian central planner. Sort of like the Democrats who shriek ‘states rates’ when the Federal government interferes with some state level regulatory scheme, but otherwise detest the idea.

Mark January 16, 2013 at 2:38 pm

One take is that higher densities would reduce the use of gasoline, which is vastly underpriced when considering the negative externalities.

Major January 16, 2013 at 2:39 pm

If those government regulations aren’t serving purposes that people value, why have they persisted for so long and in so many places? The claim that government regulations and spending favor lower densities is highly dubious anyway. Think of all the government spending on common features of dense urban areas — mass transit systems, convention centers, arts centers, museums, concert halls, sports stadiums, government offices, court buildings, “urban renewal,” etc. Highway spending is funded mostly by taxes and fees levied on drivers. Subsidies account for only a small fraction of total driving costs, so are unlikely to have a large effect on incentives (in contrast to mass transit subsidies, which account for the vast majority of total costs). And the mortgage interest deduction is based on price, not location or urban form. You’ll get the same deduction on a $300,000 high-rise condo in the city as you would on a $300,000 single-family home in the suburbs.

Gene Callahan January 16, 2013 at 5:13 pm

“If those government regulations aren’t serving purposes that people value, why have they persisted for so long and in so many places?”

A reader of this blog who has never heard of public choice theory?

Major January 16, 2013 at 5:33 pm

A reader of this blog who has never heard of public choice theory?

What about it? If you have an actual argument, then make it.

Corey January 16, 2013 at 6:02 pm

“If those government regulations aren’t serving purposes that people value, why have they persisted for so long and in so many places?”

Funniest thing I’ve read so far today. Thanks.

Major January 16, 2013 at 6:11 pm

I’m doubled over with laughter at your inability to answer it.

Willitts January 16, 2013 at 6:33 pm

Arrow’s Impossibility Theorem
Logrolling
Cycling of preferences
Agenda control
Budget maximizing bureaucrats
Vote maximizing politicians

The public choice theory to which they refer is replete with examples of how a policy which is opposed by a majority of voters can become and remain law. They didn’t answer your question because it would take too long to explain it to you.

Start with Public Choice III by Mueller, or the Calculus of Consent or a good textbook in Public Economics that includes Public Choice.

Major January 16, 2013 at 6:55 pm

The public choice theory to which they refer is replete with examples of how a policy which is opposed by a majority of voters can become and remain law.

None of which explain the ubiquity and longevity of zoning regulations. The fact that the political process is imperfect and sometimes produces outcomes that do not reflect the popular will does not explain how zoning laws that are supposedly deeply unpopular have managed to survive for decade after decade, election cycle after election cycle, in thousands of jurisdictions all across the country. No credible evidence has been presented that these laws are generally unpopular at all, let alone drastically at odds with how most Americans prefer to live and get around. Proponents of dense urbanism just refuse to acknowledge that their views simply do not reflect how most people want to live.

Peter Schaeffer January 16, 2013 at 3:16 pm

Mark,

Gasoline may or may not be underpriced. However, Europe has shifted massively to cars and suburbs since WWII, even with much higher gas taxes.

Note that gasoline has a vast positive externality that needs to be considered. Cheap gasoline is clearly associated with more affordable suburban housing which in turn drives (to a degree) middle-class birth rates. Taking into account the future tax yield, gasoline is probably overtaxed everywhere including the United States.

Ricardo January 16, 2013 at 9:48 pm

I would like to see someone at some point address the environmental trade-offs between higher-density living and converting to CNG/LPG cars. I guess this would partly depend on how one thinks about the fracking technology that makes cheap natural gas available…

ElamBend January 16, 2013 at 12:55 pm

What these figures kind of mis-represent, and maybe the entire article does, is that as a whole, the US has gotten much more population dense in general, it’s just that the growth in population density has come in the form of suburban style density. We have been abandoning the country side for a century and moving to much more population dense areas, it just that those denser areas look more like northern Virginia, than Washington DC.

Peter Gordon January 16, 2013 at 1:05 pm

Most cities (metro areas) are much too big and varied for an average density reading to make any sense. The appropriate density for any use is a matter of circumstance. There is successful high-tech in low density Silicon Valley as well as in high density Manhattan.

Slocum January 16, 2013 at 1:33 pm

Robert Bruegmann made similar points about declining densities in ‘Sprawl: A Compact History. And the declining density is not unique to American cities — the same declines have also happened in large European cities.

Also don’t think only in terms of preferences for suburban-style detached housing with green space — but also about preferences for a bedroom for every kid and young adults wanting their own apartments (rather than living with parents or roommates). I live in a city of about ~100,000. It is apparently growing and becoming more dense, but that’s deceiving. New high-rises are going up downtown (and none of the old, single-family neighborhoods are being razed). So the number of housing units is clearly growing, but the population and density are not — simply because the number of people per household has continued to shrink.

Urso January 16, 2013 at 1:50 pm

It’s amazing how our expectations have changed in less than 50 years. When shopping for houses, we turned down as “too small” for a family of four, houses as big as the one my father lived in a child with a family of six.

Peter Schaeffer January 16, 2013 at 2:15 pm

Slocum,

“Robert Bruegmann made similar points about declining densities”

+1

Urban density has been declining for literally centuries. I have read that the TGV in France has enabled the creation of suburbs 100+ KM from Paris. Unthinkable even a generation ago. The “Main Line” played a similar role in the development of Philadelphia’s suburbs, but on a much smaller scale.

mkt January 17, 2013 at 1:57 am

Yes, I had to scroll down pretty far before I found this post, which makes one of the most important points. Higher income over a few decades ==> higher demand for more living space per capita ==> lower density. But that is not the same thing as saying that people are abandoning the urban cores (although in many cases they did) out of preference for suburban living (although many people do prefer the suburbs). So while there’s no guarantee that the pedestrian-oriented and public transit-oriented visions of urban life will be successful or desirable, the generic decrease in density is not evidence that Americans are rejecting those forms of urban living. Many of them may still be happy to live downtown — but they’ll demand a 2-bedroom condo instead of the 1-bedroom or studio unit which might’ve been the norm 50 years ago.

Major January 17, 2013 at 3:38 am

So while there’s no guarantee that the pedestrian-oriented and public transit-oriented visions of urban life will be successful or desirable, the generic decrease in density is not evidence that Americans are rejecting those forms of urban living.

Huh? Of course it is. Those forms of urban living require the much higher densities that Americans have now abandoned almost entirely. That’s why transit use has plummeted over the past 50 years, automobile use has skyrocketed, and the share of the population living in dense urban areas where a pedestrian- and transit-oriented lifestyle is possible has shrunk to a small minority. The market for public transit as general-purpose urban transportation now seems to consist largely of “captive” riders — people who for some reason (age, disability, poverty, etc.) cannot drive or afford to run a car and are forced to rely on transit instead.

DW January 16, 2013 at 2:36 pm

1.92% is no doubt an extremely misleading statistic because it implies smoothness and blurs offsetting contributions.

Actual changes in urban density are reactions to technological/infrastructural change and so probably quite lumpy. I imagine the data set shows massive de-urbanization following adoption of cars and construction of roads, massive re-urbanization following de-industrialization of city centers, development of waterfronts and reduction in pollution.

Peter Schaeffer January 16, 2013 at 3:19 pm

DW,

“massive re-urbanization following de-industrialization of city centers”

Massive? Probably more like “minor-SWPL re-urbanization following…”.

save_the_rustbelt January 16, 2013 at 4:05 pm

When I read these discussions I immediately think of Blade Runner.

High rise slums for everyone! Nice plan.

Willitts January 16, 2013 at 6:22 pm

Concentration camps and ghettos. It’s easier to control huddled masses.

Ricardo January 16, 2013 at 9:55 pm

Wow, this is pretty unhinged. I guess Milton Friedman and all those freedom indices have been consistently wrong about Hong Kong then.

Taimyoboi January 16, 2013 at 4:21 pm

Has Yglesias re-considered his thesis in light of the fact that driver-less cars are on the horizon?

Major January 16, 2013 at 8:51 pm

Doesn’t look like it. Self-driving cars means self-driving taxis, which will probably mean the end of almost all conventional urban mass transit (buses and trains), and an end to associated urban forms (“transit-oriented development”) and car-free lifestyles. Not that such lifestyles are common even now, except in New York and a few other old, dense cities and neighborhoods.

ed January 16, 2013 at 4:47 pm

Nearly everything Yglesias writes on urban issues is essentially libertarian. He consistently pushes for more economic freedom to build different types of buildings, or open different types of businesses, run taxi or other transporation services, and so on.

(I think he does have a regretable soft spot, though, for some mass transit/rail/light-rail boondoggles.)

Benny Lava January 16, 2013 at 5:41 pm

Remember that increasing market choices and removing government regulations is tyranny and is advocating for “forcing” people to live in misery. At least that’s why the libertarians tell me. You know they know all about it!

Corey January 16, 2013 at 6:04 pm

It’s “libertarian” only if you view libertarianism as the embrace of specific principles about choice and markets. It is, in fact, for most people about reflexive defenses of the status quo, as this thread has shown pretty clearly.

Peter Schaeffer January 17, 2013 at 6:46 pm

“Nearly everything Yglesias writes on urban issues is essentially libertarian”

That’s funny. He is libertarian only when it suits him. He is strenuously opposed to the core libertarian agenda of limited government, low taxes, no welfare, private provision of health care, education, etc.

That doesn’t make him right or wrong. However, it does make him a standard liberal, not a libertarian.

Willitts January 16, 2013 at 6:18 pm

Why would you sympathize with greater urban density when population density is associated with nearly every social ill in our public policy debates? Even the benefits of local public goods diminish with increased congestion.

If anything, we should be deconstructing cities.

MD January 16, 2013 at 7:26 pm

You’ll take away my two-dozen-restaurants-within-walking-distance over my dead body.

Ed January 16, 2013 at 9:09 pm

Cities in the US tended to become less densely populated in the twentieth century, but that doesn’t mean that this was desirable or will continue. Unfortunately most commentary on cities tends to stop at “cities in the US have become less densely populated (the suburbs are really growing!).

Anyway, the twentieth century American city really needed cheap gas and lots of land, and so in likelihood this trend will reverse itself in the twenty-first century, and we will see more a reversion to historical norms. And if I’m right, the model wasn’t desirable because it wasn’t sustainable, though it had desirable elements (get rid of the slums, disperse the population in case of a nuclear attack) that made the attempt understandable.

Major January 16, 2013 at 9:30 pm

Anyway, the twentieth century American city really needed cheap gas and lots of land, and so in likelihood this trend will reverse itself in the twenty-first century,

We have vast amounts of land. Only 3% of the nation’s land is urbanized at all. The rest is still rural. Unless the U.S. population grows to billions and billions, which seems rather unlikely, we’ll have plenty of land for everyone to live at very low urban densities.

And we don’t need cheap gas, either. We already have the technology for massive increases in the fuel efficiency of gasoline-powered cars, and for cars that use completely different sources of energy. The average car of 2050 will probably be super-efficient, electrically-propelled and fully-automated. Car travel will be even easier and cheaper than it is today.

Corey January 17, 2013 at 9:17 am

We already have the technology for massive increases in the fuel efficiency of gasoline-powered cars, and for cars that use completely different sources of energy.

There is truly no bullshit like comments-section bullshit.

The Anti-Gnostic January 16, 2013 at 9:52 pm

The biggest factor in suburban sprawl post-1960′s was probably public school integration. That’s when the Anglos, Jews, Italians and Irish pulled up stakes and headed for the wide open spaces. The trend was reversed by the urban renewal vanguard, childless gays and yuppies, followed by high paid parents who can afford private school or can muster political pressure and the ability to navigate byzantine rules to carve out good school districts. Cities are desperate for net tax consumers. Naturally, the end result is to price undesirables out of the neighborhood. Sorry, Matt.

Bloomberg is transforming NYC into an Ashkenazi-SWPL playground, and if that means stopping and frisking young black men and keeping supply scarce, that’s what he’s going to do. That little Machiavellian would have fit right in with the Italian city-states. Rahm Emanuel should take notes.

Poor old Yglesias is raging against an awfully powerful machine.

genauer January 17, 2013 at 7:00 am

Folks,

a lot of the stuff is specific to the US and to the period before 1990.

The increasing wages with city size is not true in Europe and especially Germany.
We dont have these run down central neighborhood issues in most places here.

People with kids come into the cities (at least until the 1 M size). You can have very good public transport, high quality schools,high density of shopping, restaurants, doctors, the usual, AND good air quality, forests and meadows nearby, a 500 sqm (Schreber)garten, if you feel like it.

The City of Poughkeepsie,NY is my typical example, how you can ruin the neighborhood by running too many highways through it.

I am in general more on the libertarian side, but significant regulation is needed, to have a nicely growing city, as long as you don’t do the rezoning with american bomber squadrons, as here in Dresden, Germany..

Seth C January 17, 2013 at 7:03 am

Many cities respond to growth by increasing their city limits, thus reducing measured density even if the city center is actually becoming more dense. Offhand, I can tell that at least three of those cities would be exempt from that trend. The District’s limits are virtually set in stone, so any population growth means higher density. New York’s five boroughs are nearly sacrosanct. Nashville merged with its surrounding county in the 60s, so that the city limits are quite expansive, and any growth that could remotely be called urban is sure to mean higher density.

I say this all the time: City to city comparisons are not meaningful when city limits are set at a local level.

Ted January 17, 2013 at 10:15 am

The industrial revolution and its consequences have been a disaster for the human race.

msgkings January 17, 2013 at 2:54 pm

Early entry for worst blog comment of 2013

Major January 17, 2013 at 11:05 am

Yes, that’s right, Corey. “New York City has grown recently only because of immigration, not because Americans want dense urbanism” = “immigrants don’t count as people.”

You appear to be trying to outdo yourself in the stupidity of your comments.

Phil January 18, 2013 at 11:36 am

sorry I missed this whole debate as it interests me greatly. For some interesting commentary on the micro-economics of city planning and maintenance, check out http://www.strongtowns.org. it is an organization run by Chuck Marohn, a civil engineer and planner. he leans libertarian and fiscally conservative, but also supports “traditional development patterns” such as walkable downtowns.

The Anti-Gnostic January 17, 2013 at 11:15 am

Actually, that’s precisely the point: immigrants are people, with all their passions, preferences, diverse worldviews, etc. Sometimes they’re good, sometimes they’re bad. Like all people everywhere, they take up space and generate waste. They have to eat, work, sleep and get their kids educated. They get old and sick. They aren’t fungible cogs that can be slotted in wherever some social engineer thinks would be a good idea..

There is a bizarre, manic thoughtlessness about immigration. It’s as if immigrants are like units of Time. You can never have too many! How did immigration work out for indigenous Americans and Australians? Do Israelis want more Arab immigration, do Palestinians want more Israeli immigration? What happened the last time Palestinians immigrated to Lebanon? Sure we could get the population density of the Indian subcontinent, but do we want it?

The US imports 1.5M people a year. Can that continue indefinitely? When we’re as crowded as Europe, is that when we stop? Whether you think global warming is a climate issue or an artifact of urban thermal inversion, it’s clear that there are some real environmental impacts that nobody’s even talking about. What happens when millions of people ramp up their consumption to American levels? What happens to wages, land prices? Is cheap labor and expensive land what made America great?

Peter Schaeffer January 17, 2013 at 6:48 pm

TAG,

Sanity on immigration is a thought crime.

msgkings January 17, 2013 at 10:11 pm

Nah, what you call ‘sanity’ on immigration is just wrong.

And we know this because your arguments have been used, wrongly, for over a century.

The Anti-Gnostic January 17, 2013 at 11:04 pm

The statement “your arguments have been used wrongly for over a century” has zero rhetorical weight. Is English your first language?

Peter Schaeffer January 18, 2013 at 2:24 pm

Those “wrong” arguments were used to stop mass immigration around WWI. Our nation prospered for decades in part, as a consequence. Then we allowed mass immigration to resume with vast and ugly consequences.

msgkings January 18, 2013 at 1:21 am

Si, senor, es mi lengua primera

Comments on this entry are closed.

Previous post:

Next post: