Scott Alexander has a post where he discusses some flaws in the YIMBY arguments. The term ‘YIMBY’ means “yes in my backyard”, and refers to those who push back against the anti-development NIMBYs. While Alexander has some sympathy for the YIMBY perspective, he thinks they oversell their claims in several respects:

1. It is not always true that building more housing in a particular location will reduce housing costs in that location.

2. Many people find dense cities to be less pleasant places to live than less dense cities.

3.  He also wonders if the advantages of “centralization” (or perhaps more accurately “agglomeration”) are oversold. I did not find this argument to be persuasive.

I agree with his first two points, but I don’t think they have the implications that he assumes.  I also question some of his empirical claims.  For instance, he has a chart showing that San Francisco has already been building housing at a fairly good clip:

Interestingly, his previous two graphs suggests that the 2008-15 growth rate in San Francisco is barely half the 1.1% growth rate reported in this graph, so I don’t know which of his graphs are inaccurate.  In any case, the problem in the Bay Area is not just the city of San Francisco; it’s the entire region.  Metro areas such as Houston and Dallas (with similar total populations) produce several times more housing units each year than the Bay Area, albeit mostly in their suburbs. Compare population growth in San Mateo County (just south of San Francisco) with Clark County, which includes Las Vegas:

Since 1970, Clark County has gone from having 1/2 of San Mateo’s population to having three times its population.  San Mateo contains the heart of America’s biotech industry and is also important in other tech fields.  Population growth in Marin County (just north of San Francisco) has been even slower.

Admittedly, the San Francisco area is a rather special place, with beautiful coastal areas that need to be preserved.  But the same story applies to less scenic locations such as the East Bay, LA, San Diego, Boston, and New York.  In cities with tight zoning rules, housing construction is much slower than in areas where zoning rules are less restrictive.  And this keeps housing prices excessively high.

Alexander is correct that building more housing doesn’t always drive down housing prices.  That’s because in a few rare cases the construction of more housing can actually increase the demand for housing.  One possibility is that the new housing is of higher quality.  Another possibility is that the new housing offers enlarged opportunities for gains from “agglomeration”, which refers to the efficiency gains from having many firms concentrate in a small area so that they can interact.  (Think Hollywood, Wall Street, Silicon Valley, Broadway, Madison Avenue, etc.)

I’d respond as follows:

1.  New housing construction lowers local housing costs in the vast majority of cases.

2.  New housing construction always lowers national housing costs, ceteris paribus.

Thus from a housing cost perspective, YIMBYism is always in the national interest.

You might argue that I don’t understand the feelings of local residents.  Actually I do, just as I understand the feelings of taxi drivers threatened by Uber, or retailers threatened by Amazon, or dairy farmers threatened by free trade.  Of course I understand the feelings of local residents who don’t want to see more density, traffic, etc., in their community.  They may have moved to Marin County because it’s a lovely, quiet place.  Change is not good for everyone.

Alexander discusses some of the downside of living in San Francisco, with its traffic, crime, dirty streets, homeless people, etc.  He also points to Manhattan, where they built a lot of housing and yet costs remain very high.  Here we need to avoid “reasoning from a price change”; the extensive housing construction in Manhattan reflected the high demand in that region.  Even more construction would have likely meant lower prices (and did before NYC zoning got tight.)

I’d like to present a vision of unrestrictive zoning that I believe is more plausible than Alexander’s vision.  Let’s start with some facts:

1. America’s major metro areas are enormous, and mostly “suburban” in character.  Most Americans have zero interest in living in Manhattan-type areas, even if the price were low.  They like their backyard barbecues in Oklahoma City.

2. Manhattan itself is only 24 square miles.  It’s also far and away the best thing NYC has going for it.  If Manhattan (from Wall Street to 100th Street) had not been built, NYC would be a much more dreary place.

3.  The West Coast would be a much more desirable place if it had a Manhattan, or perhaps two mini-Manhattans.  You could imagine one in downtown LA, and another on the eastern side of San Francisco.  If these were built (say 10 square miles each), then more than 99% of the LA metro area would remain just as suburban as it is now, indeed maybe even more suburban than if a mini-Manhattan were not built downtown.  Thus there would be less pressure for 4 story apartment buildings out in the neighborhoods if developers could build dense housing downtown at low cost.

Think of it this way.  Lots of parents don’t let their kids play outside by themselves, because other parents don’t let their kids play outside.  If you choose to be the exception, then (unlike during the 1960s) your kid is the only one available for pedophiles to prey upon.  Lots of the anti-NIMBY feeling comes from a false perception of what the real estate market would look like if complete laissez-faire were adopted, based on the current distorted market. Actually, the vast majority of America, even the vast majority of expensive metro areas, would look roughly the same.  The current tight controls on development make developers anxious to build upwards on any scrap of land in the Bay Area where they can get permission to build.  But there’s only so much demand.  If they could build anywhere, they would not build everywhere.  Next time you visit Manhattan, walk around some of the residential areas.  You may visualize Manhattan as wall-to-wall high rises, but there are actually lots of neighborhoods with low-rise townhouses.  Even in Manhattan, there wasn’t enough demand to make the entire island look like the Upper East Side.

Pressures at the margin in a highly distorted market are not evidence of what a completely free market would look like. As an analogy, the “free market” part of health care in the US or UK is nothing like what a truly free market health care system would look like in the US or UK.

Density advocates will never “win” in the sense of replacing most of our low density suburbs with dense metro areas, because there are nowhere near enough Americans to make density anything other than an exotic exception to the boring suburban rule.

4.  New development can also make cities nicer. The recent construction of lots of housing in the inner parts of Boston and Cambridge (along with the removal of rent controls), has made those cities much nicer places to live (as compared to 1982, when I first moved to Boston.)

I live in a quiet Orange County suburb—does that make me a hypocrite?  No, I support YIMBY policies in my own back yard.  I would not be happy if a skyscraper were built next door, but I support the right of developers to do so.  Of course a skyscraper next door would also make my property much more valuable.  The main reason I’d be unhappy is that I’d have to take my profit and then face the hassle of moving.  I get that, but progress always involves some inconveniences.  The main point here is that suburban living in America is quite nice, and even with no zoning restrictions the vast majority of Americans would choose to live in quiet suburban neighborhoods. That’s not going to change.  But those who want density deserve a few Manhattan-type neighborhoods to choose from, in NYC, Chicago, Miami, LA, San Francisco, and a few other places.

Earlier I said that I did not understand Alexander’s point about agglomeration.  He seems to suggests that agglomerations such as Silicon Valley may reflect what Eliezer Yudkowsky calls “inadequate equilibria”, a scenario that society stumbles into, is not efficient, but is hard to get out of.  But there are smaller tech hubs in lower cost areas like Austin, Seattle, and Raleigh.  If the gains from agglomeration were not real, why would so many tech companies choose to locate in relatively high cost Silicon Valley?  Alexander understands this point, and suggests that maybe agglomeration occurred in the wrong place:

First, it could be that centralization happened in the wrong place – that, if anyone had been able to centrally coordinate, the tech industry should have ended up in Austin or somewhere else that’s well-planned and has lots of geographical room to expand into. Second, it could be that centralization is just a game of keeping up with the Joneses. If there were no San Francisco, then some company would still end up employing the best programmer. But given that there is a San Francisco your company might have to move to San Francisco or have no chance of luring them away from all the companies that have.

But then what’s the plan?  If YIMBY policies are bad because they are disruptive, then moving Silicon Valley to Austin would be ten times more disruptive, decimating the economy of the Bay Area and also destroying the tax base of California.  You might respond, “How about moving part of the industry?”  But then you’ve missed the entire point of agglomeration.  You either have a major agglomeration (with its huge efficiency advantages), or you have a bunch of small, scattered tech hubs.

PS.  Obviously I agree with Alexander that NIMBYists should not be accused of being racists who don’t care about the poor.