I also posted this on LessWrong.

If you want to change the world, start by giving people free books. Or at least, this is the approach of the Adam Smith Institute, a free-market think tank in the UK. At an event of theirs, they gave me a copy of John Myers’ little book (or long paper) YIMBY: How to end the housing crisis, boost the economy and win more votes. Weirdly, the book starts with specific solutions to the housing crisis, and then only in the middle talks about why this is a problem you should care about.

You probably already know this story: NIMBYs (Not in My Back Yard) want to preserve the value of their homes, so they lobby the government to restrict supply through zoning laws, historic preservation rules, and widely distributed veto power. Hence, when demand goes up, prices increase. YIMBYism (Yes in My Back Yard) is a recent movement of people who want to allow more building to solve this (the author is a cofounder of London YIMBY).

I feel that whether you view NIMBYism as a problem of excessive government intervention or civil society is dependent on your pre-existing biases. I take a central claim of libertarianism to be that the government is overrated, and civil society is underrated. But if we had been trying to show the opposite, we could use the fact that civil society obstructs building so much to argue that the government needs to socialise more housing.

This gets at my general confusion about decentralisation arguments. There is a well-known thesis that decentralisation is good because it allows for the synthesis of information from many different sources, in a way that couldn’t plausibly be done by central authorities. The problem of NIMBYism is a problem of decentralised town councils, part of the centralised government, being lobbied by decentralised homeowners, to obstruct the ability of decentralised individuals to pay centralised developers to build new buildings. For whom is this a win – the centralisers or the decentralisers?

There’s a paper from Hsieh and Moretti that finds that US GDP growth between 1964 and 2009 was 50% lower as a result of restrictions on building. If we take this at face value, it’s pretty staggering. A standard result in the economics literature is that restrictions on the supply of housing cause a doubling in house prices. For London, they seem to correspond to a quadrupling in rents (these figures come from the book). Apparently, we’re not in flying cars colonising the galaxy because we wasted all our money on rent…

…or because we’re living in the wrong places. There are two problems with supply constraints on housing: waste and inefficiency. The average UK renter pays 30% of their income on rent, so a crude guess is that society is paying a 15% tax on economic activity to pay for restrictions on building. But this effect is plausibly smaller than the problem of people not living where their labour is most valuable. This is why you get software engineers in the Bay Area doing their own plumbing: few plumbers can afford to live in San Francisco. Then you get an oversupply of service workers in economically stagnant areas, driving down wages… you get the idea. 

I’m not sure whether anyone actually believes that there were more restrictions on building in the past. Britain’s modern planning system didn’t exist until 1947, with the passing of the Town and Country Planning Act. Britain now has possibly the most dysfunctional planning system of any rich country (except maybe Ireland?). Is this because the Town and Country Planning Act is just really bad? Or does it work as poorly as systems from other countries, but the nicer pre-existing supply of houses in the UK inspires more NIMBYs?  

Houston famously has no zoning laws (it does however have other planning regulations including parking minimums that make the city less walkable). Houston represents one equilibrium, in which the city has enough renting tenants (44% of the population) to form a permanent bloc to vote for more housing. The South of England represents the opposite equilibrium, in which there are enough homeowners to consistently block new developments. 

This book focuses mostly on the UK and London in particular, where expansion is strictly limited by the green belt. It’s safe to assume that the primary reason for the popular support of the green belt is that it contains the word ‘green’. Most of London’s green belt is occupied by farmland, which uses pesticides and is generally bad for the environment. The green belt doesn’t exist because it’s an area that is particularly beautiful (people who think this are probably confusing it with an Area of Outstanding Natural Beauty). Nor does it exist for specific scientific or environmental reasons. Its function is primarily to restrict the supply of housing and ensure that Londoners in the suburbs have access to nice outdoor areas. Having access to nice outdoor areas is important, but this is a spectacularly inefficient way to do it. 

An angle I’m surprised the book didn’t push harder is that zoning laws are almost certainly harmful to the environment, because they limit densification, and higher-density buildings are better for the environment. For example, carbon emissions per capita in London are almost 40% below the UK average. When you think about it, this is entirely unsurprising: cities have lots of people living close together using public transport who don’t need goods transported to remote locations.

Another notable omission is the effects that high house prices have on fertility. One of the most important things I learned from reading Matt Yglesias’ book One Billion Americans is that women in Western countries’ ideal number of children is not, in general, going down. If anything, the number of children the average American woman wants is slightly up since the 1980s. What’s changed is the economic circumstances: women make more money, so the opportunity cost of having a child is higher. There is also a Baumol’s disease driving up the price of education and childcare: some sectors (like software) have increased in productivity so much that compensation has had to rise in low-growth sectors to keep people from jumping ship. Childcare may even be experiencing negative productivity growth, because of various laws restricting the number of children that can be taken care of by one carer (in my native Ireland, depending on the circumstance, the child:adult ratio can be no larger than 3:1). But plausibly more important than these effects is that couples have to wait longer until they can afford a house, and they don’t want to have children in a cramped expensive apartment. The evidence for this looks pretty solid to me, although the book doesn’t discuss it. I assume the reason that YIMBYs don’t make this argument more often is that more children are not an unalloyed good; some people even say that having children is a bad thing because of climate change. The argument that people really do want more children, but that we price them out of being able to do so through a screwed-up planning system, is a pretty good one. 

The ASI and London YIMBY propose ‘street votes’ as a practical solution to increasing the supply of housing, in which each street can vote to ‘up-zone’ and allow for denser development, thereby greatly increasing the value of their property. This would be done through a double majority: it would be approved by two-thirds of residents and two-thirds of residents that have lived on the street for at least three years. The hope is that this would replace the current vetocracy, in which small numbers of cranky residents can oppose new developments while not clearly violating any ‘will of the people’.

The author doesn’t consider the other proposed solutions to this problem, but that is mostly because the other proposed solutions are silly. Rent control, for example, is almost universally condemned by economists as being a terrible idea. Even the Freakonomics guys once released a plea for people to stop supporting rent control. Obligatory reference to the time Assar Lindbeck said that “Short of bombing, I know of no way to destroy a city that was more effective than rent control”, et cetera et cetera. 

I’ve also heard people suggest that second homeownership be heavily taxed or banned altogether. But second homeownership is already pretty rare; vanishingly so in expensive cities. Insofar as there are unoccupied units in desirable cities, it is usually because someone lives there during the week for their job and elsewhere at the weekend. The reason they do this is that rent is very high and the rental market sucks because of (you guessed it) insufficient supply. Advocating a policy with mostly symbolic value against the elites is not usually a good way to solve problems. 

You might be wondering about whether the dilemmas caused by NIMBYism can’t be solved with Coasean bargaining. Coase’s theorem says that, in the absence of transaction costs, rational actors will bargain to efficient outcomes. If my flatmate likes listening to loud music that I don’t like, theoretically she will compensate me exactly to the extent that I am harmed by listening to her music. Similarly, you could imagine developers compensating nearby residents to the extent that they are harmed by living near an ugly building and/or an irksome construction site. I’m assuming that the reason this doesn’t happen in practice is that it’s very hard to say what counts as ‘nearby’, it’s hard to make this compensation in a legal way, and that lots of NIMBYs oppose developments nowhere near where they live. 

I don’t generally recommend making people feel bad for supporting something because the initial motivation for that thing was racist. But still, it’s worth mentioning that single-family zoning (the primary form of zoning in the US) was designed to exclude black people from white neighbourhoods. Something similar is true of the American college admission system, by the way: it started out as an elaborate scheme to keep Jews out of universities.

In a competitive market, the cost of a good falls closer and closer to the cost of replacing it. And indeed, for centuries, house prices outside city centres generally hovered above replacement levels. Now, houses cost at least double their replacement cost, and many multiples more in cities. The ONS says that the market value of houses exceeds replacement costs by £3 trillion, which is 150% of the UK’s GDP. The book says £3 trillion is “more than double” UK GDP, but this is in violation of what Google tells me. 

I’m not sure how distinct a problem high house prices are from homelessness. The rent in San Francisco is very high, and there are lots of homeless people there. But presumably, most of them still wouldn’t be able to afford to live in an apartment even if the rent were half as much. Is the reason that NIMBY contributes to homelessness that it blocks the development of ultracheap tenements for homeless people? If so, this seems like an odd point to neglect in a book about housing regulations. A few years ago a man in LA built tiny $1,200 houses for the homeless, before the government shut him down, presumably to free the homeless people from oppression or something, and then they went back to sleeping on the streets. Maybe someone needs to make a tear-jerking documentary about this, Blackfish style. If only people were as cute as killer whales…

Thanks to Sydney and Gytis Daujotas for reviewing drafts of this post. 

P.S. I wasn’t aware of John Myers, Ben Southwood and Sam Bowman’s new piece in Works in Progress when I wrote this, but it covers very similar ground. 


2 comments, sorted by Click to highlight new comments since: Today at 7:47 AM
New Comment

Good post. If you're interested, the "Street Votes" idea was developed into a detailed policy paper by another UK think tank, and then proposed to Parliament by some MPs. Shortly afterwards the housing minister was sacked and a new minister appointed, so things could go either way.

Founders Pledge also did a great report on this topic, where they try to estimate the extent of the problem (in the UK) and the benefits of fixing it.

Another cost-effective intervention in this area is land records digitization in Bangladesh which has, according to the Copenhagen Consensus Center, 619x (61,900%) social, economic, and environmental return on investment. EAs looking for ITN-scoring application of their coding skills should feel free to take a note.