Latest / 80,000 Hours Podcast / #211 – Sam Bowman on why housing still isn't fixed and what would actually work

Transcript

#211 – Sam Bowman on why housing still isn't fixed and what would actually work · 80,000 Hours Podcast

• Sam Bowman: There’s a tunnel that people want to build under the Thames. This would join up Kent and Essex, and would allow the two largest seaports in England to have road haulage between them — so really economically valuable. This tunnel has not begun construction yet. It’s been in public consultations for the last seven years. The planning document has run to 360,000 pages, and the cost of producing the planning documentation has been £297 million. That is double what it cost Norway to build the world’s largest road tunnel — not the planning application, but actually to build the world’s largest road tunnel. So the cost of permitting, of approvals, has risen way out of proportion to the value that we’re getting from it. And that, we argue, is one of the really big blockers to the investment and to the building that the UK needs. Rob Wiblin: Today I have the pleasure of speaking with the well-known writer and economist Sam Bowman. Sam is currently the editor of Works in Progress, which has given itself the remit of covering underrated ideas to improve the world — so, naturally of great interest to 80,000 Hours listeners. It’s a magazine that I think started a couple of years back and is somewhat associated with the progress studies movement or ecosystem that we’re going to talk about a little bit later. I think a lot of listeners to the show will know of Sam, whether they realise it or not, because he’s one of the coauthors of this very significant and viral article called “The housing theory of everything” — which details the enormous and varied harms that are being done by restrictions to housing construction in many countries at the moment. And some others will be very familiar with Sam from his prolific and very entertaining output on both Twitter and his Substack, which is called Consumer Surplus. I found out in the process of preparing for this that he also studied a double major in economics and history, which I think goes some way towards explaining why he’s such an interesting person to talk to on almost any topic that you can pick. I’ve been hoping to interview you for years, so thanks so much for finally coming on the show, Sam. Sam Bowman: Thanks for having me. It’s great to be here. Rob Wiblin: We’re going to talk mostly about how to build a democratic majority in favour of changes to society or the economy or technology that produce a net benefit, but produce some losers. But first, what’s been the main focus of your research over the last couple of years? Sam Bowman: I think the main thing that I’m interested in is, A, why is it that almost every Western developed country has stagnated? And B, what is the relationship between that and our seeming inability to build almost anything? I think that the fact that we don’t build much anywhere — especially in the UK, especially in places like California, New York, large parts of the US (although not all of the US) — I think that’s a really important factor in this sort of broader stagnation. And I don’t just think it’s related to economic growth; I don’t just think it’s related to innovation rates appearing to slow down. I think that it’s also related to the fact that birthrates are falling, climate change is worse than it could be. There are other things — like obesity, and our health is worse than it could be. My view is that not building enough houses in particular — but also transit infrastructure, and in some cases energy supply, is really important in causing that. And the reason that I’m optimistic about this stuff, and that it kind of compels me to keep going, is that I think all of this is really eminently solvable. It’s not a technological problem; it’s really a political problem, and it’s really an incentives problem. And maybe that’s what we can talk about today. Rob Wiblin: Yeah, absolutely. I think most listeners, maybe a great majority of listeners, will be familiar with issues around NIMBY, and difficulty getting permissioning to build housing and to build other kinds of things in the US and UK. So we’re mostly not going to focus on the nature of the existence of the problem. We’re going to spend most of our time thinking about creative solutions, given that people have been aware of this issue for some years, maybe for some decades, and it hasn’t yet been solved. In some ways, we haven’t made nearly as much progress as we might like or would expect to happen. But let’s at least start by talking about how large the scale of the problem might be, to give people a sense of just how much harm might be getting caused here. Because you can understand that there’s a problem, but maybe not appreciate just the full range of damage that’s being done. So what are some of the harms that have come from restrictions on housing construction, in particular the US and UK, that you detail in “The housing theory of everything”? Sam Bowman: So while the UK and the US have, in some places, very similar problems, on a nationwide level, they actually have quite different problems. The UK pretty much has everything in one city. London is 10 times larger than the next largest city in the UK. It’s basically everything in one: it’s the political centre of the country, the financial centre, the economic centre. Most of the major industries in the UK are centred on London. So when London has a problem, Britain has a problem. Now, contrast that to the United States: the US is much larger, obviously. It’s about five or six times larger by population. It has lots of very large cities. It has two really outstandingly important cities, from an economic point of view — maybe three, depending on how you count — which are New York, Los Angeles, and the San Francisco Bay Area. San Francisco by itself is too small to count, but the San Francisco Bay Area, which is lots of cities around the Bay put together, I would say probably do count. Now, those places have huge problems getting things built, but in most of the rest of the United States, there aren’t nearly as strong of problems with getting things built, especially actually getting housing. So 80% of US cities do not have a big gap between the cost of building houses, at least in some parts of the city, and the price of selling houses. And that’s a measure I’m going to use again and again, because it’s a really useful way of using the price system to kind of estimate the regulatory constraint. So just to repeat: what I’m saying is the cost of building a house might be something like $150,000 or $200,000. Then how much you can sell the house for might be, depending on where you are, anything from around say, $200,000, $250,000 all the way up to $1 million or $1.5 million. That gap is a really good proxy for regulatory barriers to building, because if those barriers weren’t there, then people would just build until the price came down and they made lots of money. So in 80% of US cities, there are at least places in the city where that gap does not really exist. It’s a very small gap and you can probably explain it just through normal kind of economic frictions. That gap is huge in places like San Francisco, it’s very large in LA, it’s very large in most of New York. It’s absolutely enormous in London: in some parts of London, the gap is a factor of four or five: you can sell a house for four or five times more than it would cost to build. The reason I say that the US and the UK, while they have similar problems, the nature of the problem is quite different: because the US is so large, and because its economy is so diversified across the country, while these first-tier cities are constrained massively, and while that stops a lot of growth from happening (and we’ll talk about why in a minute), there is still an overflow — people, even if they can’t move to Los Angeles or to San Francisco, they can move to Austin or Chicago or Miami. These cities are not as productive as San Francisco or Los Angeles, but they’re very productive. The UK doesn’t have that. So when we talk about the UK, we’re kind of talking about the limit case of what happens when a country puts all of its eggs in one basket, and then that basket becomes catastrophically broken, to use the metaphor in a slightly funny way. So I think there are lots of interesting commonalities. And when I talk about the UK — which is where I live, and where I have done most of the most amount of research — I think it’s most useful as a kind of a limit case of how bad things could get. Maybe it’s worth unpacking the intuition about why it’s such a problem for people to not be able to move to certain cities. But hopefully that gap between price and costs is a useful frame for just beginning to understand that there is a problem — and then we can talk about maybe why it’s such a big problem. Rob Wiblin: Yeah. So the fact that to buy a house or to rent a house costs, in many cases — if you want to live somewhere where you actually have good economic opportunities — four or five times the cost price, it’s easy to understand how that is impoverishing people. That’s how you end up with people spending 30%, 40%, 50% of their entire income just on rent, just to be able to live somewhere that they find tolerable. Moving on from that basic harm, what are the major damages that it causes? Sam Bowman: I think almost everybody, when we think about housing, we first of all think about it as a budget problem. That’s completely natural. For almost everybody, your house is the most expensive thing you will spend money on, unless you’re lucky enough to have paid off your mortgage. It’s usually most people’s number one asset if they do own their house. And across the economy, it’s the number one thing that we spend money on. It’s the biggest single item that people spend money on. So it’s obvious why you think about the first-order effects of expensive housing being that we have less money to spend on other things. The second-order effects that we focus on when we talk about the housing theory of everything are what happens when it isn’t just a fact that you’re spending more money on living in a certain place, but you can’t move to that place in the first place — either because there just simply aren’t the houses there to house you, or the costs are so high that you wouldn’t be willing to spend the money on living there. What happens then is that lots of the jobs you might have been able to access are no longer available to you. Economists call these “agglomeration effects“: the economic effects of pooling people in a particular place with each other, and the benefits that you get from doing that. Cities are a particular kind of agglomeration that give you both scale — you could get lots of people — and get you complexity: you get lots of different types of skills. Now, when we don’t build enough houses in a particular city, the first or one of the main second-order effects is this productivity effect — where your wages might be 10% or 20% higher if you moved to London and did a very similar job to the one you do now. If you can’t afford to move to London, your wages would just be lower. And so overall, the aggregate effect of lots of people not being able to move to London or San Francisco caused this reduction in people’s incomes. Now, there are different ways of estimating the aggregate effect of this. Unfortunately, one of the really well-known ones turns out to have quite a lot of computational errors in it. Luckily it wasn’t a load-bearing study, because there are others that use different methods. For example, as two economists, Duranton and Puga, have done, you can try to estimate how many people would move to San Francisco, New York, San Jose, places like that, if there wasn’t a constraint on building houses. They estimate that you’d get 40 million people moving to the greater New York area, and you’d get fewer, but still very many, people moving to the San Francisco Bay Area. And the wage uplift that people would get from this, they estimate to average out about $16,000 per year per capita in the US — which is absolutely gigantic. It’s kind of unimaginably large. And I think it’s more interesting from a directional point of view than it is from a kind of precise… It’s very difficult to say that that exact number is correct, but magnitudinally it makes sense. So one cost is this static productivity cost. I think just as important, or maybe more important, is the innovation cost. Looking at patents and patent applications, there is a fairly decent body of evidence that suggests that cities and density lend themselves to more innovation. Most industries, especially ones that are ideas-based or kind of intangible-capital-based, want people to be around each other. It’s better for ideas being exchanged, it’s better for people from different departments learning from each other. I went to, in fact, the lab of a biotech company in London a few weeks ago. They work on regrowth of bone tissue and other human tissues. I found it really interesting, because it was kind of a microcosm of what I think about housing economically. Because, number one, it’s really important for them that their lab is very close to hospitals, because human tissue that they have to work from dies very quickly, and so they have to be able to transport it really close, really quickly. So that’s one that’s kind of interesting from a scientific point of view. It’s kind of interesting if you think about road speeds and things like that. But let’s put that to one side. More interestingly, for these purposes, this company primarily uses biologists, medical scientists, and machine learning engineers. And they have designed their lab, which is in the middle of London, in Euston, to have machine learning engineers and medical scientists, a wet lab and a dry lab, on the same floor, in the same premises, so that these two different people from these two different departments interact with each other face to face, constantly, again and again, and learn from each other. And what we see when we look at the empirical evidence is that unconventional breakthroughs are more likely to come from mixed urban areas than they are to come from, say, business parks, where you just get all companies of the same type being next to each other. What this kind of thing suggests, and there’s lots of historical precedent for this — mediaeval Florence, Amsterdam in the 17th and 18th centuries, obviously London, obviously Paris in the 19th century, New York in the 20th century — is that having lots of different people in a relatively large, concentrated labour market leads to lots of innovation, and lots of people exchanging ideas and coming up with breakthroughs together. So if this is right, then it suggests that holding back cities isn’t just a problem because it stops people from getting higher-paying jobs, but also that it stops people from meeting a potential cofounder of a company, or from meeting the CTO of a company that they might have founded, and from coming up with unconventional breakthroughs that they might never have thought of had they not been in a bar with somebody after work who turns out to be from a totally different field but have an overlapping interest. I think that innovation effect of housing constraints is something that’s really important and interesting. A little bit underexplored, but potentially huge. Rob Wiblin: To avoid getting stuck just describing the scale of the problem, I’ll just quickly name check some other issues that are raised by housing construction restrictions. As you alluded to, it’s a major driver of regional inequalities: it’s a big part of the reason why you have people on so much lower incomes in northern England, for example, while people in London seem incredibly wealthy. It’s a huge driver of interpersonal inequality and wealth inequality. I think Piketty and Saez documented these substantial increases in inequality in wealth and assets that people hold. And when people looked into it, it turned out that maybe almost all of that change was driven by changes in real estate values — where people who owned real estate were becoming incredibly wealthy as there was a run-up in real estate prices, and that was just making them so much richer than people who hadn’t gone into the property ladder in time. It’s also bad for health, potentially, because people are stuck in suburbs where they drive rather than walk. I think Manhattan has like a third the obesity rate of the average of the United States, and it’s almost the only place there where you can easily live a car-free lifestyle. It’s bad for fertility, potentially, because people just don’t start families if their houses are not large enough, or they can’t afford a house in which they can actually give their kids a room. Are there other themes that are missing that we could quickly mention? Sam Bowman: The one that I don’t find well founded, but I do think is an interesting hypothesis, is political polarisation. And, you know, we’re speaking in the aftermath of the US presidential election, so we will need to check to see whether this is consistent with those results. It’s too early to say because the exit polls haven’t been properly adjusted yet. But certainly when we look at the UK, when we look at France, what we find is that this regional inequality point — where people are basically trapped in their hometowns because they can’t afford to move to more prosperous places — those places seem to be the most acutely driven to what you might call “radical alternatives.” I don’t want to talk about whether they’re good or bad, but they’re definitely outside the mainstream. I think that there is quite a lot to be said for, if your experience of the modern world is a crap expensive apartment, sharing with people you don’t know; no stable relationship, because getting married and having kids, while you might have done this, and you might do this if you were living in a more easy supply area, you don’t; and a job that’s fairly low paid and insecure: that’s a pretty crap experience of capitalism. I think it’s fairly reasonable to infer that you might not be very happy with the world as it is right now. And that might explain why there is a very large and widespread discontent with capitalism. I don’t want to overrate that. I think it’s an interesting theory and an interesting hypothesis, but it’s just an idea. Rob Wiblin: Yeah, yeah. It’s probably a nonzero effect. Surely some people have experienced this. But we can’t exactly say how many from the data that we have. Sam Bowman: And the way I think we should think about housing generally, and housing constraints generally, is that they determine where people can be. And where people can be is one of the most important things in anybody’s life — and on aggregate, in everybody’s life. Rob Wiblin: OK, before we get to talking about solutions, this isn’t a show about the UK in particular, and not all of our listeners live in London or the UK like we do. But I want to elaborate a little bit on just how crazy and just how destructive not just restrictions on housing constructions, but restrictions on basically building, constructing, or almost doing anything in the UK have gotten — because it’s maybe a canary in the coal mine; it’s an example of these restrictions, when they’re taken to an extreme level, just the amount of damage that can be done on an aggregate level. So let’s flesh that out a little bit. What is the planning commissioning system, or what restrictions are there on construction or getting permission to build things in the UK? Sam Bowman: So in this paper, “Foundations,” that I coauthored recently, we pointed to housing shortages, infrastructure shortages, and energy supply shortages as being the three cardinal sins of the UK economy. With housing, by far the biggest problem (although not the only one) is what we call the planning system, and what in America you would call the zoning system. Here we have a very discretionary system where, if you want to build something, you have to go to your local authority — like your local council or your borough council — and say, “This is what I want to build. Can I have permission to do it?” And almost always you will not get permission to do it. And when I say “almost always,” I mean for any imaginable project you might have. Most of the time people don’t even bother making submissions. So you get this kind of weird compositional effect, a kind of survivor bias thing, where it appears that we have a very permissive system — Rob Wiblin: Because nobody even bothers to ask. Sam Bowman: Nobody even bothers to ask. The only times you will ask are because you are certain that it’s worth putting all the money into the application, the legal process, and so on. So number one, the problem is it’s really, really, really hard to get something built — because it’s really, really, really easy and cost free to local authorities to stop you from building. We also have things called green belts, which are areas outside of English and Welsh cities where you basically just can’t build anything; it’s kind of determined by central governments that you can’t build on these things. But I think by far the biggest problem is that local authorities basically have no reason to let you build. That’s housing. Infrastructure suffers from similar problems, but infrastructure also has a very centralised, and I would say non-cost-conscious, approach to building it. So when we do build infrastructure in the UK, we do what’s called “gold plating”: we try to build the absolute best thing imaginable. I took the Elizabeth line over here. I love the Elizabeth line. It was an amazing feat of engineering. But it’s also the second most expensive metro line ever built in the world. The only metro line more expensive is the Second Avenue subway in New York City, which has suffered from very similar problems. And that has incredibly ornate, architecturally beautiful stations. I actually took a photograph going down into one. I don’t consider it beautiful, I have to be honest, but they’re considered by architects to be beautiful. They’ve won loads of awards, these stations, and some of them have cost between £500 million and £1 billion — just for the stations. The Elizabeth line is an amazing achievement, but I would really like to have like five Elizabeth lines, or a number of Elizabeth lines, many not in London. I would like Birmingham and Manchester to have Elizabeth lines, or their equivalents of them. When you look at comparable projects that were much cheaper in building, like the Madrid Metro, which was delivered in the 1990s and 2000s and 2010s for about a quarter of a price or less per mile than the Elizabeth line: they were really, really pragmatic about costs. The stations usually had a single design that they just built every station according to. They’re not that pretty, I have to be honest, having been there last year. But there’s nothing more beautiful than an affordable and abundant metro line, in my opinion. So that’s one. Another is that we don’t queue up projects, so we don’t have a constant pipeline of new projects. We do them on a very one-off, ad-hoc basis. Another is that, because nobody really has an incentive involved in the project to keep costs down, they actually over-incentivise people to complain and they’re over-willing to buy off objectives — and we’ll talk later on about objections and things like that. And I do think buying off objections is really good and really important, but you can do too much of it, and you need to balance buyoffs with some cost focus. And if you don’t have a cost focus, then you end up with a situation like we’ve had with High Speed 2, HS2 — which is this high-speed rail line between London and Birmingham and some of the other northern cities — where a quarter of the route is tunnelled through the countryside, simply because there were objections from local residents that they didn’t like the thought of having to look at a train going past. You know, really crazy things. And this is a project that has ballooned from initially £20 something billion to now over £100 billion. So we have all of these small problems that add up to very, very expensive infrastructure, plus loads of veto points where you can challenge the project on the basis of its environmental impact, or on the basis of its effect on local residents, things like that. And then the third problem that we’ve got, energy, has some of the problems that I’ve just talked about with infrastructure. And we’ve also made, I would argue, mistaken national-level decisions about what type of energy we’re going for. So we have ended up deciding that wind and solar are going to be by far the largest share of our low-carbon energy in the next five to 10 years — whereas, had we done that with nuclear and focused on getting the cost of nuclear down, we may have been able to get, and we may be able to get, electricity at a much lower price. So that is a kind of combination of a strategic decision that I think has been mistaken, plus the costs of nuclear being allowed to get extremely high. Rob Wiblin: Yeah, yeah. We’ll come back and talk about nuclear specifically later on. At a broad level, what is the outcome? I think housing construction for about 50 years has been lower than it was in the ’30s and than it was for most of the period of the 19th century, despite population growth being about the same level or higher. We have the biggest city in Europe that doesn’t have any public transport system whatsoever. Despite there being enormous demand to live here, London’s population almost hasn’t grown at all. And we’ve been barely constructing any more public transport infrastructure; we’ve just been struggling with what we already had — apart from the Elizabeth line, which, as you mentioned, is the second most expensive line in the world. Are there any other kind of high-level stats that you can throw at people to indicate just how rough things have gotten in the UK? Sam Bowman: The biggest one is that the industrial price of electricity has more than doubled in price-adjusted terms in the last 20 years. That, I think, is an incredibly important fact, given how much we focus on industrial productivity in the UK, how much people want there to be some manufacturing in the UK, how much attention is given to industrial policy. And this is true across the Western world. The Draghi report, which is a recent report by Mario Draghi, who was the governor of the European Central Bank, into kind of what’s gone wrong for the European Union or for Europe, highlighted that even in the last three years, since the Ukraine war, there have been huge reductions in output of manufacturing sectors in Europe that are energy intensive. So things like chemicals, paper, cement, stuff like that that’s very energy intensive, has fallen in output by like 15% in three years — which is Great Depression levels of output reduction. Now, our numbers stop before the Ukraine war. The Ukraine war has made things even worse. But I think it’s important to emphasise that we are not factoring any of that effect in. We are just talking about what happened between 2003 and 2021. That’s one thing. I think another: the number of flights going in and out of Heathrow has not risen since 1990. The number of people has, because we’re using bigger aeroplanes, but the number of flights hasn’t risen. This is because we haven’t built another runway and we’re at capacity. We haven’t built a reservoir since 1992. Maybe I’m being a bit glib here, but they’re like the easiest bit of infrastructure to build. They’re ultimately a hole in the ground with water in it. We have all built reservoirs at the beach — and they’re slightly more complicated than that, but they’re not that much more complicated than that. And reservoirs really should be kind of positive for local people, like a lake. The existing reservoirs we have are tourist attractions in some cases, because they have water sports and things like that on them. And yet we haven’t built one since 1992. The nuclear reactor at Hinkley Point C that we’re building is going to be the most expensive nuclear reactor by megawatt ever built in the world. The new one that we’re building at Sizewell C is estimated to not be quite as expensive as that, but it’s still going to be far more expensive than any other reactor we had built in the UK, up until the 1990s when we stopped building them. All of these come together to point to a huge shortage of the basics in the economy — the foundations, as we call them. And I think it’s reasonable to infer that a country like the UK — that relies on people more than anything; it relies on people being able to work in a services- and ideas-driven economy — a country that has a perennial long-running shortage of investment… And when we talk about investment, we have to remember that houses and energy supply and infrastructure, they’re not just the things that get you investment, they are investment. There’s this weird disconnect between investment as people talk about it in the public debate, and investment as it actually is — which is spending money on building things. Sometimes it’s factory equipment, but it’s just as often — or just as often should be — railways, roads, energy pylons. The final example that I think is really powerful is there’s a tunnel that people want to build under the Thames. It’s called the Lower Thames Crossing. This would join up Kent and Essex, and would allow the two largest seaports in England to have road haulage between them — so really economically valuable. This tunnel has not begun construction yet. It’s been in public consultations for the last seven years. The planning document that has had to be produced for this tunnel has run to 360,000 pages, and the cost of producing the planning documentation has been £297 million. That is double what it cost Norway to build the world’s largest road tunnel — not the planning application, but actually to build the world’s largest road tunnel. So we have a situation where the cost of permitting, of approvals, has risen way out of proportion to the value that we’re getting from it. And that, we argue, is one of the really big blockers to the investment and to the building that the UK needs. Rob Wiblin: Yeah. So I can definitely recommend that listeners, if they would like to hear more, they could go to ukfoundations.co: the essay is “Foundations: Why Britain has stagnated.” Maybe this actually wasn’t in that post, but it was something else that you wrote that actually blew my mind the most: in London, there’s 90,000 square feet of lab space available; in the Boston area, there’s 14 million square feet of lab space. You can just imagine if you’re trying to grow a scientific enterprise in the London area, there’s almost nowhere that you can do it. How much is this going to limit our ability to actually be at the cutting edge of technology or just do useful work? Sam Bowman: Totally. And something that I find really interesting when I talk to scientists, especially in Cambridge, because Cambridge has similar constraints — Cambridge actually has slightly more lab space, but it has a very important biotech sector — one of the things that scientists have said to me is that lab space availability doesn’t just affect how much research they can do; it affects what research they do, and how unconventional the research they do is. Because the way grant funding works is you basically submit a kind of sure bet, something that will hopefully deliver a result. It’s probably not going to be a high variance bit of work, but you’ll make a little bit of progress with it. You get the money for that. If lab space is constrained, expensive, then that’s all you do. If lab space is free, or freely available and inexpensive, then you probably run other things in parallel. You have a bit of extra time, you’ve got the lab there, you maybe have a much more wild bet kind of idea that you would never be able to get grant funding for, but you can kind of run at the same time. So when lab space is there, you may end up getting much more radical, innovative, bold experimentation alongside the stuff that we’re paying for. When lab space is constrained, you’re just going to get the kind of boring, conventional, move the world forward by a tiny fraction of an inch kind of work — that sadly we fund, and sadly we are prone to that kind of risk-averse type of work. But the knock-on effect of lab space shortages could be that there is a considerable constraint on the kind of science that we want to get done. I think that’s really, really fascinating, if true. Rob Wiblin: Yeah. So the bottom line of the essay… Or actually, you start out pointing out that the issue is that the UK, for the last 15 years, has had just dismal productivity growth, 0.3% or 0.4% per year — while other countries have been doing 1 or 2% — which adds up to us being about 25% poorer than we would be if we had continued the productivity growth that we’d had in the previous 20 years before 2008. Then you point out that that’s interesting — so what could be the reason investment in the UK is really weirdly low compared to other countries? You ask, why would that be? And then you just look into it. And of course you find it’s because it’s all illegal — basically that most of the investments that would be most productive for the UK simply are not allowed — so of course they don’t happen, and you end up with very little investment and very little technological and economic advancement. Sam Bowman: Absolutely. I don’t want to litter this with stats, because it’s kind of difficult in audio, but the best statistic I think to kind of prove that there is a big, big, big regulatory constraint on getting stuff built in the UK is: if you have a hectare of farmland in the southeast of the country, then it would be worth about £25,000 for the hectare. If you get permission to build houses on it, it will increase in value by about 180 times. You’re almost literally printing money to get permission to build houses in the southeast. It’s even more if you take land in London. There aren’t lots of hectares of farmland — there are some, amazingly, in London, but there’s not a huge amount of farmland. For that, notionally at least, the uplift is about 1,400 times. So you’re talking about nearly printing money just by making it possible to build houses. Rob Wiblin: Yeah. So I promised that we would get to solutions and be thinking about how we can actually fix this, given that there are significant barriers to fixing, or at least the brute-force approaches that have been tried have had only middling success. So these issues exist in other big metropolitan areas — places like the Bay Area, New York, probably Sydney, I would guess Barcelona — Sam Bowman: Almost every English-speaking country for sure. Vancouver has huge problems. Toronto has problems. Dublin has one of the worst housing shortages in the world. It also has terrible, terrible infrastructure. There’s a great French urbanist called Alain Bertaud who is really good on the relationship between infrastructure and housing — because clearly infrastructure enables people to move from place to place, in the same way that living closer to that place does. So if you have free-flowing roads, or if you have reliable public transport, infrastructure can be like a proxy for building houses, and actually give people the quality of life they might want, if they want to live in a more low-density area. Dublin has terrible infrastructure. You know, one of the reasons that London is as prosperous as it is is that we have incredible infrastructure. Some of the best, I think, in the world. New York has decent infrastructure — pretty impressive for North America, pretty poor by European standards, but it’s not bad. San Francisco has almost nonexistent infrastructure, and nobody uses the infrastructure that they’ve got because of, I think, mostly law and order crime problems. Sydney has problems, Auckland has problems, although they’re possibly beginning to improve them. But this is a huge, huge disease across the Western world, especially the English-speaking world. I do think it’s an interesting question as to why the English-speaking world in particular has these problems. There’s a really good piece by my friend John Myers, who I think is one of the world’s leading thinkers on how to solve these problems, incidentally. John Myers, who works on YIMBY issues in the UK, has a great article called “The plot against Mercia” (as a play on this Philip Roth novel, The Plot Against America). It’s about how Birmingham in the postwar period was an incredible boomtown, and looked as if it was possibly going to be a kind of counterweight, if you like, to London. And the government at the time considered this to be a huge problem, because it was sucking in economic activity from the rest of the country. In the same way that we talk about levelling up today, and we worry about London sucking in economic activity, they worried that Birmingham was going to suck in economic activity. So they really, in many ways, destroyed it. They built a highway through the middle of the city. You know, Robert Moses‘s craziest dreams would be realised in Birmingham. And it’s meant that we don’t really have second cities in the same way that America does. I think we could. One of the very convincing arguments that I’ve heard is that if we could devolve infrastructure spending on a per-capita basis to England cities, then we could unlock a lot of the… I mean, a lot of these places actually do have a lot of people living, as the crow flies; it’s just that the roads are so poor and there’s nonexistent public transit, or almost nonexistent public transit. So even if you live somewhere like Walsall, which is a kind of suburb of Birmingham, it’s really time consuming; it will take you more than an hour during rush hour to get into the centre of Birmingham. If we could widen the roads, if we could build metro lines, for example, or tram lines that had priority, we might be able to get more of the agglomeration effect and allow people to travel for better jobs. We just don’t do it. Rob Wiblin: Yeah, it’s unsurprising that England’s mid-tier cities are struggling, when I think we’re building virtually no public transport infrastructure in those places. So just people end up stuck in the suburbs where they are. So I think we’ve probably hopefully convinced people that there’s an issue here, and maybe the mystery is that this isn’t cutting-edge physics: it’s not so difficult to tell that if we ban constructing houses, and we ban almost all business investment, if we ban the construction of most infrastructure, this is going to create problems. People have been aware of this, and they’ve observed it getting worse and worse over the years. And the YIMBY stuff, I first heard about 10 or 15 years ago, and it’s gone very mainstream. Many, many people recognise this, but progress on it has been very limited, despite lots of people being persuaded intellectually. Why haven’t we fixed it? Sam Bowman: There’s so much in that question. So apologies if I now try to make seven arguments at once. I will try to order my thinking well. Number one is: I think we are making progress intellectually. Fourteen years ago, my old boss, Tom Clougherty, when I worked at a think tank, the Adam Smith Institute, told me we should be doing more on planning/housing issues. Which I found really weird. I was very young, very new to the game, but I don’t think that would have been an unusual [response]. I was a little bit surprised to think that, OK, this is important, yes, but I don’t really understand why we’re treating this as important as, say, tax or trade or immigration or something like that. Obviously, he changed my mind. He convinced me, and I think lots of other people have been on that journey as well. That process takes time. I think clearly, when you talk to people who are younger, who have direct experience of the rental market, they are much more aware of how broken it is and much more aware, probably from personal experience, of the knock-on effects of housing shortages. So I think there is something going on there. So it’s possible that what’s happening is just the kind of long political transition. We had, funnily enough, both presidential candidates in the US making some vague noises about making it easier to build. Sadly, neither of them I think had great ideas about how to do it. And some of them thought rent controls were a good idea — and rent controls are a really bad dead end that that kind of talking about housing or rent prices too much can lead you down. But you know, there is a shift going on. But I think there’s a more fundamental question. I’ve been talking a lot about prices, and I’ve been talking a lot about effectively the surplus — the amount of money that is created when we give people permission to build. The gap between the price of building a house and the price of selling the house, the gap between the permissions of having agricultural land and getting permission to build on it, to build houses on it, that is a gigantic surplus. So the question is: why is that surplus not being used to buy consent? That is the huge question. And there’s a really, really important paper called “The NIMBY problem” that I think gives a really useful model to understand what’s going on. Rob Wiblin: Yeah, tell us about that. Sam Bowman: The authors of this paper are two political scientists, and they argue that, in a very simple model, what would happen is that a developer would say, “I want to build a block of flats in this area. The cost would be a few million pounds. I could sell the flats for double that. The locals will fight me and object — but I can give those locals money, and I can give them enough that they won’t fight me and object, and I’ll still have a little bit left over myself as a profit.” That’s how basically all exchanges and interactions work. You know, when I go to work, I love my job, but one of the reasons I go to work is that my employer gives me money to do it, right? When I buy something from somebody, I am effectively saying, “Will you do this thing that you don’t really want to do in exchange for money? Will you endure some discomfort from cleaning my house or producing a pen for me or whatever in exchange for money?” That is basically how all market transactions work, or almost all of them work. So why doesn’t it happen when there is a political transaction going on? I think this is something that is the key point, that makes sense of all of the other things that we talk about — things like affordable housing mandates, things like environmental assessment, things like second staircase rules that require apartment buildings to have two sets of staircases rather than one. Their argument is that people who are extremely opposed to new development will trigger all sorts of veto points that drive up the fixed cost of building up to a point where there isn’t enough surplus left to buy off everybody else. So they might issue a legal challenge. They might say, “There are some rare bats that live here, and you need to fight us in court. You may win, but the cost and the hassle and the delay from doing that will drive up your costs.” And the more opportunities there are for them to do that, and the more extra death-by-a-thousand-cuts costs there are, the less and less money there is to go around to buy in all of the other people. So their argument is that the answer is to come up with much more streamlined decision making that has legitimacy, that allows the people who would be affected by the development, or would be most affected by the development, to have their say. Because if you don’t give them their say, A, that’s not good — that’s wrong, and you’re potentially imposing lots of costs on people which you shouldn’t — but B, has the legitimacy to overrule all of these other triggers for objection and for blockages. And I think this is such a good model, because when we look at stuff like HS2 or the Lower Thames Crossing or nuclear power plants, what we see is that it’s rarely a single thing causing the problem. It’s much more often this addition of legal reviews, judicial reviews of you having to do this impact assessment. In the case of HS2, just today I was reading about a £100 million tunnel had to be built to protect the bats from the train. That’s a very high value to put on bats. I quite like bats, but… Rob Wiblin: It’s a larger value on animals than we typically apply elsewhere, I would say. Sam Bowman: It is. And their model would be that all of these things both exist because of the demand to do this, and then once they are in place, they are used to drive up costs so that you can’t do this transaction. I think that’s so powerful, because it helps to understand what we might be able to do, and actually real-world examples where we have been able to do this, to cut through that, and allow this exchange or bargaining to take place that happens in almost every other area of our lives. Rob Wiblin: Yeah. So the basic structure here is that we’re talking about projects that have enormous net benefits to society. The benefits are large enough that the people who are benefiting from them would be able to compensate the losers, in principle, if they were able to have a theoretical frictionless negotiation between them — such that everyone involved or everyone who’s seriously affected would feel that they were better off, and it would get democratic consent to go ahead. So the fundamental question is why isn’t that negotiation happening, such that projects that produce enormous per-capita gains among all the affected people can go ahead, because everyone is basically getting bought in one way or another? I think before we get to discussing that, and the different approaches that you might have for getting people to be bought in, it might be worth stopping for a second and asking about the fundamental reason why a lot of these projects aren’t going ahead is because we live in a democracy. Most of these countries we’re talking about are democracies. And many or most people, certainly a vocal group of people, are against them. They think that they’ll be harmed. They speak out against them, they sign petitions, they go to their local council. They take advantage of the legal options that they have for trying to slow these things down, make them very expensive, get them blocked. Why do these folks think that they’re going to be harmed by these projects? By construction of housing, by construction of other infrastructure? Sam Bowman: Often they will be. It would be great if new infrastructure and new housing was only a net benefit. But lots of good things, like new infrastructure and new housing, have some negative externalities: they impose some costs on people around them. I live on a pretty busy road. I own a house. The road is quite loud. The more road users there are around me, the louder that road will get. Now, that doesn’t mean you shouldn’t add more people. I think you should. I think you should massively upzone the road that I live on in southeast London. But I will definitely bear a cost from that. And as it stands, I have one option. The only thing I can do is say, “I don’t want you to build this, because it will make my life slightly worse in a very small way.” Luckily we’ve got double glazing and things like that — so don’t worry, I’m not too badly off. But almost all the time we are saying that there is a kind of dispersed benefit, where some combination of society as a whole, and the direct users of the new thing — the people maybe who live in the house — will benefit. And a concentrated cost, where local people just get the option of either doing nothing and having their lives made slightly worse, or objecting and maybe avoiding that slight worsening of their lives. Now, it’s not always the case that these things are net bad. I think often new housing can be a net positive for an area. It brings more amenities, more shops. If there are enough people, it might improve the quality of the local bus route. If you do public services properly, it might mean you get better public services because denser areas can sustain more things. Rob Wiblin: There can be better jobs; they can be more culturally interesting. There’s a reason people like cities. Sam Bowman: Absolutely. There’s a reason people like cities. Totally. But also, people like having a garden, people like quiet, people like safety, people like to be able to know that their school is going to be kind of the way it was before. And if they’re comfortable with the area because they’ve moved there, presumably they like it to some extent. They often have a lot that they can lose from it changing. And frankly, urban areas are generally higher in crime than lower-density suburban areas. That’s a very, I think, important and significant reason that people tend toward — especially in the US, where crime can be really quite random and quite dangerous. I think things like that are very important in understanding why people object to new housing. I think it would actually be kind of irrational and hard to square with people’s stated behaviour if this was just about property values. You know, there’s a very simple model, a very basic model that says that people own their houses, more supply of housing drives down the price of houses, and so people just don’t want it. Number one: I think people don’t really act in that way. And number two: I don’t think it would be very rational for them to act in that way. In the same way that I think there would be a kind of overall aggregate productivity benefit to the UK if we built more houses, but it isn’t really rational for any individual to campaign for any individual housing — Rob Wiblin: It’s a collective action problem. Sam Bowman: Yeah. And that works both ways, right? The collective action problem also means that there isn’t really an incentive to stop a particular housing development on the basis of national housing policies. And also, I think it’s not a trivial thing that almost all political parties in every Western country are at least nominally in favour of building more houses. The Labour government in the UK has just been elected on a landslide. One of its major flagship policies has been build more houses. As I’ve mentioned, both of the presidential candidates have at least said some things in favour of building more houses. The Canadian Conservative Party may soon be elected on a platform of building more houses. It’s really hard to square that with the idea that people are just trying to stop houses from being built because they don’t want a national reduction in house prices. I think much more likely — and much more rational and easy to square with people’s actual behaviour — is that they’re protecting their neighbourhood. They like the quality of life that they get from their neighbourhood. They don’t want it to get worse, they don’t want it to change. And if that’s right, then actually pro-housing advocates may often be making the wrong arguments. I think there’s a slight tendency to harangue NIMBYs — as I call them, as everybody calls them. And maybe that’s kind of politically incorrect, but I think it’s also good to be clear and use words that people understand. So I will continue to call them NIMBYs, but I don’t think they’re bad people for the most part. I think that when you frame new housing as a kind of fairness issue, and say, “It’s very unfair: you own your house; it’s very unfair that you’re not letting other people into your neighbourhood.” That doesn’t sound to me like a very compelling reason that I should want to change my mind. That sounds like you’re saying I’m a bad person and that I am going to be made worse off by this, and I don’t want to be made worse off by this. So I think the challenge is how do we make new housing in the interests of existing residents? Either financially and/or I would say in terms of the effect that it has on the area. I think if we can do that — combined with simplifying the processes for actually approving the houses, and reducing the costs of approving the houses — then we might be able to unlock a huge amount of value. The reason that I think we should be excited about this is that there is so much money on the table. We are talking about trillions of pounds in the UK, tens of trillions of dollars in the US, and trillions of dollars almost in almost every other country. New Zealand and Ireland it’s more like hundreds of billions, or tens of billions maybe, but it’s still a lot. There is so much on the table that, if we can fix the mechanisms, then that could be it. I do get a little frustrated when we talk in terms of, How many houses does the UK need? How many houses should we build somewhere? Because that’s the wrong question. It’s an interesting question, and it’s interesting to imagine how many we would be building if we didn’t have these obstructions to building versus how many we are. But it isn’t really fixing the problem. A lot of the approaches that governments take are driven by targets, where they say, “This area must take this number of houses, and this area must take this number.” Really, we don’t want to have to use targets. We really want a world where the price system is doing the targets and people are, “Great! They’re building an apartment complex down the road from us — we’re rich!” Or, “This area is going to become so great.” That’s what we want to get to, and I think we can. Rob Wiblin: I used to buy this theory that it was people trying to increase their home prices basically that was driving them to want to oppose construction. But there’s several giveaways that that’s not the case. To begin with, renters tend to be equally as opposed to new construction as homeowners, despite the fact that they benefit without losing in any way. Also, if people were just so focused on trying to extract as much money as they can from their property, why wouldn’t they be campaigning to upzone their specific property so that they can then sell it, like flip it for twice the value and then go live somewhere else? But we see relatively few attempts to do that. There’s a great book, Neighborhood Defenders. I think this poor anthropologist or ethnographer went through enormous numbers of transcripts, and I think probably interviewed as well, many people who were trying to stop development in their area to find out what was primarily motivating them. And she came away thinking that it was mostly that they wanted to maintain the character of their neighbourhood, was basically the thing that they were saying. I think most attempts to overcome this NIMBY issue have basically tried to go in the front door: you get a central government that says, “Look, I understand that you don’t want to have this thing in your area, but it’s necessary for the country. It’s terrible that we’re not constructing housing. Look at all the problems, this litany of issues that’s creating there. So we’re just going to basically force you to do it with targets, or we’re going to take away your right to object.” This typically has produced poor results. Why is that? Sam Bowman: Probably for two reasons. Targets fail in the same way that targets in the Soviet Union failed when you would say to a shoe factory, “Produce 38,000 shoes,” and they’d make 38,000 left baby shoes. And every time you try to make it more precise, people just get some way to evade or to minimise the burden of the system. So local authorities, when they have targets, will tend to assign housing away from existing populations — far from where the amenities are, where the shops are, where the roads are, things like that. So you get fairly low-quality housing when you use housing targets. That’s one problem. Much more important, I would say, is that because you haven’t fixed the underlying incentives problem, a government that does try to do a radical upzoning will face very significant political objections — because they haven’t actually made more housing in the interests of the people who live near the new housing, so those people will fight it. And even if a government does bravely manage to get something over the line and spend tonnes of political capital, it can be repealed. We’ve just seen in New Zealand that the previous Labour government was voted out partly on the basis of a very bold upzoning. I’m actually quite impressed, but the new government has repealed in some senses the old approach, but has also brought in very pro-housing measures — partly because they have an incredibly good housing minister, the new New Zealand government, but that’s very unusual. In Croydon, in south London, the local government passed a rule basically allowing individual homeowners to densify their plots. It was repealed a few years later after that government was voted out and the opposition Conservatives were voted in. Many, many of these things are either impossible to get in because they don’t actually change the incentives, or once they get in, they’re very non-durable — they’re very easy to reverse. So the result is that most governments just won’t try — very rationally, I think very reasonably. I think the only way to do this durably is to buy in residents to new housing, so that residents want new housing to be built. And we can talk about some examples of that; there’s one really interesting example from Israel. Doing that, I think, is the key to unlocking this. And there are ideas in process, somewhere between ideas and reality, in the UK, in the US — but I think the key is changing incentives, so that people want housing to be built near them, and then lots of the other problems go away. Rob Wiblin: So we’re going to talk here about basically a bunch of different approaches that you could use to get a majority, maybe a great majority of people who are affected by some change, to want it to go ahead — despite the fact that if they weren’t compensated or that they’re at some risk of being losers, despite the large net benefits to the change. We’re mostly going to be thinking about construction and about housing in particular, but it’s worth noting that, at an abstract level, this applies to all problems. It’s not only about spatial issues or about construction and land use. There are many ways in which we could potentially reform society, change laws, change tax policy, change some enterprise that is very inefficient, but perhaps some of the employees there are benefiting, that kind of thing — where society would benefit as a whole, but there are some powerful people involved in the decision or who could try to block the change, who effectively have a veto. And if the only option that you give them is to veto or not veto, they will stop the change, and society will miss out on the enormous gains that could be harvested there. I guess I’m familiar with this, having done economics: I think you used the term “Coasean bargaining.” So Ronald Coase was an economist who looked at this issue, saying that if everyone has very clear property rights about what things you need to get, exactly who you need to get permission from in order to do various different things, then they should all just be able to negotiate, bargain away, and produce the GDP-maximising outcome, no matter how the initial property rights are allocated. Of course there’s frictions. What’s the saying that economists use now? “The answer is transaction costs.” So of course this doesn’t in fact happen, because a transaction cost can be quite considerable, and the issue of transaction costs and the difficulty of negotiating and reaching an agreement can mean that it does matter who has the initial property rights and exactly what the legal setup is. But at least in a highly abstracted sense, there is always this question: If it’s clear whose permission you need, why can’t you just compensate them, and then get the efficient outcome? Do you want to say anything about that? Sam Bowman: Yeah. Should we talk about street votes? Rob Wiblin: Yeah, let’s do it. Let’s do street votes. What are street votes? Sam Bowman: Street votes are a policy that allow the residents of a street to have a vote on the design and density rules of that street. And in doing so, they can opt out of the existing planning or zoning rules, and opt into a system that they control. To give you an example: If you live on what we would call a semi-detached-house street in the UK, somewhere in outer London, at the moment you basically will find it impossible to build anything more than what is currently on your land — a semi-detached house. Under the street votes policy, you may petition your neighbours to trigger a vote. You need to get, say, 10% of the people on the street to agree that we should have a vote. Then a series of options can be submitted. Anybody on the street can submit a design code — sort of rules about what the house has to look like. And part of the design code will be a density rule. So can we build up to three storeys or four storeys? Most proposals for street votes involve some kind of hard cap, kind of an arbitrary limit on how high you can build, so that you’re not getting streets building skyscrapers that then overshadow all the neighbouring streets. You want to be able to make sure that the main effect of the decision is borne by the people on the street. And then they choose. They say, “Yep, these are the ones I’m happy with.” And if a design code passes a threshold — gets, say, a supermajority of 66%+ of the people living on that street — then, in addition to the current rules about what they can build, they also have the option of building the housing that has been voted for them and approved by them through this vote. Rob Wiblin: So why, in theory, would we expect that this solves the problem? Sam Bowman: Because the decision is so localised that the people making the decision are also the people who benefit from the surplus, who get to capture the surplus that you get when you get permission to build. So if you’re in a semi-detached house somewhere in London, let’s say it costs £600,000, £700,000, depending on where you live. If you get permission to build up to say six storeys, the value of the house that you live in and the land that it’s on could rise up to, say, £2 million, because you can build so much more value on that land. Now, most of the residents probably won’t choose to personally redevelop. They’ll probably sell their land to a developer who will then do infill — which is when you build across the gaps in between houses. Semi-detached houses usually have a little alley, they usually have a parking garage — there’s a lot of not-very-well-used space in between the bits that people actually live on. A developer that buys, say, five houses in a row can build under this proposal up to let’s say six storeys. And they have to do so according to a design code. As long as the residents have approved that design code, then you can build this. You don’t face any of the other objections to building. So there’s a win-win: we have turned people’s veto into something they can give away in exchange for the value of their property increasing. So you have a built-in incentive to say yes; there’s a built-in reason, because you are the one that captures the value of the uplift. I think this is really, really powerful. And when we look around the world, there are similar — not identical, but similar — proposals or actually policies that have worked in this way. But the key mechanism, and you’ve talked about this already, is to take away a situation where all you can do is veto something, and to be able to transact that veto — to be able to not quite sell the veto, but have the ability to give it up in exchange for some of the benefit that would otherwise go to the developer, or the people who get to live in the apartments, or whatever it might be. Rob Wiblin: So what’s the experience been in cases where people have tried something along these lines? Sam Bowman: So in Israel, there is a currently existing policy called TAMA 38. There’s another policy called Pinui-Binui, but let’s focus on TAMA 38. What this does is allow the residents of an apartment block to vote to either reinforce the block and add more storeys, adding more units, or demolish the block and then just replace it with a taller one, adding more units. What they get from that is new apartment units that then are sold to fund the redevelopment, and they personally get larger and hopefully nicer and more modern apartments for themselves. This effectively allows the existing residents of these apartment blocks to get the benefit of adding new units to the apartment block. And this now accounts for a third of new housing being built in Israel every year. This is a huge, huge policy. Just recently I was talking to somebody who has an Israeli husband, who mentioned, “Yeah, my husband’s family did that a few years ago.” So this is really important. There are other examples. In Seoul there was a policy called the Joint Redevelopment Policy, which allowed very low-density, effectively villages on the outskirts of what was then Seoul in the 1980s, to vote for redevelopment. I would say they actually made some pretty big mistakes, because they didn’t have height limits — so some of the most high-rise, dense parts of Seoul were delivered through this policy. It was kind of street votes on crack. I think that’s bad because at that point, the street is no longer the kind of legitimate deciding body. Once you get over a certain height, the people around you are affected by that, not just the people on the street. There’s a very limited but quite fun example of the Squamish Nation outside of Vancouver, the First Nations people that owned the land that now is Vancouver. They still own some of the land, and a decision by a court ruled that they have the rights to develop their land, and they’re now undertaking a huge redevelopment of their land. And effectively, having voted for massive redevelopment, they will now collect a lot of the benefits of that. I think that’s pretty great. It’s a great way of allowing people who are currently on land and currently own land to share in the benefits of density and development that would otherwise usually be captured by the developer or by the local authority or whatever it might be. There are other examples that are not directly analogous to this, but are somewhat analogous. So in New York City, Mayor de Blasio passed a rule that allowed air rights to be sold by historic buildings that were not going to ever build up. And you ended up with a situation where Katz’s Deli was lobbying people to vote in favour of this ordinance, and the Catholic Bishop of New York had his parishioners handing out leaflets in favour of this, because churches — Rob Wiblin: The churches are not going to develop up. So it’s a windfall for them. Sam Bowman: So they were able to sell their air rights to other developments. And some of the super tall towers that are being built have been enabled by this transaction of air rights: 450,000 square feet of housing has been enabled by this rule. And what these policies have in common is buying in people who kind of have a de facto right — but the right is at the moment not transactable — and allowing them to transact it. You mentioned Ronald Coase, one of my favourite economists. I call it Coasean democracy because the democratic element of this is so key. We could have a system where every homeowner effectively had complete control over what happened over their plot. That would get you a lot of building. That would be effectively a zero-constraint system, but it would also get you a lot of negative externalities. It will get you a lot of stuff getting built that was very unpopular, that maybe was very ugly. There are reasons that we have controls, and they’re not just, “Oh, it’s very unfortunate that we’ve ended up with these rules about what you can build” — it’s good that we have rules about what you can build! You are affected by what is built in your area. Some of the nicest parts of London were built by single owners who capture all of the externalities and so build in a single, consistent manner. Regent Street is an example. Lots of Bloomsbury. Lots of the very lovely, dense old parts of London were built by a single person because they internalise the externalities, and so they have an incentive both to build beautifully and to not build things that obviously make the internal value of what they have worse. So what this attempts to do is to approximate having a single owner. We obviously are not going to, and people want to own the land that they live on primarily. But using democracy, if we do it at a really localised level like this, we can approximate having a single owner. It’s almost like we’re turning a street in this respect into a kind of joint stock company for the purposes of making these decisions. And I think it’s a really elegant way of allowing bargaining to take place via democratic decision-making processes with lots of protections for minorities — “minorities” as in people who lose the vote. So doing it via supermajority; letting renters have a say, not just the owners of the property; if you want, depending on where you are, saying people who’ve been on the street for more than 10 years, maybe their vote counts double; or maybe you need to get a majority of people who’ve been there for 10 years rather than just people who live there right now. There are different kinds of contextual rules that you might want, depending on where you are. But the core point is if the decision happens at a local-enough level, then the incentive will be really strong: the surplus available will be there for people to be delighted to vote for more density. And as long as you do it with things like design codes — so that they can have some confidence that what gets built will enhance the quality of the area, and won’t just be horrible glass monstrosities that make their road feel really bad — then you can make it better for the people who choose to stay there. And you can create a huge, huge financial incentive for people to vote for upzoning that just would not exist via most other systems. Rob Wiblin: The thing that surprises me about this working is that I guess currently most of these decisions are made at the suburb level. And maybe that’s kind of the worst jurisdiction to make this decision — because it includes a whole lot of people who don’t really benefit at all from a development going ahead, but includes a lot of people who might be harmed very slightly because congestion has slightly increased in their area. But if you have a single street that decides to upzone itself, wouldn’t all of the surrounding streets be really annoyed by this, and basically campaign to prevent that one street from upzoning itself? Because they don’t personally benefit financially from an adjoining street having apartments or tall buildings, but they would imagine that they would be worse off because of the congestion and public services and so on? Sam Bowman: I think that is a risk if you don’t design the policy in the right way. I think that you are correct that the residents of a street where a new development takes place are not the only people who are affected by the new development. But there are mechanisms that we should introduce that can make it such that the street is by far the most affected negatively. One thing you can do is make sure that some portion of the uplift from the collective uplift goes to the local authority, or goes to public services, so that you’re immediately getting the money that you need to make sure there isn’t this extra demand for public services. Another is controls on car ownership. And this will depend on where you are. In parts of London, that will be very straightforward and make a lot of sense. Same with lots of parts of New York. In San Jose that probably is less practical. But you could, for example, require that off-street parking be part of the deal. There are things that you can do to at least reduce the parking burden. And part of the deal of local taxation is also that you are funding the maintenance and expansion of roads. So a lot of the stuff that one would worry about can be built into the policy, so that you are either cushioning or preventing outright this extra negative externality that goes beyond the street. In terms of the size of the development, this is why height limits are important. People being able to street vote themselves up to 30-storey buildings would be a terrible idea. You would get people from at least a kilometre around getting furious about this new tower that was being built. When it’s six or seven storeys, the impact is much less. In some places, you should be thinking more in terms of a block. The street is not the only unit that can do a vote; a block could also do a vote. (A block being a contained group of houses that sort of back into each other; I don’t just mean a block in the US sense of like a unit of distance or something like that.) That’s another unit that could do a vote. And block votes are actually very high potential if you have places with underutilised car parks, for example, or the kind of urban wasteland that a lot of cities have between houses. The flat I lived in before the one I do now had a car park right on the edge of Brixton that basically was unused. This apartment building had like 50 apartments in it, so hundreds of people living in it, and maybe like three cars were there at any given time. It was really kind of shocking how much of a waste it was. Now, if you could build on that, you could easily double the footprint. You could easily double the number of homes that unit occupied. And I didn’t need a car; clearly most of the other residents there did not need a car. So I think that there are mitigations you can put in for many of the costs. I think in general, the question is the right one: where are the costs most acutely felt? Partly from just a pure economic justice and economic efficiency point of view, you want the decision to be made by people who bear the biggest costs. But partly also from a political pragmatism point of view: people will fight a policy if they stand to lose out from it a lot. But we can see from Israel, for example, actually what happens is people lobby for more dense rules around what they can upzone into. In Israel, there was a proposal to redistrict the height rules and things like that in an area of Tel Aviv. I think it was Tel Aviv, it may have been Haifa. And the residents were furious that it wasn’t going higher, because they all wanted to use this policy to upzone themselves. So if you do things right, you will have countervailing demands for everybody to be able to upzone themselves. Rob Wiblin: I think maybe part of the reason why this isn’t such a devastating problem is that, sure, it might be the case that some of the residents in the neighbouring streets don’t love this, but they’re only affected moderately. They’re affected a little bit, maybe a medium amount. The people on the street expect to make maybe a million pounds if we’re talking about London here. So the people who are a little bit harmed, who might go to the council and complain a bit, they now have to confront this incredibly powerful, incredibly motivated, incredibly coordinated interest group that desperately wants to have the street vote thing go through. And it’s easy to see how they might basically lose out in that fight, because they just aren’t sufficiently motivated to prevent the enormous amount of value being created. Sam Bowman: I agree with that. I think it’s also important to recognise that a lot of people’s objections to developments that are happening not immediately near them, but further away: they’re not directly affected by that development, but they have a rational interest in not conceding a principle that that kind of development can happen. So when people object to a tower in Brixton, they’re often not personally affected by the tower in Brixton at all. What they’re objecting to is a slippery slope where you can build that kind of tower anywhere, including near them. And one of the good things about street votes is that hopefully it gives people confidence that even if this street over here decides that it’s going up to six storeys, the people on your street will still have control over your densities, and you can opt not to do that. Unless a supermajority of people on your street also want this, then you can opt out of it. I think it’s that kind of opt out that gives people some confidence, or will give people some confidence that this isn’t going to be a blanket thing that affects them. Rob Wiblin: So my understanding is that we’re in the process of maybe actually getting street votes in the UK. Do we have a sense of how many different streets or how much enthusiasm there will be for this actual upzoning? Is there any chance that it actually wouldn’t get approved in many places? Sam Bowman: Yeah, the primary legislation, the actual bill that needed to be passed by Parliament, has been passed. The King signed it into law. The law for street votes exists. What now needs to happen is what’s called the secondary legislation — in America this would be called the rulemaking — where the details, the nitty-gritty of what this actually means in practice has to be delivered. We do actually have something that currently exists called estate regeneration, which I think points to the very high takeup that we may see from street votes. Estate regeneration applies to public housing, council housing estates in London that maybe were built in the 1950s or ’60s quite cheaply. They are usually quite cold, quite small, draughty, sometimes unsafe — and really they need to be overhauled. They need to either be significantly improved or replaced. Now, there have been numerous proposals by private developers to take the estates, to build new apartments — both for the existing residents, and also private apartments to sell and fund the whole redevelopment. This has been done, and was being done throughout the 2010s because huge amounts of central London housing are social housing — up to 40%, 45% in places like Southwark, Lambeth. Under [London Mayor] Sadiq Khan, a rule was brought in to say that the residents of these estates should have the final say on whether these happen or not. So not just the housing associations that notionally represent them, but the residents themselves should have a vote. And somewhat unexpectedly, these votes have passed with almost North Korean levels of support. There was a case in a place called Aberfeldy Village, which is not that far from Canary Wharf, where a proposal would have taken 330 homes and added about 1,200 more, giving all of the existing residents bigger, nicer, better apartments — and paid for it all with these 1,200 extra homes. A few of them would have actually been socially rented, but most privately rented. With 91% turnout of the residents, 93% voted in favour of this. And this has actually happened 30 times across London over the last six or seven years. What this points to is, number one, that this policy by itself is a really powerful way of unlocking really central land for new housing and giving better housing to the existing council of public housing tenants. But number two, it suggests that in the existing mechanism we’ve got for letting residents vote themselves into higher density and higher quality homes, there’s a huge, huge takeup. And remember that these voters do not get any kind of financial benefit from this; they don’t stand to benefit in any way other than better apartments. Occasionally there’s a slight inconvenience payment, but it’s not like they’re getting a fraction of a million pounds or anything like that — they’re really just being compensated for the hassle of having to live offsite while the new home is built. But there is still this huge, huge takeup of it. I do think that, by the way, we should sharpen that policy. I think it would be great if residents themselves could initiate those kinds of votes, and in principle kind of declare to developers, “Come to us! We want this to happen.” They can’t do that right now. And also I think that there should be a benefit of the doubt given to residents. If they want something to go ahead, the local council can’t obstruct it on grounds that you’re turning a road into a cycle path or something like that — which has actually happened on a few occasions. But I think it’s much more important to look at this policy and see both how much more we could be doing with projects like this, and how much existing takeup there is. So we don’t just have to look to Israel today, we don’t just have to look to South Korea in the 1980s and ’90s: we can look at London right now and see that there is a gigantic demand for better quality housing delivered via these kinds of higher supply mechanisms. Rob Wiblin: Is there any appetite for doing something like street votes in California or New York or other countries? Sam Bowman: I think there might be. California is a really interesting case. California YIMBY I think is one of the best pro-housing organisations on Earth. They’ve been amazingly impressive in some of the things that they’ve done. They have taken what is arguably the most NIMBY place in the developed world, California, and they have actually been passing bills in the Californian state legislature. It’s really impressive. Some of them have been really, really effective. One of the things that they have been doing is allowing what are called “accessory dwelling units” — which are like mini houses that you can build in your back garden. And the number of people and the number of accessory dwelling units that have been built since the laws that California YIMBY has helped to bring into existence has really skyrocketed. You’ve gone from like a few thousand a year to tens of thousands a year. It may enter the hundreds of thousands a year at some point. However, they have still fallen into certain roadblocks they have still faced. I mean, California really, really is a NIMBY place. There’s a California law professor called Chris Elmendorf who did a really good analysis of why, even though these really pro-housing bills are being passed, why is it that you still aren’t actually seeing a really big uptick in the delivery of housing? He argues that a lot of the times to get the bill across the line, they’ve had to assemble a coalition of interests that have added all sorts of extra costs to development — most importantly, affordable housing mandates that say X% of any new development has to be subsidised and given to people at a much-below-market rate. And he argues that adding in these kinds of provisions has meant that, even though in principle it’s easier to build, the actual cost of building has now become so high that it isn’t viable to do that. Other things are like you have to use union labour — you know, other kinds of things that sound really good on the face of it, but then you realise that what you’re really doing is imposing a direct tax on new housing, the last thing that we want to tax. In the same way that we talked earlier about the NIMBY principle — by adding in all sorts of triggers for challenging these things, you just create other avenues for stopping things from being built — the underlying problem is that you are still trying to impose housing on places that don’t really have a reason to want it. I think it’s great, by the way. I think what they’ve done is spectacular and incredibly important. But my hope is that policies like street votes, that try to kind of buy in local residents to new housing, might be a way of circumventing the death-by-a-thousand-cuts problems that some of the Californian bills have faced. And I think that you could see that in many other places as well. Look at Brooklyn in New York state. I saw somebody describe the Brooklyn Heights area, which is these sort of beautiful brownstone homes. It basically has the best suburb in the world. I think there is no suburb that is more beautiful, that does density in a more livable way than Brooklyn Heights. It’s amazing. There is no reason that all of Brooklyn shouldn’t look like that. And you can go further out — and this becomes less and less viable the further away you are from the Manhattan core — but there are so many parts of Brooklyn that do not look like that, that should look like that, and that could look like that — reasonably affordably, if there was a mechanism that allowed those residents to say, “Yes, we want this kind of design, and yes, we are willing to take more density.” It’s not even really the tradeoff, because they benefit from the density as well, but it’s kind of part of the package. You know, Brooklyn Heights is a pretty dense area. “Suburb” is slightly misstating it, if people actually go there and experience what it’s like. But it’s very livable. I think that there are probably many, many places in almost every city in the Western world, especially in the English-speaking world, and especially in American cities, that probably are underdensified. I am by no means anti-car. I think cars are really great in lots of ways, and especially cars that don’t emit CO2 are absolutely wonderful things. I’m very pro-car. But if we look at what people demand: people want to live centrally, they want density. The most high-value homes that you can build are high-density homes in cities. So A, I think allowing those to be built via this kind of mechanism will get you a lot more livable stuff, and B, it will give this really strong incentive to existing residents who right now don’t want that to happen. And change not everybody: the key thing here is that not everybody has to do this. Places that really like the way they are right now, great. You will have a direct way of not opting into this higher-density system. But places that do think, “Yeah, we want a share of this uplift, and yeah, we want to beautify our areas”: give them a path to doing it, and do it in a really simple way that doesn’t come with all these other costs and riders. Rob Wiblin: Let’s push on and talk about the second idea which I think is kind of a generalisation of street votes, or a generalisation of the idea of compensating people who perceive themselves to be losers. I guess there’s different names that you could give. We talked about compensating veto points or buying out the losers. I think maybe “benefit sharing” is the nicest term for this idea. I suppose you’ve alluded to this idea quite a lot through the interview so far, but can you elaborate on what the mechanisms might be, and how this would work in practice? Sam Bowman: Yeah. So I think that there’s a widespread problem that we have of political incumbency. In markets, if the market is working well, incumbents should always be looking over their shoulders — because new entrants can come in, compete their economic rents away, displace them if they’re more efficient than them or if they have a better idea than them. And that works pretty well. The problem that we often have in politics is that we don’t have a really good mechanism for replacing incumbents or changing a system that favours political incumbents with one that works more effectively. I think there are a bunch of ways to think about how to change that. One is to share some of the benefits of the improvement with the incumbents. So there are three examples from British history, but they’re globally relevant. The first is the NHS, the National Health Service in the UK, which has a lot of problems. I think anybody who’s experienced it in the last few years will agree it’s far from perfect, but it was a very, very impressive piece of political craftsmanship to introduce it. What it was able to do was cut out incumbents who would stand to lose from its introduction. In the UK case, it was GPs, general practitioner doctors, who basically would lose their practices, and, were they nationalised with the rest of the service, would basically become kind of civil servants: paid not very much, maybe wouldn’t be able to control their quality of life or the way they worked. And the NHS architect, Bevan, the guy who came up with this, literally used the term, “I will stuff their mouths with gold.” So to this day, we have a system where, although many people believe that the NHS is a completely public system, their general practitioners, probably the person from the NHS they have the most direct experience with, is a privately run institution that is contracting with the NHS. And it’s very funny: you get GPs talking about, “They’re going to privatise the NHS” and so on and so forth. They are private. They are operating on a private model; they are running their own surgeries as private businesses. And really what that was was allowing them to opt out of the NHS system. Then, on the completely other side of the political spectrum, you have, under Thatcher, two really important reforms. One was right to buy, which was that there was a lot of public housing that was not particularly well allocated: lots of people living centrally, but maybe not being very economically active, or having relatively low-wage jobs. And there was a view that people in public housing are kind of inherently dependent on the state in a way that they didn’t want. Now, one option would be to say, “We’re going to kick you out. We’re going to privatise all this public housing, we’re going to sell it, and we’re going to make loads of money for ourselves. Tough luck.” That I think would never have been a goer, and I think even Margaret Thatcher would have struggled to get that law passed, and remain in government had it been passed. Because this is people’s lives you’re talking about, right? You cannot evict people from homes they’ve lived in for years, even if they don’t own the home. They do have a claim to the home, whether it’s a public home or a private home. The right to buy approach said, “You, who have been living in this home for decades, you may buy the home at a very significant discount” — in some cases, 80% off the market price — “and you will then own it; it will be your property.” And huge numbers of homes were sold in this way. It was only really after Labour got in and reduced the discount that this kind of stopped being a feature. It still exists in law, but it’s a much less important factor. But what that did was say, “We’re going to share some of the benefits of this reform with people.” Many of those people still live in those homes, but many more of them have sold those homes at very high profits to themselves and have moved to, maybe if they’re older, quieter places, made some money. This has been a very large transfer, both out of the hands of the public sector and also to the people who were the recipients of this kind of de facto subsidy. The final example is the privatisation of public utilities in the UK: things like British Gas, the Electricity Supply Board, things like what became British Telecom and Vodafone. The employees of those companies were given the chance to buy really cheap shares in those companies, at like a 90% discount. Effectively, they were given shares in the companies, because they would be the ones who would stand to lose out most from those sectors becoming more efficient. The problem with monopoly is that it gives you a quiet life. The big problem with monopoly is that it gives the companies who are monopolists, whether they’re public monopolists or private monopolists, a quiet life. And if you’re an employee of those companies, that’s often great. That’s a very nice life to have, so you have a strong incentive to fight something that would change it, and either impose a more tough working day on you, or maybe involve layoffs and maybe involve you losing your job. And part of the way privatisation could happen and avoid these people from fighting tooth and nail… Which, by the way, there was no such inducement for coal miners: they did not get this kind of bribe, and they fought it to spectacular degrees; it was almost civil war in some parts of the country. The privatisation programme worked much more smoothly and avoided that kind of conflict because it involved sharing some of the benefits. So I think that’s one approach: share some of the benefits of the reform that you want to make, and you might end up getting much less opposition than you would if you tried to go in the front door, so to speak. Rob Wiblin: I’m glad we’re talking about non-housing examples, because it helps to demonstrate just how general this principle is, and how widely applicable it could be across countries and policy areas. But thinking about how this benefit sharing could be useful in the construction case, I guess you could imagine a developer wants to build a big block of apartments somewhere. They’re printing money here. We’re imagining it’s London, so this is making millions, tens of millions, conceivably hundreds of millions of pounds from this large development. But the local people are against it because of all kinds of reasons, I guess. They think it’s going to make them worse off. So I suppose we would imagine that they would go into a negotiation with the people who live within some boundary around this large development, and they might say, “How are we going to get a majority of these people, maybe a supermajority, to be in favour of this development? How much gold do we have to stuff in their mouths exactly in order to make them not just tolerate this through gritted teeth, but actually be keen, be glad that this is going ahead?” You can imagine the people who are immediately adjoining, they compensate them 5% of the property value, basically. They could even compensate the renters as well as the landowners. And then the people who are maybe 10 to 100 metres away, they get 2%, and then you have some sort of decreasing compensation schedule — to the point where not everyone is going to be in favour of this development that has an enormous net benefit, but you could get quite a large fraction of people on board. And on net, the people pushing it will outweigh the people who are against it. And even the people who are harmed, we might not feel so bad for them, because they’re at least getting partially compensated. Is that the basic idea? Sam Bowman: Sort of. But I think it’s important not to do kind of ad-hoc collections of people. I think it’s important in advance, before you have a specific project, to define who gets to have a say, and to what extent they get to have a say. Otherwise you’ll get people declaring that they have a big problem with it to be bundled in, and then you get right back to the problem we’ve got right now. I think one way of thinking about the problem we’ve got right now is that there are different kinds of objectors to things. There are some people who are going to object to something under any circumstances. You could give them the most beautiful new development, they could become multimillionaires from it, and they will still object because — Rob Wiblin: They hate change. Sam Bowman: Yeah, they just hate change. And that’s fine. That’s fine. People are entitled to feel that way. Most people, if they are immediately around and affected by it, they don’t want it to happen because they’re going to be made worse off under the current regime. But they are reasonable, and they could negotiate, and you could buy them in. And I think it’s important to talk about: they should be able to share in the benefits. It’s really not about bribery. People don’t like the feeling of being bribed. There really isn’t a mechanism like that. And giving cash to people actually doesn’t work at all. People hate the feeling of being given cash. It makes them feel cheap. It makes them feel like mercenaries. You know, people care so much more about the nature of their area and the feeling of that joint uplift than they do the, “You’ve just cut me a big cheque.” And then the third — Rob Wiblin: I feel like I would just take the cheque. Sorry, go ahead. Sam Bowman: Well, you’re an economist. But people just do actually really dislike it. It’s interesting. Rob Wiblin: It feels dirty somehow. Sam Bowman: It feels dirty. People, for example, quite like being given cheap or discounted electricity from a new nuclear power plant or new power supply being built near them. That is a thing that works, and that is a thing that happens. The same cash value of just being given a cheque is much less popular. People like the feeling of symmetry and of it not just feeling purely transactional. But then there’s a third group of people, and I think these are the people who it’s really important to… The real challenge in the design of this type of project is cutting out the third group of people, who are cheap talkers. They’re people who will sign a petition or who will weakly express some kind of objection to something. But if they don’t have a forum to do that and they don’t have a route to doing that, they just won’t care, because it doesn’t affect them directly. It’s like the Instagram campaigning thing. Rob Wiblin: Slacktivism. Sam Bowman: Yeah, slacktivism. Like I’ll post a meme on my Instagram story about this and then never think about it again. And the real problem that current decision-making structures have is that it’s impossible to differentiate between those people and the people who have legitimate objections, who will fight tooth and nail. I think the goal of anything — like street votes, or estate regeneration, or we could imagine something like this for approving new nuclear power stations, or even new runways at airports — the really challenging thing, I think that becomes harder and harder the bigger the project is and the footprint of the project is, because it becomes harder and harder to differentiate between the people who are genuinely adversely affected and the people who say they are — either because it’s easy to, or because if they do, they might get some of this compensation. And that’s the policy design challenge. I think the street or the block is quite an elegant and intuitively satisfying way of doing it. And empirically, it seems to work the same way that the housing estate seems to work. But were we talking about doing a vote for a new runway at Heathrow Airport or something like that, it would be much more difficult to do. I don’t know that it’s impossible, but I certainly wouldn’t be able to say right here, “It’s this group of people.” Rob Wiblin: Yeah. I just want to pause for a second to point out how crazy it is that with our current permissioning system, basically people can only say no to things, and in general they can’t be compensated to say yes. Just imagine that there’s some joint project that needs both you and me to consent for it to happen. And I would benefit: it’s 100 units — enormous, life-changing benefit to me from this project going ahead. To you, it’s barely an inconvenience: you suffer like 1 or 0.1 units. But you don’t perceive yourself as winning; you perceive yourself as very slightly losing. Now, in the sane world, in the normal economy, of course, I would just compensate you enough such that you’re above zero. You perceive yourself as slightly benefiting, and then we both consent. Maybe there’ll be some issues with the negotiation, but we would expect something with such an enormous net benefit to proceed. But if literally you say it is impossible for me to compensate you in any way, it doesn’t matter how large the ratio of benefit to cost becomes. There could be 10 people benefiting enormously with life-changing gains. But you’re just going to say no, no matter what. I think that helps to explain just how large the costs have gotten from this structure in the economy. Year on year, the costs just get worse and worse. But we’re still just stuck here because all that people can do is veto, and they continue to do it. Sam Bowman: The extra thing to talk about here is I am very interested in direct democracy, in using quite local decision-making, because I think it really aligns the incentives well: the people who are most harmed by it also get to be the people who make the decision, also get to be the people who get some of the benefit from it. But we have local government. And local government should be doing this fairly well, but it isn’t. And one of the reasons is that local government used to both feel the costs and the benefits of new developments in its area, and now it experiences neither. The big change that happened in the UK in the 1940s, the Town and Country Planning Act, with respect to in-city densifications — not the green belts around the cities, but with the densification within the cities — it wasn’t that they gained the power to stop things. They had those powers for much longer; they had those powers for decades. It was that before 1947, they would have to compensate you if they stopped you from building on your property. After 1947, they had no compensation requirements, so they lost any reason not to say no. And then on the other side, it used to be that property taxes were retained locally. So it used to be if a prison was built in your area, or a new factory or something like that, you would pay what are called business rates — just business property taxes — and it would be the local council that would retain that money. So you had competition between local areas to allow these things to be built in their areas then, in the postwar era in the UK. But this exact process happened in the United States — especially in California, especially New York, in the kind of more liberal states — driven by a kind of sense of equity. Areas that were collecting large amounts of property taxation began to be forced to put some of that property taxation centrally for it to be redistributed to poorer areas. You can completely understand why: the fairness element of it feels very intuitive. But the effect was that in both the UK and in parts of the United States, the incentive to allow more housing and to allow more business premises was eroded. You know, it even used to be the case that in the US they would sometimes have higher rates of property taxes for new houses. They would call this, I can’t remember, it was something like the welcome tax. It’s sort of ironically named, you know, we’re going to tax you more to build it. Rob Wiblin: Well, that actually makes sense conceptually, right? Because the existing taxpayers have paid for all of the assets, the accumulated infrastructure that belongs to the local council. Imagine someone who’s going to come in and start benefiting from that. Maybe they should have to pay a bit extra to compensate for all the past investments. Sam Bowman: Absolutely. There’s a friend of mine named Judge Glock — best name in the world. Judge Glock writes about this, and argues that restrictionary zoning is specifically driven by a free-rider problem that emerges when you don’t have this level of property taxation. When you can’t price new entrants at a higher rate than existing entrants, then you have a situation where people have an incentive to move to areas with good public services — like good schools, whatever it might be — who are not paying their fair share, or who are not compensating the existing residents. So it might be that apartment buildings pay less overall in property taxation. So you get areas that have good schools, or maybe they’re safe or whatever, and the apartment building is built. You can move there, you can pay less in property taxation, and kind of free ride on the existing services and amenities. And the only thing that the local government can do is use zoning to keep people out and to stop this free riding from going on. I think that’s a really interesting theory. I don’t think that explains all of the problem by any means, but I think it’s a very interesting part of the story. And in terms of the timeline, it coincides really well with this. It was roughly the mid-1970s that this began to really take force, this kind of movement in law. Now, I’ve written about it in the UK, and I think this applies everywhere. In the UK, because there is so little local retention of property taxes paid by businesses — and paid by houses, but let’s focus on businesses — it means that new prisons, new nuclear power stations, new data centres just have no real benefit to an area compared to what they could if they paid 100% to the local council and then there was more than 100% compensation — say, a top-up from central government. So we have some infrastructure, where prisons are a great example. Nobody really wants to live near a prison. Maybe it’s irrational. Most prisons don’t have escapes. Rob Wiblin: I think it’s understandable. Sam Bowman: It’s understandable. It used to be the case that prisons would pay so much in property taxation that you would get some benefit: your school would be better or your infrastructure would be better or your parks would be better. Now that is not the case. Now there is really no uplift, or there is very muted uplift, to your local taxation, to the wealth of your local council. What I’ve suggested is to make sure that the local council can keep all of the property tax bill, top it up from central government, and then hopefully you would get a situation where there was a large enough incentive for local authorities to want these things to be built in their areas. And I think this is going to be so important as data centres become more and more… I mean, we already need lots of data centres to be built. The problem with data centres is they don’t employ that many people. They tend to be quite ugly. That’s fixable. And although you can, in principle, build data centres in quite remote areas, most of the companies that are building them want them to be built near population centres for reasons to do with servicing them, things like that. So we maybe are missing… When I talk about street votes and things like that, all this is great. But we shouldn’t forget the basic incentives that local government should have, but doesn’t have — once had, but then had taken away — that we could restore I think fairly easily. Rob Wiblin: Yeah. I’m slightly worried that people’s eyes will be glazing over as we start talking about different levels of business rates and so on — but hook up, because this could be one of those subtle policy issues that doesn’t really get anyone going, that people barely even know exists, that could just be having an enormous effect on our built environment. You could imagine arguments in favour of “we should stick it all in a common pot and then distribute it,” or “we should keep it at the local level.” There’s good arguments either way, perhaps. But the effect of taking away all of the business rates, all of the property tax, basically, from developments would be the councils will reject them across the board, basically — because it’s all cost, no benefit to them. You have to compensate them somehow for the additional costs and issues that they’re going to be incurring. Maybe it’s not so obvious on the tin, but it could just be an incredibly important issue. Sam Bowman: Yeah. And I know that property taxation is boring. Local government is just fundamentally quite dull. So I do get it. But here’s another way of thinking about this. When we think about not being able to build things, we can get really bogged down in the thousands of pages of rules and reviews that need to take place. We are talking about one of the most insanely complicated systems — and this is true in the US, as in the UK, as in most Western countries, especially English-speaking countries. There’s an unbelievable thicket of problems. Now, we do need to solve that problem. I would argue that the solution is to create new structures for approving them that are simple. I think we should just cut our losses, declare bankruptcy on the old systems. They just don’t work. But we need people to fight at a micro level for stuff to get built who are not currently fighting for it. So the reason that I’m so obsessed with changing incentives is that I don’t want to be the only person to want more data centres to be built. Rob Wiblin: You can’t be at every council meeting. Sam Bowman: Yeah. I don’t want us to have to persuade the population of the country that it’s good to build data centres, and so they should just put up with ugly data centres near them — although they should. But what I want is for people who have no personal interest in AI or data centres or in the UK’s technological progress to just say, “Well, this will mean that my school will get a new classroom,” or, “This will mean that they’ll fill in the potholes on my road. So of course we should do this.” I think it’s aligning incentives like that that will help us to solve all of the other problems we need to do. And I think that’s why we should do it first. We should focus on getting incentives fixed, creating an army of people who will directly benefit from new things being built, and then we can say, “Now let’s look at the environment agency’s review process. Now let’s see how much political power the people talking about a bat tunnel have when they are not just making the country worse off, but the local people around that bat cave directly worse off” — because they have, for example, eroded the amount of surplus that could be paid to them. Rob Wiblin: Yeah. So the business rates, the property tax rate going towards local people, the local council, so that they can provide services or lower taxes or whatever, that seems like kind of the low-hanging fruit here. But it sounded like you also wanted to then move beyond that, and also start thinking about how can we stuff more gold into more people’s mouths in order to get more things done? I guess especially the tax revenue stuff is going to be going to a medium-sized jurisdiction — to a borough, to a city council, that kind of thing. But often we’ll probably need to compensate the people who are really close to something — that is, the people living right next to the prison who might be quite frustrated by it. But you’ve kind of been alluding to the ways that that can be quite problematic. So how do you distinguish the people who are really harmed by this — who do deserve compensation, and will fight really hard to oppose it; they really will try to exercise their veto to stop this data centre being built right next to their house — versus the people who are just pretending, because maybe they haven’t really thought about how much they’re going to be harmed, but they will speak up against it in cheap ways though they’re not difficult ones? It’s very hard to distinguish them, because they’re not going to announce themselves. You would have to put costly barriers for people, like make it very hard for them to object, in order to distinguish the people who really care versus the people who only nominally care. And you can also imagine that once compensation is on the table, even the people who don’t intrinsically care about the thing have a reason to pay the cost and try to pretend that they really are harmed, that they’re super upset about the bats or whatever, because that might be a way of extracting money. They might just be quite selfish in that way. So how do we get around these practical issues? Sam Bowman: Sometimes it’s easy. Sometimes it’s straightforward. Sometimes we have empirical evidence that people on a street care much more about what happens on their street than what happens on a street two streets down. We just can tell, by looking around the world and looking at what happens in the UK right now, that is just what people tend to prefer. So that’s reasonably easy. I think once you start to try to expand the specific model of votes, then the problems you’re talking about arise. Personally, I think that the vote-type model probably is only really scalable for really, really important projects — where there is so much surplus that you probably can buy in or bundle in large numbers of people. Because I agree with what you’re talking about. Otherwise we would have to be coming up with really precise definitions of adverse ill effect. So I could imagine, for example, nuclear power plants being subject to a local referendum. That would be one example. And we need a lot more nuclear power plants, but we don’t need thousands — we need dozens, or maybe hundreds if I’m being really ambitious. I mean, talking about hundreds would be genuine energy abundance. So we need dozens, but we would like over 100. Though I think the best way to think about this kind of approach is a general approach to getting policies working well. So when it comes to energy, there is a policy called nodal or locational pricing. At the moment, you have a single price for electricity across the entire UK. And most countries have this. Some countries don’t, incidentally. New Zealand has locational pricing. Rob Wiblin: I think Australia too. Sam Bowman: And in that case, you can price electricity according to where the generation is, and the cost of bringing the electricity from the generation to the user. That means that you have a really strong incentive to be OK with a new wind farm or a new nuclear power station or an electricity pylon, or whatever it might be, because that will mean directly cheaper energy for you. Now, that is not being done via a vote. That is being done via normal approval mechanisms. But because the incentives are a bit better, you should get people who are just generally more amenable to this kind of thing being built around them. So it isn’t that everything needs to be done through votes. I think sometimes it does. Sometimes you need the legitimacy of a vote to cut through a thicket of other rules and regulations; you basically need to create a new approval, a new institution for making approvals. But sometimes you just need to slightly improve incentives along the margin. I think business property taxes are an example of that. I think locational pricing and electricity is an example of that. As I believe has happened in France, when people who live near the new electricity supply get discounted energy bills, that’s an example of that. Sometimes you just need to be pragmatic about who is going to object and how do we make it so that they can benefit from this instead. It doesn’t need to be through this single model of votes. Votes are just a very powerful way of replacing a really dysfunctional existing institution with one that might be a bit simpler. Rob Wiblin: I could see two different ways that you could approach this. One would be to think about it in terms of justice and fairness and equity: thinking that we need to figure out, as accurately as we can, who is harmed by exactly how much, and then compensate them that much and then a bit extra for their trouble. You could also imagine someone who’s just more ruthlessly realpolitik, who is like, “I want to build this thing here. Who do I have to pay off, and exactly how much do I have to compensate them to basically get a sufficient number of people on board that it will be approved by whatever committee has to approve this damn thing?” How do you think about it? And are there any proposals for doing maybe the latter thing a little bit, and trying to get developers to just be able to provide whatever compensation people find most appealing in order to get a majority in favour of something to go ahead? Sam Bowman: I think about it as a mixture of the two, honestly, and I don’t really think about the first one as a fairness issue. I sometimes do, because fairness is kind of unavoidable as a thing that we care about. All human beings care about fairness to some extent. But I also care about it just from an efficiency point of view. It would be inefficient if we were building high-rise towers in suburban areas without concern for the effect that would have on that area, right? Rob Wiblin: Because it could end up being a net harm. Sam Bowman: It could impose huge costs on other people, to the point where if we correctly priced those costs, it wouldn’t be worth doing it. When we talk about carbon taxation, we are basically saying that it looks like burning coal and getting cheap energy is really great, but once you factor in all of the costs of doing that, it’s actually not great at all. It’s actually pretty bad compared to other ways of generating electricity. I see other externalities as being less severe. Coal is a pretty extreme case of a negative externality. But I think a really ugly building, or a building that stops you from getting sunlight where you used to have sunlight, those are pretty important negative externalities if you’re affected by them too. So partly it’s fairness, because those issues do relate to fairness as well, but partly it’s just that it would be a bad system to let anything be built anywhere, and it wouldn’t be the system that anybody would opt for if they could choose. We see this in the United States a lot: there are housing covenants, there are homeowners’ associations. People want to have collective control over what can be built and done in their area. They want quietness. They don’t want parties going on until 5:00 AM every night. They want to have some control over the area outside of the footprint of the land that they own. And I don’t think it’s that easy to generate those sorts of rules from nothing. If you have an existing built-up area, you would need everybody to opt in, and you’ll almost always get holdouts and things like that. But where we can approximate that kind of collective decision making over things where there is a legitimate collective interest, then that’s the model that I have in mind. When it comes to being ruthless, I mean, the goal here is to build a lot more housing, right? If this idea of improving incentives turned out not to deliver lots of new housing, then it wouldn’t seem to me like a great idea. But empirically, looking around the world, looking at examples that are a bit like this today, it seems to work really, really well. It seems to be the case that people have different levels of objection to new homes and new developments and new infrastructure. It seems to be the case that people who object will do so very loudly, and in a way that kind of deafens out all of the quiet people who are happy to consent. When we look at somewhere like Houston — which managed to upzone the entire city, and go from single-family housing across the entire city to one where many more townhomes and many more apartment buildings are being built now — the way they did it was to just let the people who really objected opt out, to let the block just say, “No, we’re not going to do that. We’re going to keep our old single-family zoning approach.” That is solving for the thing you are trying to do within the political constraints that you are in. And you know, the thing I always say in the UK — and I think this applies to California as well, and lots of the English-speaking world — is that we imagine that we’re going to smash the NIMBYs, but the NIMBYs smash you. So I’m often frustrated by my friends who think, “I’m really radical, I’m really bold, I think we should just do upzoning for the entire world or the entire country” or whatever. It’s like, even if that was a good system — which I’m not convinced is the optimal system, but let’s say it was — you’re not going to get that done. There is no government that is going to pass that, or there are very few governments that are going to pass that. Rob Wiblin: And if they did, they would be immediately voted out and it would be reversed, because most people are against it. Sam Bowman: Right. So why don’t we consider this to be the huge civilisational problem that it is, and rather than talk about the thing that we’d like to do in a fantasy world where politics didn’t exist, talk about the thing that we can do — using the surplus that is on the table to change the rules of the game so that people will do the work for us. That’s the challenge. Rob Wiblin: This opting-out system was the third proposal that we want to talk about. Can you elaborate a little bit more on that, and explain why it makes sense in theory? Sam Bowman: Yeah. Most of the things I’ve talked about so far I think we can conceptually think of as opting into a different system. Many of the problems that we’ve got are a very bad system that favours incumbents, or at least some incumbents, or vested interests if you prefer, who for one reason or another — either because they’re mobilised, they’re energised, or because they have a loud platform or whatever it might be — are able to keep everybody in this old bad system that immiserates people in general, but makes them better off. So one solution — the street votes solution or the TAMA 38 solution — is to opt in to a different system: keep the old system as it is for most people, and allow people to opt into a new one if they want to. A different approach, which is what Houston did, is to allow those people to opt out of a nationwide reform. And this is very contextual; this depends on the politics. This says, we are going to upzone this place. In Houston, they went from a system where you had large setbacks, so you couldn’t build a house close to the road; you had large restrictions on the minimum size of a house — I think it was 5,000 square feet. And although Houston sort of famously says it has no zoning, what that means is it doesn’t have rules on whether you can use a particular plot for a business or for a house or something like that. It doesn’t mean that they don’t have rules about what you can build and where: it does. And these rules, which were very, very geared towards single-family homes, were then repealed or weakened to allow much more dense housing: townhomes, or what we would call terrace homes in the UK, or apartment buildings — not skyrises, but medium density. Now, this kind of proposal probably would have faced huge political objections had it been imposed citywide. So instead, what they said was that homeowners’ associations — or, later, other agglomerations of people, other groups of people in a local contiguous area — can sign a petition, have a local vote, and if a majority of them does not want this change to what you can build in this area, they don’t have it. The old system will apply to this area. And you can look at a map — either on Works in Progress or on the Houston town planning authority’s website — of the places that have opted out. And it’s pretty sizable, but it’s less than a third of the city — 20–30% of the city. And that’s really intriguing. When you think of most of the cities that we would like to upzone, number one being San Francisco — probably nowhere in the world that would benefit more from an upzoning — and London may be number two; there are probably a few others, like Seattle and Washington, DC: the number one problem is these incumbents, who like things being the way they are, who stand to lose from them changing, will fight tooth and nail to keep their neighbourhood the way they want it to be. And if you are planning on doing a citywide upzoning, the way to fight to keep their neighbourhood the way it is is to keep the city the way it is. Now, opt-outs allow you to kind of have your cake and eat it too. The rest of the city can do what it wants. People who really, really object to that can keep their neighbourhood the way it is, provided their neighbours agree. And that gets you possibly a kind of greasing of the wheel, and possibly allows you to get most of the benefits of an upzoning without having to fight this huge terrible political battle. Rob Wiblin: Right. So if you’re someone who doesn’t want your street developed, you don’t have to get involved in the citywide politics. You can let the city policy be what it is. Instead, you choose to go and coordinate with your neighbours to set the rules for your street. And that’s a less generally harmful diversion for their energies, and actually allows them to satisfy their preferences without causing so much collateral damage elsewhere. Sam Bowman: Yeah. Or your block. I think in Houston it was blocks rather than streets specifically. Rob Wiblin: Maybe the thing that doesn’t quite make sense about this to me is: if we’re saying the NIMBYs are everywhere, they’re among us on every street, and they’re so powerful, why did only 20% of the streets opt out? Wouldn’t most of the people have wanted to not have their neighbours or other people near them building medium-density housing? It’s a bit confusing. It suggests that maybe there is a larger group that is kind of indifferent, and if the default is that it’s permitted, then they aren’t going to take massive action. Sam Bowman: Yeah. I don’t think we should generalise too much from the Houston experience. Hand on heart, I would like somewhere like London to try this, but I would not predict that it would pass. Or at least I would suspect that this kind of approach would be considered to be… People would have both city-wide objections to this, and you would get much more opting out. London is a much more historic city. There is much, much more of London that was built prior to 1914, is historic listed conservation areas, yada yada. My guess is that this is somewhat related to the fact that Houston is a fairly new city, and that Texas in general has a fairly permissive attitude to building things. And famously, Texas builds far more renewable energy than California does, because it’s much easier to build there. Not because they love renewable energy, but just because they have a different attitude to building things. So I think that probably it’s related to that, and I don’t think that the opt-out system is the magic wand or the magic secret sauce, because I don’t think it aligns incentives that well. In certain contexts, it may be a good idea. It might be worth trying or doing polling in various cities, but I wouldn’t expect it to be universally acceptable. And I’m not sure that it’s that great of a tool, because as we’ve seen in California, if you don’t also solve the other costs that we impose on new housing… I mean, affordable housing is something that we should probably talk about — because it sounds so good, right? I talked so much about the surplus available from building new housing. People recognise that and think, great, we can use some of that for affordable housing. But we don’t need to get into the general question of, does it even make sense to think about affordable housing? Rob Wiblin: All housing is affordable if someone’s there. Sam Bowman: Yeah. And surely the goal ought to be to build enough housing that it is just cheap — because it’s cheap because we’ve built a lot of it, not because we’ve sidled off a little bit of it to be subsidised. But anyway, let’s put that to one side. Affordable housing, if we want it, would make so much more sense paid for out of general taxation. Because it’s a social thing that we have said, as a country or as a city, we want this to be built. At the moment — Rob Wiblin: Why does it have to be funded entirely through a tax on new housing? Sam Bowman: On the thing that we want, on the thing that would make housing more affordable if it was built. And citywide upzonings: great, fine; I’m certainly not against them, and I think it’s really impressive when they happen. But if they don’t have some way of fixing those other problems, then they’re probably going to deliver less housing than you might want, depending on where you are. Houston didn’t have that problem. I suspect San Francisco might have that problem. I suspect London might have that problem. We’ve talked almost exclusively about within-city densification. With greenfield development, I talked about this gap between the cost of farmland and the cost of the price of housing-developable land being 140x. There is a slight temptation — and I think Labour are experiencing this right now — to basically just see dollar signs when you see that. It’s impossible not to. There’s an entrepreneur in all of us, and it’s impossible to not see some dollar signs. But then to kind of eke out, eke away every single dollar between the farmland and the housing land, and spend it on affordable housing, spend it on infrastructure — which it should cover its own infrastructure; that is a very legitimate thing — spend it on a tax for the general local betterment and things like that. And before you know it, you have actually run out of surplus, and you’re back to where you began — because it’s not viable, and the project doesn’t pencil. Like we have rules now for biodiversity net gain. New housing has to increase biodiversity. It can’t be neutral. It has to actually increase biodiversity — not necessarily on the site, but generally. We have, like I mentioned earlier, second staircase rules which effectively stop you from building the kind of New York-style dense apartment building, the kind of Friends-style apartment building. Rob Wiblin: It stops exactly the kind of apartment building that I imagine most listeners would like to live in. Sam Bowman: Yeah, yeah. The second staircase rules are like a ban on mid-rise development, because once you have to add the second staircase, you need to make the building a certain height. And you know, citywide upzoning is great, absolutely — but unless you are getting rid of that kind of rule as well, then you might be surprised at how little extra housing you get. Rob Wiblin: Yeah, second staircases might sound like a slightly boring niche issue to listeners, but I’ll stick up a link to a great YouTube explainer of just how destructive that rule is. And didn’t the UK government analyse what would be the benefits and costs of having a second staircase rule, and they concluded that the cost will be like 100 times the benefit? The argument is that it’s better for fire safety to have like two routes out of a building. But not that many people die in house fires anymore. We’ve done wonderful things, actually, on preventing fires from destroying property. Buildings are much more fire resistant than they used to be, and so the benefits of the second staircase are really negligible. But they went ahead anyway and imposed a second staircase rule. That’s how grim things have gotten in the UK: a government that literally thinks the costs are a hundredfold will still approve it. Sam Bowman: And very funny, because this has been the subject of a very successful campaign in the US to scale back these rules. So just as the US is getting rid of these crazy rules, the UK is introducing them — partly because there was a terrible apartment building fire. This was in 2017 at Grenfell. Many people died. Awful, awful event. And it’s understandable that no government wants to either let something like that happen again, or be seen to let something like that happen again. I actually suspect the latter is a much more important factor here, because the impact assessment that tried to estimate the benefits of this found or estimated that a tiny number of lives would be saved. Actually, I think the costs are 240 times larger than the benefits. And these impact assessments are usually kind of whitewashing exercises. Rob Wiblin: They wanted to pass it, they couldn’t even make it look at all — Sam Bowman: Yeah. I mean, these impact assessments are not being done by — not that I would actually want this — but by this sort of independent body. It’s a junior civil servant usually doing this. But still found basically no benefit at all. And it means that you end up with, you just cannot make a mid-rise development “pencil” economically — that’s the word people use. With this kind of requirement, you have to give up way too much floor space for it to work, so you’re forced to build a high rise. And to be fair, the second staircase rules only kick in above a certain level. So we’re still going to get six-storey buildings; we might not get nine-story buildings; we certainly are not going to get 12-storey buildings. But we’ll end up with a huge gap where you can build up to six storeys or you can build from, I don’t know, 15 storeys — but you get nothing in the middle. You get none of that kind of middle. And just to bring this back to this whole point about incentives: what we want is for every homeowner in the land to say, “This is crazy! This is holding back the prospect I have for doing a vote on my property and upzoning my street.” We want every single person to have a stake in a rational regulation of safety. Safety is so important, but we need to do it rationally, and not pass things that cost billions and benefit basically nobody. It’s so crucial to fix the rules so that it isn’t just me sitting here with you and telling you this is a bad idea. It’s everybody saying, “This is bad. This makes us worse off.” Rob Wiblin: OK, so we were talking about the opt-out system, which you’re saying could be good in a place like Houston. We could imagine this is a tool that could be quite useful in emerging, growing cities that have maybe a more entrepreneurial, freewheeling mindset, but perhaps it’s not going to work in a place like London or New York. So we’ve talked about three different approaches that are conceptually related. What are the prospects for getting these things up? And maybe can we think about how listeners could help: what work is being done to try to solve these problems using these tools, and presumably others that people like you are dreaming up? Sam Bowman: I think there’s a lot that people can do. The best thing that people can do, I would say, is look at the international examples of successful approaches. You know, one of the things that we’re trying to do at Works in Progress is to produce these case studies. We have published about Israel’s programme. We have a piece in the pipeline about South Korea’s programme. We published about Houston. We’ve got a great piece on New Zealand‘s nationwide upzoning that partly got reversed, but then they’re doing other things instead. And we’ll probably do a followup piece on that. We’re doing a really interesting piece on Japan’s programme of land readjustment. I’m very interested in how much this transfers out of Japan, because as probably most listeners know, Japanese housing stock has to be very young because they tear it down after about 30 years — because of earthquake risk, because there’s a culture of just building relatively cheaply and then replacing it. Very different from the culture, obviously, in the UK, for example. Rob Wiblin: Never take down a house, ever. Sam Bowman: I live in a house that was built in 1830. This is one of the oldest houses in Japan. Rob Wiblin: But the longer it’s been around, the more you have to keep it around for as long as possible. Sam Bowman: But the Japanese programme — as I understand it, and this is ongoing research — allows landowners of usually not-built-on land or usually very-lightly-built-on land, you’ll understand why in a second, of adjoining land, to effectively redivide up their land and collectively sell a little bit of all of their land for development. So you could do this for train lines, high-speed rail lines, you could do this for new housing. And what it does is it avoids holdout problems, because you make these decisions via supermajority. You say that the local government specifies, “This is the group of people, and this is the land that will be affected by this. This is the proposal. You will all be affected by this in an equal way.” So you don’t have the problem that you usually have with holdouts, where one person gets their life taken away from them and like, lucky you, you’re going to get a bit of market compensation for losing your family home and everything. But I think it doesn’t really work other than on either non-built-on land or in a place where people are kind of expecting to knock down this house in 10 years anyway. So we have a piece coming on that. Anyway, to answer your question: number one, help us find more case studies like that. There are cities that are working on this, there are places that are working on this. Help us find them, write them for us if you like. And then try to do this sort of thing in your own city. There is so much that can be done from people picking up these ideas. Street votes is one, estate regeneration is one. There are loads of other ideas that can be developed. I really feel like this is an intellectual frontier, a really exciting frontier of thought and research about what is being done and then adopting and remodelling some of those ideas to really sharpen them, and then just bring them to local government, bring them to local politicians. And we need more and more, as a friend of mine put it, “polities worth copying.” We need more. We don’t just want it to be Seoul and Tel Aviv. We want it to be Bilbao, we want it to be Munich, we want it to be Vancouver. There are so many places around the world that just a few places getting this right can be absolute tonic for the rest of us to be able to say, “Look, they did it! And look, it does work!” Rob Wiblin: It’s such an odd situation, because I guess this policy area of land use reform has turned out to be a $10 trillion issue, or it turned out to be one of the most important policy issues of our time. I guess in some ways it might be a difficult policy area to pursue a career in, because there hasn’t been such historical interest, there probably aren’t tonnes of academic programmes on this, and at least historically, there haven’t been many think tanks, I would guess, with big programmes on it. At the same time, as people have realised just how large the stakes are here, I think there’s been increasing philanthropic interest. So I imagine that if you’re someone who is extremely talented, you could find yourself at the top of this topic area, at least in your country or in your city, relatively quickly. Maybe I might be pissing people off by saying that. Sam Bowman: This is a very 80,000 Hours mentality. Rob Wiblin: It’s somewhat neglected, at least historically. Sam Bowman: I do think there are a lot of really good people working on this. And we have a piece on homelessness and housing shortages: it sounds, on the face of it, completely obvious. Why would housing shortages cause homelessness? It sounds completely, “Duh, because there aren’t enough houses.” But then you think about it a bit, and you realise that many people who are homeless have really serious problems with drug addiction, mental health issues, other kinds of individual problems that don’t easily map onto just not having a place to live. The piece that we’re running is by a guy called Salim Furth, who works at the Mercatus Center where I’m on the board, and really interestingly argues that the reason that many people get pushed into homelessness is that they have some of these underlying problems and when housing is scarce, they don’t have a friend or a family member that they can stay with during times of crisis, they don’t have a spare bedroom. So rather than being able to rely on somebody who does have a more stable life, who does have a spare bedroom, who can help them through that time of crisis, they end up relying on either much more unscrupulous, nasty people or just directly having to go onto the streets. Rob Wiblin:And that makes a negative cycle, right? Sam Bowman: Exactly. And that to me was a revelation. Reading that was like, “Ah, I see.” And he’s got really interesting evidence to support this. It’s a really interesting piece. So there is some really great work being done. Some of the people I’ve cited already are really great. But I will say I think the solutions part feels like a very fertile bit of ground. And having worked for a long time in public policy, it’s a little bit boring sometimes when you get into an area where all the solutions are done — like we know what the optimal tax policy is, and it’s really hard to imagine how to implement the optimal tax policy. The really interesting stuff is where you either don’t know what the solution is, or getting it into existence is the really challenging bit, and that’s quite contextual and that requires very careful research and design. And I feel like housing, anything to do with the built environment, is an example of that. Rob Wiblin: OK, let’s push on from housing permissioning, but think about another kind of cross-cutting, maybe generally applicable approach that you’ve been thinking about recently for greasing the wheels of positive change and overcoming incumbents who might be against it. This is setting up parallel new and old institutions, and allowing them to run at the same time. Can you explain this whole idea? Sam Bowman: Sure. The underlying intuition is that when I’ve worked at or with companies that work well, and when they have a problem, or a team that isn’t performing that well, they will often not try to fix the team: they will often set up a new team. And we don’t really have anything like that in the public sector. I mean, this is how markets work. It’s almost never the case that a failing, struggling company gets turned around. Sometimes it does. Microsoft, good example — but it’s a very notable example. Most of the time, a company just ends up becoming very bureaucratic, becomes kind of gripped by the problems that it’s had and just dies, and a new one comes along to replace it. And there is a substantial period of time where the new one coexists with the old one. Because basically that’s how markets work. And also, if we were planning markets, we wouldn’t want to just put all of our eggs in the new basket. Even when it comes to something like legacy software, the big disasters that happen when a company is reliant on an old software system and moves to a new one is when they try to do the switch overnight. They try to say, “OK, the old system’s not working, everybody tomorrow is going to move to the next one.” Almost always that goes wrong. When it works — and I spoke to somebody where I was very reassured, because when I talked to him about this recently, he gave the example of this and said, “This is how you should do a legacy transfer, is have these things coexisting” — you have the new software, and people can slowly inside the company move to the new system, and you get a kind of a discovery process and you find the errors before you have switched entirely to the new errors, to the new system. But we don’t do this in public policy for the most part. We certainly don’t do this as a general approach. And where we have done it, it seems to me that it works pretty well. The one that I think is really interesting: the UK science funding is not in a great place. I would argue UKRI, which is the UK Research and Innovation Agency, is really not that good. It gives money to reasonably unambitious projects. Most people who interact with it don’t like it very much. It’s slightly gripped by these cultural battles that lots of grant-giving organisations seem to be gripped by. And it has more than £10 [billion] to give out every year. So one approach would be to fix that, put in new people, try to fix the institution. That would be great. But every single person who benefits from the current setup will fight you. And those people are pretty powerful, right? They’re powerful within the organisation, they’re influential, they can say things that make you look like an idiot if you try to fix it. It’s really hard to do that. Another approach is to say, why don’t we just set up a new thing? And that’s what we did, right? So we set up ARIA, the Advanced Research and Invention Agency. I love the fact they use the word “invention” rather than “innovation” — so much better as a word. And ARIA is, for its part, loosely modelled on the DARPA model, but I don’t think that’s the thing that’s interesting about it. I think it’s really cool, and the stuff that they’re doing is really interesting. They have some people working on robotics as inspired by the animal kingdom, some people working on stuff like machine-brain interfaces and noninvasive versions of machine-brain interfaces. Just really interesting stuff. But that’s interesting on one level. I think on another level what’s interesting is we said we’re not going to try and fix UKRI and make it more bold and adventurous. We’re going to make a new institution. And if it works, it works. If it doesn’t work, it doesn’t work. We’re going to protect it basically from public scrutiny. It has been given a mandate that it has a certain number of years where it can do what it wants, and only at the end do we judge it — which I think gives it a lot of freedom, and it’s a really intelligent approach and actually was my biggest concern about setting it up before it happened. But I think it speaks to a bigger solution to the problem we have of both regulatory and public agency dysfunction, where fighting to fix the regulation or to improve the thing we’ve got is both hard and slow. When we have any kind of time pressure, that can be really, really bad. So sometimes just setting something up new and having it work in parallel seems to work well. So ARIA is one example. The UK’s vaccine procurement programme basically was just set up as a whole new thing. They recruited a brilliant woman named Kate Bingham from the private sector. She was a venture capitalist, and she — unlike the European approach, which made big bets on a very small number of vaccines and said economies of scale are the way to go — made smaller bets on a very wide number of vaccines, assuming that one or two might work but many others would not work. And that’s exactly what happened. And the result was that we got in the UK, much earlier than the rest of Europe, access to vaccines. And so we were able to lift lockdowns and things like that. In the US, many people who are listening who are in the US either will have or should have a 401(k). This is a tax-protected retirement account. This is a very important thing. I’m really into personal finance on the side, by the way. And if you don’t have a 401(k) and you’re in the US, get one — look into it and get one, because they’re very useful. The 401(k) was invented by accident. It was invented when there was a need to clarify the rules around employer-funded retirement counts. This is at the end of the 1970s, beginning of the 1980s. The expectation was that this would be a tiny bureaucratic clarification of a problem. This would have no impact on overall budget, yada yada. It clearly has ended up becoming one of the most important savings vehicles on the planet. It’s a crucial part of most Americans’ retirement planning. The solution — and this was kind of accidental, but the effect was the same — wasn’t to say, “Let’s take people’s existing accounts, let’s change them, let’s let them do this,” because there will be winners and losers from that. The losers are often more likely to fight you than the winners. It was to create a new vehicle that people could just opt into. And I’ve talked about this a lot, but when I think about stuff like street votes, estate regen, whatever, again and again, one of the elements that I think is really appealing is we are building a parallel setup that has legitimacy — legitimacy is really important — and can make a decision that considers the project in the round, considers all of the benefits, all of the negatives, and just says yes or no. When we think about building the tunnel under the Thames or building a high-speed railway from LA to San Francisco, again and again these things just get tripped up on this incredible thicket of objections, reviews, agencies. And a very simple solution might be to come up with a really straightforward… And you know, sometimes people laugh at this, because I’m not talking about standard setting. Standard setting needs everything to use the same thing. And there’s a very famous XKCD comic where, you know, “We’re going to come up with a new standard” and there are like 14 standards. So this is not standard setting. This is effectively opting out of a system that will be too difficult and complicated to fix by itself, and just creating a new parallel system that it is easy to move into if you want, and may get you a better agency. And one of the reasons that I think economic growth is so essential to doing broader public sector reform is that if this is true, then we need quite a lot of extra money flowing around — because it’s expensive to fund new institutions, and we can’t just have like parallel health systems. I don’t think this would work for that. But in a context of economic growth and in a context of prosperity, it becomes much more viable to have these sort of competing institutions than in the world we’re in right now, where everywhere except the US has had almost nonexistent economic growth for a decade. Rob Wiblin: The thing that doesn’t quite make sense to me about this is, I think probably it does work in reality, but I almost don’t know why. Rather than single anyone out as doing a bad job, let’s just talk about Old Org and New Org. So if the worry is that we can’t reform Old Org, because the people inside it benefit from the way it is now, and they’re going to fight tooth and nail to maintain the way that things are currently done. So you couldn’t just shut it down right away, you couldn’t change it internally. But we’re saying that they will tolerate you setting up a new organisation in parallel, knowing that this could plausibly end with their old organisation being sunsetted basically, or being gradually scaled down over time, their funding decreasing and the new thing taking over. I guess maybe it’s just that it’s not so salient to them that things are being changed. Or I guess you could imagine if the change is going to take decades, then maybe people don’t really care because they expect to retire by the time that the organisation that they’re in is finally shut down. So you can just kind of decrease its funding so that as people retire, basically no one has to be specifically laid off or you’re taking advantage of people’s short-termism, perhaps. That they think about their job for the next few years, but 10 or 20 years out, they don’t have the bandwidth to think about it. Am I along the right lines? Sam Bowman: I think it’s a fair objection in theory. For one reason or another, it doesn’t seem to be the case in practice. Often because we’re not literally talking about vested interests, I think. So I’m worried about talking a little bit off reservation here, because I don’t know enough about the details of this, but at some point in the last 20 or 30 years, France liberalised its childcare regime — but it didn’t do so by changing the rules for existing childcarers. It created a new thing that you could be as a registered childcarer that would be better for everybody. It would mean you could have multiple childcarers working in a single property, you wouldn’t have to rent out new premises. You could have a better ratio of staff to children, as in fewer staff per child or more children per staff member, basically cheaper. But obviously anything around childcare comes with huge risks, because parents are worried and so on. And in doing a kind of parallel institution — again, I haven’t done deep research in this, so I don’t want to say this too authoritatively — they seem to have been able to get pretty much everybody, I think, switching over to this new kind of vehicle, but not having the political problem of, “By the way, the government is now saying your childcarer can do this.” It was a kind of a voluntary switchover. Rob Wiblin: So you can stay in the old childcare arrangement if you like it, but you also have the option to switch over, and then eventually you just notice that almost everyone switched over. So you can, at that point, once people have conceded overwhelmingly that the new thing is better… Yeah, I see. Sam Bowman: I think when it comes to a science funding agency and a parallel one, I think they will fight you. They’ll probably have less direct incentive to fight you. There is a slight collective action problem that they face, because for the time being, their incomes are protected. And they probably will fight you. I think the powers that be did fight ARIA. So I don’t want to claim that this is a magic solution. Rob Wiblin: They’ll fight you a lot less than if you said, “We’re firing you tomorrow.” Sam Bowman: Yes. And I often kind of joke, but it’s not really a joke, that often we should just be pensioning off civil servants and saying, “You can have your income for the rest of your life. Here’s a bond that will pay out your salary. You don’t have to do any work anymore.” That would be expensive, but it would also mean that we don’t have to replace them. There’s no self-sustaining. Effectively, you can fire a lot of people without them caring, and you don’t need to rehire people in their place. I don’t think that’s obviously always scalable. And I also don’t think that I, sitting here, have this big list of these agencies that I’m going to afuera these ones out and get rid of all these. So I don’t want to overstate that as the magic solution. And obviously, all this costs money. But I do think that there is a really important intuition that if we want to get improvements in regulation, then fixing the old regulations is often way, way more complicated and difficult than creating new ones, copying ones from other countries and just letting them exist in parallel. If we want a new regulator, yeah, the regulator will fight us — but it won’t fight us nearly as much as if we try to change the practices of the regulator. Some of it’s cultural like that. Rob Wiblin: This is completely off piece, but I think the way that we got the nobility to give up their arms and to accept a more centralised government and eventually being less powerful was just basically buying them off, compensating them for the loss in the short term. So they accepted that their grandchildren wouldn’t be as powerful as they were, but at least they were protected in the short run. And they were impatient, so you were able to kind of buy them out. Sam Bowman: I think the Meiji Restoration involved a lot of this as well, actually, with the samurai. Rob Wiblin: But I guess this could be our next episode. Sam Bowman: Yeah. Yeah. Rob Wiblin: Pushing on: you’ve told me that you think aesthetics is underrated as a way of getting people to go along with stuff. Tell us about that. Sam Bowman: Yeah, this has been a personal journey for me, because I’m quite a utilitarian person. I do care about aesthetics and I do care about beauty — but I also just love optimisation, and I’m a cold economist at heart. I have definitely been convinced by people — like my friends Nicholas Boys Smith and Samuel Hughes — that a really underrated factor in why people oppose new housing is that they consider the new housing to be ugly. There is a very famous study by a guy called David Halpern that asked people on a university campus to rank 10 buildings according to how nice they looked. They got the non-architecture students to rank them, and then got the architecture students to rank them. And the ranking was exactly opposite: the architecture students had literally the exact opposite taste to normal people. And it’s architects who make most of the decisions, or town planners who are influenced by architects. If it’s true that this is a big factor, then this is a really low-hanging fruit in getting more development: by making the development fit with what normal people want, not what architects want. Rob Wiblin: Maybe we can’t crush the NIMBYs, but we can crush the architects. They’re a smaller group. Sam Bowman: Absolutely. Exactly. And there’s a really great piece by Samuel Hughes that talks about architecture being agreeable. He has basically two different kinds of art: art can either be sort of challenging, or it can be easy or agreeable. Challenging art: you almost always want art to be challenging. If you like music, you almost always want to listen to sophisticated music. You probably wouldn’t have liked it when you were 10 years old. You probably wouldn’t like it if you were just coming to that genre or that artist for the first time. But if you really appreciate that genre, you want the 17-minute-long Appalachian harp song, or whatever kind of black doom metalcore people like. If you’re a fan of literature, you don’t want to just read a Patricia Cornwall novel — you want to read a really great Orhan Pamuk novel or something like that. And almost always this is OK, because you enjoy these things privately: it doesn’t cost anybody else if you enjoy these sophisticated things that no non-connoisseur would appreciate or get. The one exception he argues is architecture, which is almost entirely enjoyed or experienced by people who have no interest in architecture, who cannot opt out of seeing this building because they live near it or they walk past it every day or whatever. And yet architecture is no question designed by, designed for, and decided on by people who have sophisticated tastes in architecture. So when you talk about a new building in the public space, in any public space, you’re almost always witnessing a conversation between architects. And when something looks ugly, you’re almost always looking at very sophisticated people talking to each other in a very sophisticated design language in a way that is often using ugliness as like a part of sophistication. A lot of modernism generally, like atonal music, is intentionally not meant to be easy to enjoy. That’s the point. And I think it’s kind of one of those things that when you see it, you can never not see it. And you realise that so much of the public space is being designed by people who almost actively want it to be ugly — or ugly to a philistine like me, anyway. Rob Wiblin: It’s as if we’ve allowed professors of literature to break into everyone’s house and force them to read Finnegans Wake at gunpoint. Sam Bowman: Yeah. Or like the way Soviet films were often these incredibly avant-garde films, because they were being designed by the most brilliant film students on the planet. But no normal person could enjoy it. So I think this is a really good insight, a really important insight. It helps to explain why it is that so much of the modern built environment is really not very attractive. And I think there will be some people listening who will say, “I really like brutalism.” Brutalism is the one that everybody focuses on. There are good examples of brutalism, no doubt, but it’s also quite a sophisticated thing to like brutalism. And you don’t say you like brutalism without a bit of you in the back of your head thinking, “I do actually know quite a bit about this, and I am quite clever and I am quite sophisticated.” Rob Wiblin: Or just like, you know, I could enjoy looking at this occasionally, but do I actually want to live next to this, be looking at it all the time? Don’t you want to feel more comfortable at home than looking at brutalism? Sam Bowman: Absolutely. But some people do. Some people do. And some people find pastiche really horrible and really ugly. They think it looks like Disneyland, and they think that’s really horrible and bad. That’s fine. Good for you. I mean, I’m glad I’m not that sophisticated. That’s great to be that sophisticated, but most people do not feel that way. And this definitely, I think, affects the experience and the feeling and opposition that people have to new buildings. Because this is such a direct effect. Rob Wiblin: And it affects such a large number of people. Because even if you live quite a long way from somewhere, you might have to look at it when you’re going to work. Sam Bowman: And think about London’s skyline. London has a specific rule that a new tower must be architecturally significant to get approval if you’re above a certain storey limit. So we have built into the system that every tower has to be a special unique snowflake kind of thing. And the result is that you have no background towers. I think it’s something that Manhattan and Chicago both do really well: they have statement towers. They have really striking buildings. Obviously Manhattan has the Empire State Building, the Chrysler Building, and there are others that I am less au fait with and all the rest. But it also has a lot of background generic towers. It has something for them to stand out against. Every single tower in London is a kind of wannabe Chrysler Building or a wannabe Empire State Building. Actually, if they were a wannabe Empire State Building that would be OK, because that’s so good. But we’re trying to make everything stand out, and stand out in a way that is impressive to architects. But I think it’s not that beautiful or impressive to normal people. This is why when I first encountered street votes, a long time ago now, I found the design element very quaint, and thought, “That’s nice that you’re putting in a design element. I don’t really understand why. I don’t really get it.” Now I get it. Now I understand. That’s actually, I think, most people’s… And I think if and when street votes hits the real world, I think most people’s experience of it is not going to be primarily as a density rule — although that might be economically the most important effect — I think it will primarily be as a renewal of town streets and city streets that are at the moment quite ugly, and built in a way that is really not pleasing to the eye for anybody, and replacing them with probably Georgian, some Edwardian, some Victorian. Georgian is the really, really powerful one that people love in the UK — and it’s very simple, so it’s not expensive to build. And when I think about the effect that a policy like street votes might have, I don’t want to get too obsessed with a single thing, but I can kind of imagine a London that looks far more like the most beautiful parts of London. You know: Belgravia, Bloomsbury, Fitzrovia. I think there’s no reason Clapham shouldn’t look like that or Camberwell or Kentish Town. Why shouldn’t those places look like those really beautiful areas that we used to build but we don’t anymore? And it’s so exciting to think that just letting people have a say over their street might get you that. You don’t need master planning, you don’t need a top-down decision about the way the city looks. London should be a patchwork of design. London really should not be a single design across the city, but it would be nice if it was a single design across a given street, I think. Rob Wiblin: The way I was thinking about this economically is that we were talking earlier about it’s not as if it’s realistic to go to everyone in the borough of Tower Hamlets and give them £10 to accept a building that’s going to be built miles from them. But you could make it beautiful, so that when they occasionally see it, they think, “Isn’t that great that we built this new beautiful building, and it improves the skyline?” Or they’re happy to see it when they’re travelling around the area. And they might value that about £10. And maybe this development that’s quite a long way from them, maybe it only cost them £5 in terms of harm, but by just giving them this small benefit, you’ve made them people who perceive themselves as overall in favour of the project, with much lower transaction costs. Sam Bowman: Yeah. I do believe a lot in momentum, in policy and in politics. Roger Douglas, who was the finance minister of New Zealand in the 1980s, has this incredible line: “You have to let the dog see the rabbit.” His point is that the momentum exists, and people are not going to vote for economic growth. You have to give them stuff that’s good, and they will vote to continue that. So I think getting a couple of really great projects around street votes, let’s say, off the ground — and letting that happen, and hopefully happen well; hopefully the design works and hopefully the people vote for a nice design and stuff like that — once people can see that with their own two eyes, that’s when the momentum, I think, will kick in, and demand that for themselves. You know, I’m from a city called Cork in Ireland, and I wouldn’t say very much of Cork is beautiful, but it has good urbanism. It’s very dense in a nice way, like narrow streets. Some of it is beautiful. It isn’t, I would say, a very historic city. There are some bits that are historic. But coming to London is just mind-blowing. And so many people I know from around the world, when they come to London, the main thing they want to do is just walk around the nice bits. And the nice bits are not the rich bits necessarily. They’re kind of rich now because rich people want to live in those bits. But they weren’t built as the rich bits. They were thrown up in a few years by the Victorians because they just needed stuff to build and they thought, “Of course we’re going to build this in a beautiful way.” Rob Wiblin: That would make it nice. Sam Bowman: Yeah, “Why wouldn’t we want to build this beautifully?” And I think that the prospect of general renewal that can come from that is really exciting. And it’s actually, to me, in some ways the lowest-hanging fruit of all. Because who could possibly object? Rob Wiblin: Well, I mean, architects — but I guess we’ll just have to outvote them. Architects: a city is not an art gallery. We live here, I’m afraid. Sam Bowman: And yeah, it’s weird. It’s very weird that we have come to this. There is some value to architecture, I guess. But there’s a guy on Twitter called Coby Lefkowitz who just tweets new buildings, mostly in America, that are really beautiful. And often they’re art deco; there isn’t a single style that everything ought to be done in. But it is really inspiring seeing this is happening. These are real buildings being made that are done in very beautiful styles. And maybe we’ll come up with new styles. I worry that architecture is in a very bad local equilibrium, in the same way that literature is as well, and some modern classical music: I don’t think many people listen to it, and I don’t think it’s that much fun to listen to, compared to 19th century classical or early 20th century classical. And I worry that architecture has fallen into this same kind of trap of doing stuff for other producers rather than doing stuff for consumers. But maybe we’ll break out of that. Rob Wiblin: Yeah. I mean, if architects want to build stuff that most people find incredibly ugly to have a conversation with other architects, that’s fine, but they can do it out of town. We shouldn’t be reserving all of these spaces and forcing the buildings to be ugly to most people. I guess one other idea for getting lots of change that we considered talking about is advance market commitments, but we’re going to skip that because Luisa did a fantastic interview with Rachel Glennerster about that a couple of months ago. So if you just scroll back in your feed, look for #189 – Rachel Glennerster. It was an excellent interview, and I think it pretty much covered the territory. So you edit Works in Progress. I guess you commission a lot of the pieces and decide what goes in it. The remit is “new and underrated ideas to improve the world.” And I guess some of the articles are about policy, the kinds of things that we’ve been talking about, but many are about emerging technologies or promising technologies — things that could be coming along that people are perhaps underrating and not investing as much in as they could. What’s one technology that you think is particularly important, and at the moment a bit underrated? Sam Bowman: I mean, it’s hard to say this is underrated because everybody’s talking about it, but I think that nuclear is so easy to fix and so important. And I worry a little bit that techno-optimists are too geared on California, and the brilliant minds of people who are in the southwest of the United States, who are probably rightly very bullish about solar for their areas. I worry that people who live in dark old northern European, British climates are forgetting that, you know — Rob Wiblin: It’s not California. Sam Bowman: Yeah. We get about a quarter of the amount of sunlight in the winter as you get in the southwest of the United States. And there are basically two really big problems that we’ve talked already about. Energy costs in the UK are absolutely astronomically high. Industrial energy prices are the highest in the developed world, and absolutely gigantic. And even before the Ukraine war, they were very, very high: they were more than double in inflation-adjusted terms what they were. Nuclear power, unlike wind and solar, because it runs 24 hours a day, doesn’t need massive improvements in battery technology; it doesn’t need massive overbuild of electricity pylons and things like that. When you have a megawatt solar farm in the UK, the capacity factor, you get about 100 kilowatts from it. You get about 10% on average across the year — arguably 20%, because that includes nighttime, which I think is slightly unfair on the solar. The amount of land that solar takes up is absolutely huge. You’re talking about eight acres for a megawatt, or between three and eight, depending on what technology you’re using. Rob Wiblin: And that’s fine in Texas, but almost everywhere in England, at least south England, is kind of spoken for one way or another. Sam Bowman: Yeah. And it is undeniable that the cost curve of solar especially is incredibly exciting. We may get to a point where photovoltaic cells are basically free. Like where they’re in paint, or they’re just being rolled out like paper. It’s just so cheap that it’s almost free. And in that case, maybe batteries can absorb the energy that we need so that we’re OK. Although I think in a place like the UK that has not very much sunlight and often has periods where it’s incredibly overcast for weeks, I think even then we’re not going to be all right. So on the one hand you have something like solar. I respect the solar bulls. I think they’re very brilliant people. I have learned a lot from them. But what they are saying will not work for the UK, and may not work for a lot of northern Europe, and may not even work for the northeast of the United States, which a lot of people live in. If so, we need something else. Wind is just not going to get as cheap as solar. Lots of elements of wind are no longer following anything like the kind of cost reductions that they were following because we were getting economies of scale. Certainly not getting to a point like photovoltaic cells, where we’re talking about this energy being free. And wind suffers from different, but the same kinds of intermittency problems. Nuclear doesn’t. But most importantly, nuclear used to be cheap. In the United States in the 1960s, nuclear power was three cents a kilowatt hour in inflation-adjusted terms. Three cents a kilowatt hour is really, really not much. It’s cost competitive with coal. The challenge is, how is it that we have gotten from nuclear being about the same price as coal to nuclear being by far the most expensive form of electricity generation? By some measures it’s more expensive than wind and solar in the UK — probably arguably no: if you factor in the need for constant power, it’s actually a bit cheaper. But that’s the challenge. So one thing we can do is look to history and say, what is it that changed? Clearly, Three Mile Island — which didn’t kill anybody, but did lead to very significant concerns about radioactive materials being released — was a problem. Obviously Chernobyl was a huge problem. Fukushima, which again, didn’t really lead to massive loss of life — actually, I think was kind of a model for the safety of nuclear, because if you can imagine a more unfortunate coincidence of events than a tsunami and an earthquake and a nuclear meltdown still not leading to anything like the kinds of deaths… I don’t want to put an exact number on it, because there arguably were some radiation releases. Nobody can really say exactly how many people died, but it’s closer to zero than it is to 100. So we’ve gone through a series of events that have led us to be very, very concerned about the cost of nuclear in terms of human life — without ever thinking about the alternative costs in terms of human lives from other ways of generating power. And if solar and wind are not viable options, then that means gas or coal, depending on where you are in the world, which create CO2 emissions, which causes global warming, which can kill people. Rob Wiblin: Releases particulates as well. Sam Bowman: Releases particulates. Air pollution and so on. So one issue is that we have a very disproportionate attitude towards nuclear power safety. Another issue is the way we build nuclear: very similar to the infrastructure projects we’re building, we’ll build the absolute most advanced reactor in the world. We’re building one of them, maybe two. We’re not building a pipeline of them. And when we build them, we create huge absurd environmental objections to them. Recently it was reported that Meta was unable to set up a nuclear plant to power a data centre that it wanted to because of some rare bees that were protected. In the UK, Hinkley Point C, which is the power station that’s being built, has had to install an underwater megaphone to play loud noises to scare fish away from the cooling intake system. And incidentally, there’s no danger to the reactor of the fish being brought in. The danger is to the fish. Rob Wiblin: We need to do some benefit sharing. Write them a cheque instead. Sam Bowman: Something like 120 fish per year that would have been brought in would have been from a protected species. So they were ordered to come up with a way to deter them from coming in. And this underwater megaphone, incidentally, because the underwater megaphone requires a diver to do its maintenance, actually will increase the risk to human life. And I know that some people listening will think that there is some kind of exchange rate between fish lives and human lives, but it’s definitely not 120 to one or anything like that. You know, it’s probably a few more hundred, because not all of them are protected. But then we can look to countries around the world that are building nuclear at scale, and building it pretty cheaply. One country was France, which did a huge shift to nuclear power in the 1970s. After the oil crisis, the UK went for North Sea gas and oil; France went for nuclear, and said, “We do not want this to happen again.” And France is now by far the most nuclear dependent. Something like two-thirds of French electricity comes from nuclear. The other country that does this very well is South Korea. France, unfortunately, has kind of ended up in the same place in terms of nuclear power costs that the rest of the Western world is in. South Korea has not yet reached that point. The South Koreans can build nuclear reactors for about a quarter to a fifth of the price that we build them on a per-megawatt basis, and they’re exporting them. So they built them for about, I think, a third of the price that we build them in the UAE. They have agreed to build two in the Czech Republic, Czechia, for half the price that we’re building Hinkley. And they want to build here. That’s the interesting thing: the South Korean company, called KEPCO, it’s a state company, they are lobbying to build in the UK, but so far I believe on deaf ears — but I don’t want to speak to what the government’s doing or not. So what do they do? They build a pipeline of them. They build fleets of reactors. They’re not building the best reactor every time. They don’t have the same thing we do, where we are approving and treating each project as a distinct entity. They say, “This is the reactor design. This is safe. You can build as many of them as you want” — within certain restrictions around population areas and stuff like that. So I think in the UK, nuclear is really exciting. It’s an old technology, it’s a fairly proven technology. I would love to sit down and talk about all sorts of interesting drone tech and stuff like that, but what I really care about is getting the basics right for the UK, and getting the basics right for the whole world. I think the UK is as good a place to start as any. And nuclear has so much going for it: we know how to do it cheaply; it’s a purely regulatory problem. There is no technological breakthrough that needs to happen. When we talk about small modular reactors, they’re great, they’re really interesting and really cool — but we don’t need them. We just need to be able to do large reactors affordably, as we used to and as already happens in South Korea. So one thing we could do is change regulation and try to fix it. Or — in line with my view that we should just set up new things — we should just say the South Koreans can build, and if it’s safe in South Korea, it’s safe here. I think that would be a very quick, easy way, and we could probably start it tomorrow. Rob Wiblin: Won’t locals object to this? I mean, we’re talking about the difficulty of even getting apartments built in the UK. Just starting to chuck nuclear reactors everywhere seems like it’s probably quite a bridge too far for us right now. Sam Bowman: Funnily enough, locals actually really like nuclear power plants. Rob Wiblin: Really? Sam Bowman: Yeah, because lots of jobs come from them. So the dynamic is places that have not had a nuclear power plant are pretty resistant to getting a nuclear power plant. But if you have had a nuclear power plant, or if you have a nuclear power plant — Rob Wiblin: Why not two, why not three? Sam Bowman: You’re desperate for them. So there’s a map that we put up in “Foundations“: it’s like a cartogram, so it’s a colour-coded map of wages across the country. And as you would expect, London and the southeast is a sea of red. The rest of the country is blue, except for in the northwest, by the coast, there’s a little island of red — and that island is Sellafield. Which is, if you’re Irish, notorious — because a lot of Irish people were worried about this nuclear power plant blowing lots of radioactive material onto Ireland. It’s being decommissioned, unfortunately, and the decommissioning is requiring a lot of money to be spent on very high-skilled jobs. So temporarily, there are lots of jobs here. But nuclear power plants generally just require quite a lot of skilled labour. They pay well, they’re very popular once they’re there. And places that have had nuclear power plants, but no longer do, seem to really want nuclear to come back — these are good jobs. So actually, I think that it might be that the nuclear power plant approval process, from a local point of view, could be a lot simpler than we think. There is still the absurd thicket of environmental approvals. The nuclear regulator, I would say, does not see its job as being to get nuclear built; it sees its job as being to get what nuclear is being built to be built in as safe a way as possible. You may be familiar with this concept of ALARA — as low as reasonably achievable — which is the idea that there is no level of nuclear background radiation that is safe. Even if we get, as we have with nuclear reactors, background radiation to the same or less that you would get in Cornwall — because of the sunlight that Cornwall gets — that’s not safe enough. Any productivity improvement you get, you have to reinvest in reducing background radiation. Rob Wiblin: I think this is something that many people might not know: that often this regulation, they’re saying you have to make it basically as safe as you possibly can — and that means that whenever you get any cost savings somewhere else, you figure out some way of making it cheaper, that means that you freed up money now to make the concrete even thicker, add another wall, add another pump, another redundancy. Anything, another redundancy. You’re really saying we can’t accept any above-zero risk, and then there’s almost no limit, no amount of money that you could spend that would be too great. Sam Bowman: Right, exactly. Literally that. We’ve literally said that you may never get either more profit or, thanks to the magic of competition, lower prices — because you always have to reinvest more and more. And I don’t want to use the word “safety,” because a certain level of background radiation is unsafe. But when you’re talking about the background radiation that you get on a transatlantic flight, and when you’re talking about a tiny fraction of that, that is not unsafe. And we have not built that in; we’ve built the opposite into our system. So getting nuclear power costs down, to me, is the holy grail of getting a lot of non-London economic activity going again, because energy costs are just so high that if we want green industry to exist, if we want industry to exist full stop, and for it to work via electricity and not via fossil fuels, or to work via heat that’s transferred directly from a reactor not via electricity, nuclear just has to be done. We have to fix it. So I’m pretty excited about that. There are really cool small modular reactors. There are microreactors that are 20 megawatts, and the components can fit on the back of a truck and they can then be assembled. So you really don’t need to do very much assembly on site at all. You’re just putting these shipping-container-sized things together, and it’s like the size of a football field or something. That’s really cool, and that’s really exciting. But that is solving for a really messed-up system, a really broken regulatory system we’ve got. Because the hope is that if you can get regulatory approval for these, then you don’t need to keep approving them at scale. So it’s like a way of dealing with this really kind of mad, mad system we’ve got. Which is great, because I do think that that’s like regulatory arbitrage. I think there’s more to it than that, but regulatory arbitrage is a very legitimate strategy that produces great things. Airbnb and Uber are both great and they effectively or in some ways are regulatory arbitrage. So I’m definitely not diminishing SMRs, and there is a lot of impressive technology around small modular reactors. Rob Wiblin: But the UK needs a lot of stuff very quickly. Sam Bowman: The UK needs a lot of stuff very quickly. And the data centre factor I think may focus our minds very rapidly. If it works out that for a lot of sectors we need very low-latency local data centres for the purposes of inference — not for training models, but for actually just using models — we might need to have a lot of data centre capacity domestically. By no means certain that we do need that, incidentally. Everybody kind of assumes that will be what happens, but I often find myself talking to people from frontier AI companies who are like, “You can probably just buy it from Iceland” or something. Rob Wiblin: I see, yeah. With the geothermal plants. Sam Bowman: Yeah, yeah. But there’s definitely an important world where, just for the purposes of normal economic activity, we need loads of data centres. Certainly from a strategic national security point of view, I think it will be a little bit like not having any domestic energy supply. Rob Wiblin: No country wants to be completely dependent on something absolutely essential from other countries. Sam Bowman: And I think where there is possibly some hope is partly in that there’s so much urgency to this that it could be a couple of years before this becomes one of the big defining economic differences between countries. And also because it’s possible, and increasingly it seems as if very large data centres are going to look to, behind the metre, build their own or procure their own electricity, and not bother working through standard electricity prices on the standard grid. If that happens, then we might be able to say that nuclear, for the purposes of powering a data centre, has its own regulator. Rob Wiblin: That has its own regime. Sam Bowman: That’s got its own regime — again, setting up a new thing rather than trying to fix the old thing. But you know, let’s try to give the whole country the benefit of cheap power, not just one sector. Rob Wiblin: Not just Amazon. What’s another technology that you think is underrated and important? Sam Bowman: I feel like “underrated” is bad, because I feel like… Rob Wiblin: What are you excited about? Sam Bowman: So I have been excited about GLP-1s, like semaglutide, for a few years, for more than most people have been. Rob Wiblin: This is Ozempic, Mounjaro. There’s a whole bunch of different names. Sam Bowman: Yeah, yeah. Mounjaro is actually not semaglutide; it’s a different thing that is just like semaglutide. But they are very, very exciting, because I have a lot of sympathy for people who have obesity or people who are overweight. Dieting is absolutely miserable. It’s one of the worst experiences one can go through. And the fact that anybody can diet, I think, is amazing. But the fact that lots of people cannot diet, and find it so unpleasant and so difficult that they would rather be seriously ill, I think is extremely sad. And I have deep, deep reservations about compulsion, and about making food worse to stop people from getting obese — and I don’t really think it works either. But we’ve solved the problem! It’s unbelievable! For almost everybody — they don’t actually work for everybody, seemingly. But these drugs have been used for I think more than a decade in diabetics, so we have a very long track record of their safety and adverse effects and things like that. And there are a few adverse effects. They definitely have some side effects. But in a way, what’s so exciting about this is that technology has come to a thing that was treated as a social problem. The people who are very interested in obesity and in obesity policy are obsessed with this concept of food environment: that the reason that people are fat is that there are fish and chip shops or chicken shops or McDonald’s everywhere. Or when you go to a supermarket, the supermarket is telling you to buy sweets and stuff like that. And they, in my opinion, get the causation slightly backwards. The reason those things are everywhere is that they’re delicious. Rob Wiblin: Everybody wants them. Sam Bowman: Like, fried chicken is my favourite food. It’s absolutely delicious. I’m really showing how little of a connoisseur of things I am. Fried chicken is absolutely, incredibly delicious. Sweets, candy, chocolate: they are incredibly delicious. And genuinely, a lot of anti-obesity campaigners or public health obesity campaigners genuinely don’t understand markets or economics in the slightest. They genuinely think that there’s a sort of corporate interest in selling people fattening foods, and if there wasn’t a corporate interest… They’re starting from a position of anti-capitalism — they don’t possibly realise it, but they are — and they are then saying the reason that obesity exists is that capitalism has wired itself to give people these high palatability, high calorie foods. And the answer, in their opinion, is either reformulate the food to make it taste disgusting and not have as many calories because you’re using artificial sweeteners, or to have bans on where you can put fast food restaurants, or bans on what you can advertise and so on. I suspect none of that will work, but that’s not the point. Even if it did work, I think it would be a bad thing to do, because those things are tasty and they do taste good. And most people who do not suffer from obesity should not have their enjoyment… Even people who do suffer from obesity should have the option of eating fattening foods and being obese. And semaglutide — Ozempic, Wegavi, Mounjaro — these things have just solved the problem. It’s absolutely astonishing. I think if we weren’t living in the decade of AI, we would be considering ourselves to be living in the decade that we conquered obesity. I know many people who have taken it. It’s been kind of amazing how many people I have spoken to — strangers at parties or events who know I’m a big backer of semaglutide — who say, “By the way, I’m on it.” And it’s like, how is it that we have just spent the last 30 or 40 years wrestling with this problem and then just discovered a cure for it? It’s miraculous. Personally, I think that it should be basically available to anybody. I think people should take it if they want to for cosmetic reasons. I don’t know enough about the physiology of it to know whether the related health effects are independent or just driven by the fact that obesity causes loads of health effects. A lot of studies say it helps heart disease, but does it help with heart disease in people who are not obese, or just via… I don’t know. I don’t know. It does seem to reduce people’s dependence on alcohol. It does seem to reduce people’s smoking. Rob Wiblin: Just general compulsive behaviour. Sam Bowman: Yeah. Drug taking. It’s very, very exciting. And it seems as if we are just on the brink of a much more effective wave of these drugs being developed and released. I mean, everybody listening will know about this stuff already. So I feel bad I’m not coming with a really obscure kind of invention that… Rob Wiblin: Well, something can be highly rated but still underrated or not fully appreciated. Sam Bowman: I think it is. Yeah, I think it’s highly rated but underrated. And I would rather just give you the straight answer about how I am so excited about this than tell you there are other potentially very exciting drugs and things like that — but this is the one that, I think, it works. You will be surprised by how many people are already on it. Rob Wiblin: Yeah, I know some people who rave about it. I mean, this isn’t a medical programme, and there can be side effects. Sam Bowman: And I am absolutely not saying anybody listening should take it personally. Rob Wiblin: But I think if I was overweight, I would definitely take it. If I was at risk of diabetes, I would absolutely take it. Or I would certainly try it. I guess you’re saying for some people perhaps it doesn’t work, a small minority. The way that I conceive of it is that we’ve had lots of technological advances in food to make food much more delicious than it used to be. But human biology hasn’t changed, and that’s like slightly set our appetites out of whack, or I think that’s one of the reasons that people tend to be more overweight than they used to be. And now we’ve had advances in technology on the appetite side, so that now these things can compensate for one another and we can have the delicious food without having such overactive appetites that we make ourselves overweight when we don’t want to be. Sam Bowman: Absolutely. Rob Wiblin: It’s brilliant. It was a technological advance that created some negative side effects that now we have figured out a way of mostly solving. Sam Bowman: Yeah. And just to be clear, there are gigantic market rewards: there’s gigantic profit on the table for palatable low-calorie foods. Rob Wiblin: I guess I like Diet Cokes, so that’s kind of an example. Sam Bowman: I recognise that I am a bit biased in my obsessive hatred of artificial sweeteners, because they taste really bad to me, and I wasn’t brought up with it. Rob Wiblin: Yeah. No, I wasn’t brought up in that. Sam Bowman: I never got the taste. But you know, there is a holy grail in the food world, in food development, of an artificial sweetener that doesn’t taste like an artificial sweetener. And clearly, if you could do that, then you would make bank — provided we give adequate intellectual property protection, which it may be difficult to protect. I don’t know. There are huge, huge incentives. They’re like pretty good ice creams that kind of taste like ice cream, or they taste a bit like the ice cream that’s three or four times more calorific. But I think it is really funny how you have an organisation in the UK called Nesta, which has an endowment of £400 million or more, a really large endowment. It was set up as a state innovation agency and then effectively quasi-privatised. And one of its three main areas of research is obesity, like healthy living. And it’s a little bit like the economists who didn’t predict the financial crisis — which is a very unfair thing; I don’t think most economists’ job is to predict financial crises — but they just write a single word on Ozempic. I think it took them two years to write about Ozempic after Works in Progress had written the first article that we wrote about Ozempic. And their first article was to just say, “This is too expensive. We need to do the things that we want to do” — like sugar-free chocolate and rules requiring supermarkets to reduce their customers’ intake of food. It’s a really weird situation. Rob Wiblin: When it first came out, there were just journalists falling over themselves to find some side effect to complain about, and say how terrible it is that people are taking Ozempic. I couldn’t believe that this is still going on. There was an article in the Financial Times recently about how it looked like obesity and overweight in the US was actually finally, after going up for decades, it’s finally turned around and started going down. Why? GLP-1 agonists for sure. But I was reading the comments, and people are like, “It’s terrible that people are taking Ozempic. What we need to do is change the food environment.” Like ordinary Financial Times readers were just saying it’s outrageous that we’re solving the problem this way; we have to solve it a different way. And also kind of judgement and condemnation of people who don’t have the self-control to avoid eating, when it’s like evolution for billions of years literally has been trying to get all organisms to take in lots of calories, because that was the main way to die. Of course it’s extremely difficult for many people, most people, to not overeat, when they’re just surrounded by absolutely delicious food. Sam Bowman: I do wonder, in the same way that I hate artificial sweeteners, if some people just enjoy food less. And I don’t mean that as a dig against them — or maybe I do, slightly — but I just find it kind of bizarre that they don’t see that food is as much about pleasure as it is about sustaining your life. Like, food to me is everything, in terms of when I want to spend time with somebody I like, I want to have food with them; when I’m feeling down and I want to cheer myself up, I’ll make myself buffalo wings. It’s like food is so much a part of the richness of my life that I can only see the people who want to fight obesity… I’m not obese. This sounds selfish, but it feels pretty awful to say to the 70%+ of the population that does not have a significant problem here, “We’re going to make your life significantly worse because this much smaller fraction of the population has this problem.” That feels very badly targeted. I would guess probably inefficiently targeted, and also just an affront to civilisation to say, “We’re going to make the world worse in order to help this small number of people — when there is a way to help those people in a targeted way that doesn’t make the world worse for everybody else.” Rob Wiblin: So there’s all these studies coming out suggesting that these GLP-1 agonists are not only good for obesity, and maybe not only good for cardiovascular health or diabetes through people losing weight — but also maybe it’s good for arthritis, maybe it’s good for preventing dementia; it seems maybe it’s good for extending people’s lives in ways that we have not even yet identified. Do you believe these results, or do you think that it’s possible that there’s some distortion in the method here? That perhaps there’s a file-drawer effect, or people are biased towards wanting to find interesting positive results about something? Sam Bowman: I’m not in any position to say. I am very interested to know whether there are independent relationships going on. It is like obesity is to health what housing is to economics? Like it’s this sort of overweight theory of everything. Yeah, like I totally believe that. And it’s miserable to be obese, and I have so much sympathy for people who are obese. It’s really awful for most people. Some of them probably don’t mind. But I have so much sympathy for them. But I cannot accept the mindset that we have to make the world worse for everybody else in such an important way to try to help obese people. And thank god these GLP-1s have come along and avoided us having to have that kind of discussion or even think about that. Clearly there is some tradeoff. I would be willing to get somebody to flick my ear once a year if we could cure people from cancer or something like that. You know, I’m not making any kind of like hard claim about tradeoffs here. Rob Wiblin: Whether it would be a net benefit or not. It’s not going to happen. We’re not going to ban delicious food. I think no place has come even close to that. Sam Bowman: I think that so many people in the Western world are really uncomfortable with just saying that. There are some people who, for one reason or another, are not able to look after themselves — adults who can’t look after themselves in this way. They’re not people with severe mental health problems; they’re not people who could be institutionalised if we did that at scale. But really, just the normal kind of liberal principles that we’ve got, they kind of break down for some people. And I think that it’s very challenging, if you are not willing to say that, to solve any of those problems — because you end up saying that everybody should be affected by this. You know, you saw a study recently, I’m sure, that showed that sports gambling being legalised in the US eroded a pretty large portion of people’s savings? Rob Wiblin: It was leading to a significant increase in bankruptcies, I think. Sam Bowman: Household savings rates for a pretty large group of people — I think like 9% or 10% — their savings just collapsed. Really tragic, horrible. Gambling is such a horrible addiction if you cannot manage it, if you cannot do gambling. Now, in that case, gambling feels so catastrophic if you have a gambling problem, and feels so trivial if you like gambling, that I can very much see the argument for not having sports gambling. But to me, a much more parsimonious solution would be to say that you have to take a difficult test: you have to show that you are actually capable of understanding probability, you’re capable of understanding the risks — that you get that you can’t game the system and that you are losing on average when you do this. People who can do that, why not just say they can gamble and protect everybody else from themselves, while allowing people who actually don’t need protection to kind of opt out of that? That feels, I think, to a lot of people like a very unfair system. But that feels to me like a lot less unfair than saying that we need to pass a law that applies to everybody, including people it doesn’t need to apply to, in order to protect an unfortunate, in this case, minority of people from themselves. Rob Wiblin: Unfortunately we’ve gotten so into all these topics around policy reform and nuclear and so on that we’ve run out of time to talk all that much about progress studies. But I guess I’ll put one thing to you: This progress studies ecosystem that has been growing over the last five or 10 years, it seems like — despite the fact that it’s quite online and is very debating, it has an agenda that it’s pushing or at least promoting — it hasn’t collapsed into just a totally oversimplistic perspective on things. Progress studies folks will say technology is generally good, but it’s not good in absolutely every instance. Technology in general is the way that we’re going to make the world better, but that doesn’t mean that there’s no point looking at potential downsides or trying to reduce the downsides while we harvest the upsides. And there could be specific cases that deserve extra scrutiny, because they have unusually large effects. Of course, artificial general intelligence or superintelligence is key to me, but I think many people in progress studies look askance at gain-of-function research, for example, and some other biotechnologies that could be negative. How is it that it hasn’t collapsed into an oversimplistic mantra that technology is always good in the way that perhaps some other groups would be inclined to do? Sam Bowman: That’s a good question. I think that the intellectual leaders of progress studies — Tyler Cowen was one of the two coauthors of the piece that originally launched it, along with Patrick Collison — a lot of people look to them for cues about disposition and stuff like that. While I sometimes worry that progress studies might end up being felt to be quite simplistic — you know, we like good things, but we like them fast — I think that in practice, hopefully it ends up being quite thoughtful and quite kind of granular. All of those things lend themselves to a certain type of person that cares about detail, cares about pragmatism, and cares about stuff actually working. You know, I just went to this really good progress studies conference run by Roots of Progress in Berkeley a couple of months ago, and I think you would have been seen as quite cringe and quite embarrassing if you were there just tubthumping, like, “Technology is great; it’s going to solve all of our problems.” I actually think of progress studies myself as being about finding high-leverage reforms and high-leverage improvements you can make — as it relates to policy, at least. I think that there’s this common idea that a lot of politics sort of cancels each other out: a lot of politics is this sort of trench warfare or tug-of-war where there is an equal and opposite side to anything you’re doing. And I would say that the progress studies view is that actually, often pulling the rope sideways gets you really significant improvements, and focusing on areas that do not have these equal and opposite interest groups might be a way of getting significant improvements. And the Institute for Progress, which is run by Alec Stapp and Caleb Watney in DC, is really good. They’ve done a lot of work with the NSF, the National Science Foundation, to kind of change how science grants are done and experiment with how money is given to researchers and measure whether it improves things or not. And Open Philanthropy does a lot of work like this — so this is by no means an exclusively progress studies thing. But this idea that we are very far from the potential that we could have right now, part of that is by doing technological and scientific research better, but part of it is just generally improving the way we are doing decision making and improving the institutions that we’ve got. So even though housing is a fairly non-technology-heavy policy area, I think of it as being quite progress studies related because it’s about, could we get really dramatic improvements to the world via fairly small, simple, hopefully quite elegant solutions? And I think an interest in science for that. I wouldn’t say this is a general progress studies interest, but I’m very concerned about falling birthrates. But I’m also concerned about compelling anybody to have children, or compelling anybody to live a life that they don’t want to live. I think the good news is that most people would like to have more children than they end up having. The bad news is that we don’t really know how to do that. It isn’t a simple case of “just make housing cheaper” or “just give them money” or blah, blah — although all of those things would help. I think that a good progress-studies-informed approach thinks across a lot of different margins: housing, maybe cash transfers, also technology. What could we do to make it easier for women to have children later, for example? To me, that’s progress studies at its best. When you get that quite pragmatic, quite broad-based, holistic… It is very neartermist. I love effective altruists, but I think that the fundamental difference between effective altruists… I think it’s often a good question to ask ourselves: Effective altruists think very hard about doing the most good they can in the world; why aren’t they doing the thing that you’re doing? And I think one difference is progress studies is much more neartermist. Not exclusively, and there is some concern about existential risks and so on, although you can be pretty concerned about x-risk as a neartermist. Rob Wiblin: On short-term grounds. Absolutely. Sam Bowman: And I think probably also believes that the upsides from good policy reform are very, very high — and potentially a lot higher than even many effective altruists believe, in areas that effective altruists often don’t think about, I would say. So I think it’s so far so good, but I definitely think it’s incumbent on anybody who is in this world — which Works in Progress is part of the progress studies world; I wouldn’t say it is exclusively about progress studies or anything — but part of the vocation that you’ve got is to try to maintain high intellectual rigour, high standards of what you’re doing, and hopefully not attract people who are just looking for a slogan or a thing to put in their Twitter bio or an emoji or something like that. Rob Wiblin: Yeah. I’ve got so many followup questions I would love to ask you, but I’ve been enjoying this so much we’ve run absolutely up against the end of our booking at this place. They’re about to kick us out. So I guess I’ll just have to keep my appetite whetted for a second interview that we can do at some point down the line. My guest today has been Sam Bowman. Thanks so much for coming on The 80,000 Hours Podcast, Sam. Sam Bowman: Thanks for having me. This has been great.