Edward Glaeser

EF: In your new book, Survival of the City, you argue that technological changes in the postwar period were mostly "centrifugal," leading people and companies to move away from urban cores, while technological changes in the 21st century have been "centripetal," leading to more concentration in urban cores. Please explain.

Glaeser: I see urban growth as almost uniformly a dance between technologies that pull us together and ones that push us apart.

Technologies of the 19th century, like the skyscraper — which is really the combination of a steel frame and an elevator — the streetcar, the steam engine, all of these things enabled the growth of 19th century cities. They brought people together. This was a centripetal age.

In the mid-20th century, we had technologies that were major jumps forward in transportation cost. In transportation technology, like the car, and in technology for transporting ideas and entertainment — television and radio — these were centrifugal forces that basically flattened the Earth and made it easier to live in far-flung suburbs or even rural areas.

Those centrifugal technologies were the backdrop for New York's decline during the 1970s. They were the backdrop for the exodus of people from dense cities that had been built around streetcars and subways and to suburbs that were built around the car.

But then in the late 20th century and early 21st century, the tides turned again. And it was somewhat surprising. With this shift came a vision in which the rise in these forms of information technology would lead the knowledge workers that still existed in cities to follow the path of industrial workers and essentially deurbanize. Knowledge workers would work remotely from electronic cottages.

But for most of the last 40 years, that hasn't been correct at all. That hasn't been happening in cities. We've started to see the electronic cottages become a force during the pandemic, and suburbanization has continued, but downtowns are vastly stronger than they were in the 1980s. And I think the primary reason is that globalization and new technologies have radically increased the returns to being smart, and we are a social species that gets smart by being around other smart people. That's why people are willing to pay so much to be in the heart of Silicon Valley and why they're willing to pay so much for downtown real estate in Chicago or New York or London.

EF: What does the future of small towns look like after the pandemic?

Glaeser: I think that's going to be a tale of two towns. If you are a small town like a college town, a place with high levels of amenities and beautiful scenery where rich people want to go, I think that the combination of the ability to do work remotely and perhaps some enduring pandemic fears means that you are as strong as you've ever been, if not more so. These places are poised to benefit.

Take your Silicon Valley startup with 15 smart, hungry young people. Do we truly think in five years these people are just going to be Zooming it in from their suburban bedrooms? That sounds totally implausible to me. That sounds like a totally different work model that will lack all the energy and high quality in-person connections you get from being in the same room as one another.

But on the other hand, are these 15 people going to decide, "Well we all love skiing, we're tired of paying Silicon Valley prices, should we relocate to Vail?" Or say, "We don't want to pay taxes, let's relocate to Austin." Or, "We want better surfing, let's relocate to Honolulu." That feels entirely plausible to me. The technology supports the mobility en masse of these groups to some different area. Places they're most likely to relocate to are high-amenity places that will appeal to them along one of these dimensions.

These would be probably the best index right now of whether or not a place is likely to benefit: Among small towns, is it a skilled place already? Prior to COVID-19, did it do a good job of attracting large numbers of college graduates or people who had advanced degrees?

On the other hand, if you're talking about small towns in relatively low-amenity places, places that are low density, farmland, low levels of education, these places have been declining for decades, and I see little reason why the decline would be reversed anytime soon.

EF: Place-based policies — that is, policies aimed at improving specific areas — are often criticized by economists. You've argued that they may be justified for some areas of the American heartland where men have low labor force participation. Why is that?

Glaeser: I still maintain my traditional aversion to place-based policies. I have not completely gone away from it.

But I do think that the persistence of prime-age male joblessness is worrisome. If you look at the relationship in the Public Use Microdata Areas between joblessness in 2010 and joblessness in 1980, the correlation is over 80 percent. We have an enduring level of local economic dysfunction. I don't think that this makes the case for large-scale redistribution in these areas. I think there are many good reasons for being wary of that, but at the same time, I think it's necessary to think about policies that are more place-specific.

Take housing. You really don't need to subsidize the production of low-income housing in most of Texas, because they have an unfettered market that does a great job of providing lots of low-cost housing to middle-income residents. If you have Detroit, you don't want to produce more low-cost housing, because they've got an abundance of low-cost housing there. But on the other hand, there's probably a good case for doing something about low-cost housing in San Francisco or New York or Boston.

That suggests to me, at least, that you want policies like the low-income housing tax credit that subsidizes new housing construction. You want that to be spatially limited. You want it to go in areas where there's a genuine dearth of low-income housing. At the same time, you could have more housing vouchers in the areas where housing supply is elastic. You can have the right policy for the right place, which is something that America has traditionally found very difficult to do. But it's just basic economics.

Likewise, if you're going to have a large-scale employment subsidy, it makes more sense to put that subsidy into places where it will actually encourage the most employment. That's much more likely in places that start with high levels of nonemployment, like the eastern heartland.

Now, my own preference is that you make it feel relatively budget neutral and figure out other things to deliver more of to other areas. But I think we have to realize that the eastern heartland, particularly, has very specific circumstances that need distinct policies. So, when we're discussing a $15 national minimum wage, that's fine in Seattle; we've had a whole bunch of papers on this over the past five or six years that show it won't hurt employment much. If you have a $15 minimum wage in West Virginia, you're risking much more of an increased joblessness problem there because you're starting at a very different place.

EF: What is your view of opportunity zones as a policy in this regard?

Glaeser: We have a paper on this that does try and compare them with truly comparable areas. We're focused mainly on the effect on housing prices, which we thought would be one measure of whether things are getting better in the area. We find almost no effect from these zones on prices.

I think in general that the needs of these places are really about human capital problems rather than a lack of entrepreneurship. But the opportunity zone structures are very oriented toward more investment in physical capital, including housing and apartments and such, which does not particularly seem like the right thing to do for these disadvantaged areas. When I think about what I would like to see in a zone, it's much more focused on the human capital than the physical side.

EF: Why have major investing institutions recently started buying single-family homes in large numbers?

Glaeser: Traditionally, single-family homes were overwhelmingly owner-occupied in the U.S. More than 85 percent, I think, of homes were owner-occupied. The usual view of the housing economics community was that the agency problems involved in renting them out were huge. There are estimates that suggest that renting out for a year involves a 1 percent decline in the value of the house, or something like that, because the renter just doesn't treat it properly.

By contrast, traditionally more than 85 percent of multi-family housing was rented, at least once you get to over five stories. It's much easier to manage a multi-unit building when you have one owner. One roof, one owner, because otherwise you've got the problems of coordination of the condo association or the co-op board, which can be more fractious.

So those were the things, I think, that were responsible for tying ownership type and structure type so closely together. We are starting to see that break down, which is quite interesting. I don't know if these buyers have fully internalized their difficulties with the maintenance that goes into rental houses as a long-run issue. Or if technology has changed in such a way that they think that they can actually solve that agency problem and that they can figure out ways to deal with the maintenance costs in some efficient fashion.

I'm happy to see an emergence of a healthy rental market in single-family detached housing, but I'm keenly aware of the limitations and difficulties of doing that. So, we'll have to see how this plays out. I can't help thinking some part of it just has to be that investors are simply searching for new investment products.

EF: In Survival of the City, you write about development restrictions as favoring insiders, by which you mean homeowners. Homeowners might say, "Don't we want local government to be responsive to local interests?" Is this a good thing as well as a bad thing?

Glaeser: I certainly don't want local government to be nonresponsive to local interests. The problem is that local majorities are not necessarily going to take into account everyone. They won't necessarily take into account minorities, and they certainly won't take into account people who are not in the locality right now.

As always in the case of democracy, we want something that empowers the majority, but also slightly restrains it in different ways. We want the Bill of Rights as well as the grants of power in the Constitution. And in the case of localities, I think that the idea of placing some restriction on what localities can do is quite reasonable to me.

For example, in the area of housing, you have Massachusetts Chapter 40B, which basically enables developers to work around local zoning if the locality has almost no affordable housing and if the developer is providing some low-cost units to buyers. They're able to take advantage of the state process rather than the local process for getting approval. So, you're not getting rid of all local power over zoning. But if your local power means that you produce nothing that's affordable, we're going to maintain a way for the builders of affordable housing to basically bypass you. That sort of hybrid model seems like a good one to me.

The new California law, SB 9, on speedy permitting of two-unit buildings will also be a state restriction on local power. This was the first time in many years we've had a victory for the "Yes In My Back Yard" movement — the YIMBYs. It may be a small victory, but you know, when I started thinking about this stuff in 2001, there was nothing. There was no popular movement of any form to reduce the local straitjacket on building.

EF: The famous 1961 book by Jane Jacobs, The Death and Life of Great American Cities, has been influential over the years in people's thinking about urban development. What was she right about, and what was she wrong about?

Glaeser: It's a wonderful, wonderful book. It's so wise in its understanding of urban neighborhoods and the street life and the ability of neighborhoods to just work and the dangers of trying to engage in planning. All of that stuff is fantastic.

Where she kind of screws up is when she gets into a discussion of how cities need old buildings. It's not that she's totally wrong, not at all. It's great for cities to have some form of inexpensive space. And in most cities, that inexpensive space comes from older buildings that have not yet been upgraded. The way that affordability is supposed to work is that you build new buildings, which are usually not that cheap, and that reduces the pressure to gentrify old buildings. So that's where affordability comes from.

But that led her to the policy view that you actually want to prevent tearing down old buildings and replacing them with new buildings. Now, if the two buildings are exactly the same size, I agree that replacing the old building with the new building will not promote affordability.

But that's not typically what developers want to do. Typically, they want to take a short building and replace it with a tall building. And if you stop that process, as she was so instrumental in doing in the Greenwich Village historic preservation district — which was part of a great popular wave of opposition to rebuilding New York — you aren't promoting affordability, you are freezing a neighborhood in amber. In the case of Greenwich Village, you have created a situation in which townhouses start, at least before the pandemic, at over $7 million. That's no recipe for affordable urban space that scrappy startups and young families can buy, which is how she was thinking about it.

That was, I think, the thing that she got most wrong. She was right in many of her ground-level observations, but economics is really helpful for thinking through the long-run implications of a policy intervention.

EF: You wrote in your book that the 1970s were a catastrophic decade for much of urban America. Why were the 1970s catastrophic for cities?

Glaeser: It's that combination of the centrifugal technologies — the fact that container ships and highways made it easy for factories to relocate to places where land and labor were cheaper, and highways made it possible for wealthier urbanites to leave for suburbs.

That process collided with the dreams of progressive mayors, like New York's John Lindsay or Detroit's Jerome Cavanagh, who were trying to fix the very visible woes of cities. What that meant was that just at the time cities were asking businesses to pay more, asking the rich to pay more, it was becoming easier for the rich and for businesses to get out. And that's exactly what happened.

On top of that, of course, there was an inability to deal with the crime problem. That's something that I worry about today. There is a totally understandable desire to try and deal with the terrible inequities of urban life. There's a totally understandable desire to want to do something about policing and the experience that many people have with the police. But if the changes that occur either end up targeting the taxpaying members of the urban community or end up leading to a significant deterioration in the quality of life — for example, with an increase in crime rates — that risks replaying the 1970s.

One of the reasons why I wrote the book was to emphasize that we had paths to try to reduce those inequities and could make those cities not just more functional but more humane. But at the same time, cities have to do that in a way that respects the need to continue to attract the taxpayers who ultimately fund things.

EF: You wrote that court-ordered school busing and the reaction to it played a role. In what way?

Glaeser: The way that busing got implemented was the court requirement getting rid of segregation in cities. But there was also a court ruling saying that you could not force desegregation across city boundaries.

For example, the Supreme Court decision in Milliken v. Bradley in 1974 said essentially that federal courts could not require desegregation across school districts. What that meant was that if you wanted to avoid busing, either for racist reasons or for some other reasons, you could get that only by leaving the school districts — by leaving the city. And so, for thousands and thousands of parents, that's when they moved, sometimes just outside the city's school district.

If you had metropolitan-area-level desegregation efforts, that would not have created the same incentive. Or if you had something that was more like a charter school system or like a voucher school system. Anything that breaks the link between where you live and where you go to school would've been less harmful for cities. But as it was, this was yet another huge incentive for parents to get out. And a harmful one.

And ex post, if you look at the Opportunity Atlas data created by my colleague Raj Chetty and his co-authors, there was a clear and discernable break in upward mobility at the border of central city school districts across America. Opportunity jumps up if you're just outside the central city school districts as opposed to just inside it. This is a cohort that was born between 1978 and 1983, and you can see it in their adult earnings. They are also significantly less likely to be incarcerated as an adult.

EF: Winston Churchill said in 1943, "We shape our buildings and afterwards our buildings shape us." Was he right? And if he was, what are the trade-offs in regulating architecture?

Glaeser: Yes, he was right. It's one of the greatest of all lines about architecture, absolutely.

The regulation of architecture is probably the most straightforward element of this in some sense. I have a paper with Nikhil Naik, who is a computer scientist, where we look at the connection between the appearance of your neighbor's home and your housing value. We find significant effects, and the effects are much stronger if you are on the same street as opposed to being, say, half a block away outside of the sightline. People value having nicer homes to look at, and that creates at least some scope for benefits from regulating architecture.

On the other hand, I believe people are willing to pay for architecture. Moreover, as the son of an architectural historian, I continue to be skeptical about the abilities of states to regulate architecture well. [Glaeser's father, Ludwig Glaeser, was a curator in architecture and design at the Museum of Modern Art.]

There are special cases. The vista of a city or the elegance or the magic of, for example, some streets of Barcelona or the Place des Vosges in Paris — these are magical urban spaces. And certainly, I am OK with protecting those pieces of architecture that are widely accepted to be part of the patrimony of the city.

But when it comes to building new buildings, we have to be aware that very rarely does great architecture happen by committee. And very rarely will a zoning board be the best judge of what design is going to create joy in people for decades to come. It doesn't take much to convince me that there's a market failure in many cases, that cities are chock full of externalities, but saying that the government is actually going to make it better is a much heavier lift. I'm not sure it's here in this case.

In terms of how buildings shape us, I certainly believe that the structure of our offices shapes who we interact with and end up shaping our lives. There is evidence to show that people who are physically proximate to each other end up influencing each other. I think about, say, Bruce Sacerdote's work at Dartmouth on randomized roommates. There's an older MIT study that suggests that if you are randomly put closer to each other, you are more likely to form friendships. Plenty of evidence suggests that human proximity shapes interactions. None of this is surprising, but it is powerful and reminds us that physical proximity continues to be highly important.

For those younger economists who are reading this, I think that would be a great place to think about using, say, randomized controlled trials or something else to figure out to what extent Churchill is correct when it comes to day-to-day office life and street life and how our buildings are shaping us.