Addressing Gentrification, Displacement, and Housing Shortages with Economics
Many locus points of innovation and growth in the United States are heavily regulated and controlled in regards to spatial expansion. As economic prosperity and high earning, highly skilled jobs increase in an urban area, so too do zoning regulations controlling types and levels of growth. These regulations are often associated with a sense by residents that an area has achieved a tipping point where the balance between density and accompanying amenities is in danger of producing urban congestion, with limited additional amenities. These regulations have a negative externality by concentrating social and economic capital while limiting access to the opportunities in these urban areas – especially to supporting labor markets, such as service industries. Can you imagine a future where the economic and social distribution of benefits from growth industries – technology, health care services, biotechnology, pharmaceuticals – are shared across a diverse ecosystem of supporting actors?
Devin Bunten, Assistant Professor of Urban Economics and Housing, has a clear vision of this future and she sees housing as a pathway to achieving a more equitable outcome for our innovation hubs. Her research uses economic theory and tools to study a range of urban topics, including gentrification and neighborhood change, restrictive zoning, and urban economic history. Prior to joining the faculty at DUSP, Bunten was an economist at the Federal Reserve Board in Washington, DC and helped co-found the group Abundant Housing Los Angeles, which promotes more housing production in the LA region.
Q1: What role do you see economics playing in urban planning and how can it help promote the values of equity, fairness, and inclusion?
Economics has a lot of useful tools, including empirical methods that have helped us to better understand the world. However, the piece that really drew me in was how economics at its best thinks about theory. There are a few critical pieces for a theory: what do people want, and what constraints do they face? And then, how do these pieces interact to produce the outcomes we see in the world? I find the dual focus on wants and constraints to be clarifying in many circumstances. I asked my class recently how many of them thought they were gentrifiers, or would be soon, and a large majority said yes. But then I asked how many of them want to gentrify--and of course no one did! But they feel constrained by the cost of housing to act in a way they don’t enjoy. I think economics offers a useful way of framing the core of issues in straightforward ways. Of course, the real world offers many complications--but starting from a simple model and layering on nuance is, for me, an easier task than the alternative.
Q2: With growing concentrations of urban populations, gentrification is widely discussed as intensifying and driving displacement in our cities. Would you discuss your model for measuring the impacts of gentrification?
There’s a clear sense that gentrification is a key driver of displacing people from their neighborhoods, but it has been hard to see this in the quantitative data. We’d really like to see it in a quantitative setting, as that would enable us to evaluate the effects of policy alternatives. If we grant that gentrification displaces people at a higher rate, then why haven’t we seen it? Three reasons: it’s hard to define gentrification, it’s hard to define displacement, or it’s hard to define which people are at risk of displacement. My paper, joint with Shifrah Aron-Dine, suggests that it’s primarily the last point: it’s hard to separate early-stage gentrifiers from the gentrified!
Suppose that a typical gentrifier is relatively young, and maybe more likely to be engaged in artistic pursuits. If we look at the data, this person probably doesn’t have particularly high current income and wealth. They may not have completed their education yet. They might not have great credit, and they likely rent. If we want to assess outcomes in a large-scale dataset, it’s going to be really hard to separate this gentrifier from the people they are displacing. And if we get it wrong and classify this person as “vulnerable to displacement”, then we are going to get our estimate very wrong: if a gentrifier is moving in on purpose, because they like the direction of the neighborhood change, they will likely stick around much longer than average. But then, we will estimate that gentrification lowers displacement!
Shifrah and I take a new approach. We say, if you lived in a neighborhood for at least 2 years before we see signs of gentrification, then: you aren’t a gentrifier. You’re at risk of displacement. You were living in a low-income, relatively low-house price neighborhood, and it later started to gentrify. With this approach, we then ask: once your neighborhood starts to gentrify, are you more likely to move out than you were before? And we can compare your likelihood of moving to other folks living in neighborhoods that haven’t yet gentrified. We see a displacement effect very clearly: after gentrification begins, residents with roots in the neighborhood are about 3.9% likelier to move out than residents in neighborhoods that haven’t yet started to gentrify.
The trick to the paper is that it’s really helpful to have a definition of gentrification that we can check annually. The usual approach is to use Census-year changes, but those are only available every 10 years--or maybe every five years, if we use the American Community Survey. But most people who rent in low-income neighborhoods move more often than that, so it becomes hard to know who was already there! Our approach is to use a new definition based on neighborhood house prices and incomes. Each of these pieces is available every year down to the neighborhood level. Looking at the data, it appears that in gentrifying neighborhoods, house prices rise ahead of average incomes. This gives us our indicator: low-income neighborhoods are starting to gentrify the year when house prices drift above incomes by some margin. Why do we see this happen? Well, a house price is an asset price, and so it reflects the current conditions of the house--as well as expectations about future changes. In a gentrifying place, those expectations are quite large, and so buyers are willing to “overpay” a bit in advance of rising prices.
Now that we’ve identified a clear effect, we hope to understand more about the context: do certain policies seem to ameliorate the effects? Do other policies exacerbate them? We hope that this approach can eventually help evaluate the effectiveness and inform anti-gentrification policy at a wide scale.
Q3: Living in Boston and Cambridge, with their associated cost of living, it is easy to see cities continuing to act as concentrators of wealth and opportunity -- with increasing economic barriers for entry. Could you speak to Tokyo and Toronto as models of the future you see successful housing policy unlocking?
Toronto has been in the news for a few years as the site of a possible housing price bubble--price growth as high as 10%. And yet, if you look at Toronto real estate listings, you can find brand-new condos listed for sale below $300,000 (US dollars). They might not be the hippest neighborhoods in all cases, and the cheapest condos are usually relatively small--but they exist, and often along subway or light rail lines. If you look for that price range in Boston, the first things you’ll find are listings for parking spaces, and the homes that you find are 300 square foot, 100-year-old basement units in Allston or foreclosure properties where the price is simply the starting point for the auction. Cambridge and Somerville currently have no Zillow listings below $300,000. Tokyo has hundreds of listings below $300,000, including new construction and units in close-in and hip neighborhoods. If you’re willing to downsize, you can even find a home for under $100,000 built in the 1970s or 1980s.
So what’s going on here? In a word, Tokyo doesn’t have a shortage. Boston does, and it’s self-inflicted. For every 1,000 residents, Boston permits about 2.6 new homes per year. For Toronto the number is 6.0, and for Tokyo it’s 12.2. In eight of the last 20 years, Somerville has permitted zero new units. Even with the Assembly Square project, this number in recent years amounts to about 4.1 units per thousand people. There is no building boom, and so of course the wealthy win out in the fight over the scraps of housing left over.
In terms of policy, one big difference between US cities and Tokyo--and until recently, Toronto--is the branch of government responsible for regulating whether you can build new houses. In the US, this is a strictly local affair, meaning the folks with the most to lose from new construction--noise, dirt, and traffic--have the most power to stop it. In Tokyo, policies controlling new housing are a national affair, meaning that some sway is given to the folks who stand to benefit: most especially, people who would otherwise be priced out of the market. (Toronto was an in-between case: ostensible local control, but a province-level review board that was willing to overrule local concerns. This setup was changed in recent years, and time will tell whether Toronto is able to maintain the affordability it has).
What does this mean for Boston? If we want to get past the housing shortage and promote access to the region, we will have to build more housing. Looking abroad, the best hope for this appears to be a state-level campaign aimed at making it legal to add triple deckers to single-family neighborhoods, and 7-10 story buildings (or more!) to existing triple-decker neighborhoods. These sort of buildings dot our region today, but we need more. We have the technology, all we have to do now is care to use it.