Anya Martin explains why most strategies to tackle gentrification fail, and how building more homes might work, for Issue 15. Read it on our website here.
The neighborhood of Edgehill is just south of downtown Nashville, Tennessee. Established by Union forces as a camp for fugitive slaves during the American Civil War, further waves of African Americans arrived during the Great Northward Migration in the 1900s. But since that early period of the twentieth century, its status as a historic and predominantly black neighborhood began to change. Demand for housing there began to rise due to economic growth and new jobs in surrounding neighborhoods like Music Row, where Elvis Presley and Dolly Parton recorded their hits, from the 1950s onward, and the Gulch, an upscale area that has been heavily redeveloped since the 2000s.
In an attempt to preserve its character, Edgehill was mostly zoned for single- and two-family use in 1974, and then to mostly single-family use in 2000, meaning homes cannot be easily redeveloped into apartment blocks, and in 2018 the city designated it as a historic conservation area.
As an attempt to preserve the architectural character of the neighborhood, it was a failure: over the course of the late twentieth century and early twenty-first, the original homes were replaced by larger, single-family zoning-compliant homes. If the intent was to prevent the displacement of the local community, to say it failed to work is an understatement: the number of black residents fell by 80 percent in just 20 years. Local news described it as a perfect storm of gentrification ‘[sweeping] away affordability for half of Nashville’s residents’.
Gentrification – broadly, existing lower-income residents being displaced from their neighborhood by incoming higher-income residents – is widely recognized as a serious problem. It inspires more activist engagement, protest, and artistic depiction than virtually any other policy issue in housing. Villainous gentrification is a recurring trope in TV and film. Lower-income residents dislike rapid, uncontrolled change in their neighborhoods just as much as the wealthy, but have generally had much less success at blocking it.
Most attempts to slow or stop gentrification have failed as resoundingly as Edgehill’s. Migration to, between, and within cities as economic opportunities shift is inevitable and almost a human universal. Attempting to stop people from seeking better lives in a new area generally does not work, but we can avoid most of the harms of gentrification by making space for them in the places they want to go. Housing supply is not a panacea, but removing constraints on housing supply growth can make inevitable migrations benefit communities, rather than harming them. And in some cases, like London’s estate regeneration programs, they have done it in ways that benefit and enjoy the enthusiastic support of the existing community.
The effects of rent controls
One common approach of anti-gentrification campaigners is to lobby against new housing development, on the assumption that new development attracts higher-income households to the area, increases rents, and pushes out existing residents. This was the approach taken by Edgehill, which catastrophically failed on its own terms.
A similar example is the Haight-Ashbury neighborhood in San Francisco, California. This formerly 40 percent–black community was downzoned in the 1970s. Advocates dismissed concerns from black residents about the impact the downzoning would have on housing affordability. Thanks in no small part to the downzoning, which essentially eliminated new multifamily apartment blocks in which black residents were more likely to live, the Haight-Ashbury neighborhood is now just five percent black.
These approaches failed because new development is not the cause of gentrification, but a way of absorbing new residents. The cause for Haight-Ashbury was rising demand for homes in the area, due to San Francisco’s booming economy, and since this demand was not met with enough new housing supply to match, rents rose, houses were bought by new owners, and the area gentrified.
Some communities have tried rent controls as another tool to stop gentrification. In a narrow sense, rent controls may work: beneficiaries of rent control often have materially lower rents than they otherwise would, and are less likely to move. This could reduce unwanted displacement, but it could also mean people staying in a property that no longer otherwise meets their needs because they don’t want to lose the discount.
A wealth of academic and historical evidence shows that rent controls also reduce apartment quality over time, worsening a neighborhood for its residents, especially ‘first-generation’ rent controls, which apply perpetually to a given unit. France had extreme rent control between 1914 and 1948, after which the system was softened substantially. Real rents fell 60 percent between 1914 and 1948 while construction costs rose 40 percent. Famously, 63 percent of French dwellings still lacked running water by 1946, and only six percent had indoor toilets in 1951.
In New York, the controls currently in place are ‘second-generation’ rent controls, which allow rents to change between tenancies, and are therefore less likely to disincentivize maintenance. Still, rent-controlled low-rise apartment buildings are nine percent more likely to be in a dilapidated or deteriorating condition. In addition to reducing the incentive for landlords to invest in them, rent controls push houses away from the long-term rental market and into the ownership market or the short-term rental market, meaning fewer rental homes appropriate to those on lower incomes.
San Francisco’s system of rent control, which is also second generation, may have actually accelerated the gentrification of Haight-Ashbury as rental supply dwindled and housing supply shifted to less affordable types of housing like condominiums and large, high-end single-family homes. In some cases, finding a new property to rent becomes almost impossible as no one is willing to leave their property, as Stockholm’s 20-year waiting list for homes in heavily rent-controlled areas shows. This leads to black markets for those who can’t wait, where tenants end up paying the real market rate but without any of the legal protections.
In other words, rent controls work for some renters in the short term at the expense of nearly all of them in the long term, and they poorly address the symptom while allowing the cause – a mismatch between demand and supply – to get worse. In theory, one could try to design mechanisms to counter these damaging side effects, but the long history of failed attempts suggests that this is much easier said than done.
Another common strategy is to block existing business premises from changing use, or to subsidize the businesses to give them advantages over newcomers. The logic is that existing local businesses are run by and for existing local residents, that competition from newcomers is likely to shutter some of them, and that preventing this will help existing local residents. Empirically, a 2023 paper by Ed Glaeser, Michael Luca, and Erica Moszkowski found that areas that are gentrifying experience slightly higher rates of both business closures and new openings than areas that are not. But overall, it found that the result was more businesses and stores overall, without finding evidence that the type or price point of stores changed.
The cliché image of gentrification might be a new wine bar, likely of little direct appeal to very low-income families. But this ignores that in many cases the changes are more amenable, as in the example of a new supermarket, established after new housing was built nearby, being welcomed by open arms by locals. As a general rule, more businesses means consumer benefits (in terms of more choice in where to shop) and – even if it is a wine bar – employment options for low-income residents, and greater local taxation with which to fund better local services. While necessarily true that competition will always result in winners and losers for business owners, which includes a small minority of locals, the benefits of local economic growth can be an important counterbalancing feature for the bulk of existing residents.
Another approach is blaming and shaming gentrifiers themselves. Ideologically, it can feel quite successful – how many times have you met people who recently moved to an ‘up and coming’ neighborhood join in with denunciations of other recent in-movers as gentrifiers or ‘soulless’ office workers ruining the local character? But outside of violent destruction and riots, these strategies are rarely very successful at stopping waves of incomers, because those people are themselves responding to the need to find somewhere affordable to live. Social unrest may prevent an area from gentrifying, but the cure is worse than the disease: few things are more certain in the social sciences than the negative impact of violence on local economic development and on the people who experience it.
Some anti-gentrification strategies are even less focused. Seattle, for instance, in recognition of the city’s historic use of ‘race-based zoning as an exclusionary practice’, voted to rename its single-family zoning designation to ‘neighborhood residential zones’, without actually changing any of the rules.
The gentrification mystery
There is an apparent mystery in gentrification studies: many of them show very little displacement impact. One reason for this may be that poorer residents often exert huge effort to stay in place when rich new residents move in, because they too want to benefit from the additional amenities that come with this. Some studies show lower rates of evictions in neighborhoods considered to be gentrifying.
Moving is more often driven by personal conditions than neighborhood trends. Most people move a number of times throughout their lives because of changes to what they can afford, where they work, or how much space they need. Poorer people move at especially high rates, regardless of whether or not their neighborhood is gentrifying. Even in neighborhoods that are a long way from being gentrified, the most disadvantaged people are frequently forced to move away from their communities by unfortunate circumstances and the demands of their landlords.
This high background churn helps explain why existing residents do not seem to move much more out of neighborhoods seeing rising numbers of college-educated residents and increasing house prices – that is, places that are gentrifying – than others. The base rate of people moving out is already high. For example, according to a 2019 study in the USA 60 percent of lower-educated renters move out of non-gentrifying areas in any ten year period, while 66 percent move out of gentrifying areas.1 In other words, the vast majority of these movers are moving because of their own personal circumstances, not because of gentrification per se. The study also finds that the people who do move tend to move to neighborhoods with similar characteristics to their old one – there isn’t a clear pattern of people downgrading to less desirable neighborhoods.
So why is there such a big movement against gentrification, if it has such a limited impact on displacement? There is a clue in the studies. The ones referenced above look at the whole of the USA. But around four fifths of American cities have no gap between house prices and building costs at the edge of town, and have flat or slowly growing house prices, because they have space to grow outward in all directions, and relatively few restrictions on building (at least on sprawl).
By contrast, when we look at studies of neighborhoods in cities with scarce housing and a large gap between prices and costs – think cities like San Francisco, New York, or London, where the loudest complaints about gentrification arise – we do see significant displacement.
A study of house moves in Los Angeles by Ashley Qiang, Christopher Timmins, and Wen Wang found in Los Angeles found the effects that opponents of gentrification would expect. Looking at lower-income residents specifically, they found that renters were indeed pushed out of their neighborhoods by rising house prices, and worse, that they tended to move to neighborhoods with lower school quality and higher crime. Lower-income homeowners, on the other hand, were more likely to remain in their neighborhoods and benefited enormously from house price increases and better amenities.
These differing conclusions about displacement reflect one key issue. Los Angeles, and indeed California as a whole, faces a severe housing shortage. California ranks 49th of 50 states in terms of housing units per capita. Across the state, many of the most unaffordable counties supply very little new housing at all. As a direct result, it faces the highest housing costs in the country as families compete for what is available, and those who can’t compete are pushed out to states that do build more adequately,
like Texas. Gentrification bites hardest in places where housing supply is constrained.
The narrow corridor of gentrification policy
Good anti-gentrification policy would deliver stable or lower rents, jobs that are appropriate for current residents, and improvements in neighborhood quality like better shops, less pollution, and less crime. This requires careful navigation between two parallel risks: displacement and disinvestment.
As we’ve seen in the cases of Edgehill, Haight-Ashbury, and Los Angeles, desirable neighborhoods that restrict housing supply will experience displacement.
But neighborhoods that don’t bring in new residents at all can face the problem of disinvestment: the trap of concentrated poverty that accompanies areas without infrastructure investment; access to local amenities such as essential retail, quality education, and employment opportunities; or the safe environment required for people and their businesses to thrive.
A specific example of disinvestment came after white flight, where more affluent white families in mid-twentieth-century America moved from urban centers outward to the suburbs, leaving ethnic-minority families behind. Most famous is the example of Detroit, which saw 310,000 white residents leave the city between 1970 and 1980; over 30 years, the city lost half its population and saw entire blocks boarded up and left vacant. The remaining Detroiters did not benefit from this outflux of people. The population collapse came with a catastrophic crash of the city’s economic base. The unemployment rate in the central city reached almost 20 percent by 1990, and with tax receipts down and following decades of struggle, the city eventually went bankrupt in 2013.
What happened to Detroit was, essentially, reverse gentrification. It was no solution to the problems disadvantaged residents faced, and in fact made things dramatically worse for them. It highlights a key challenge in battling the ill effects of gentrification: if your main goal is to keep more affluent residents out of an area, then you are also risking worsened economic opportunities and lower public and private investment. Finding ways to make sure gentrifiers contribute to the community might be more effective. And doing this may mean building homes for them to live in without taking the homes of anyone else.
Letting supply rise to meet demand
If supply is elastic enough, new housing can alleviate or even prevent the forces that push existing residents out of an area. The more new homes you have, the more rising demand from higher-income movers can be absorbed, and the more low-income households are able to trade up into better housing. This can feel counterintuitive to many, because nothing looks more like gentrification than new-build housing, which is typically built to higher standards than existing stock.
But if new higher-income residents want to move to an area – for work, education, family, or simply because they themselves have been priced out of somewhere else – and they aren’t moving into new builds, then they outbid others for existing stock.2
We saw already how Haight-Ashbury failed to prevent gentrification by refusing to allow new homes. Contrast this with evidence from a study of Helsinki that tracked moving chains into newly built homes. Using the Finnish whole population register, the authors, Cristina Bratu, Oskari Harjunen, and Tuukka Saarimaa tracked who moved into newly built units and, crucially, who took the homes they left. They found that while newly built homes were initially occupied by higher-income households, this triggered moving chains where middle- and lower-income families eventually moved into the higher-quality homes that had been freed up by the higher-income ones.
Even more interestingly, while the first residents of newly built homes were often higher income, after those first inhabitants moved out, the newer homes were then occupied by those on lower incomes. Ultimately the study finds that for every 100 new centrally located market units that are built within a neighborhood, 66 become available for below-average-income households through vacancies within just two years. Far from pushing lower-income households out, the new homes make space for more of them.
Building more directly reduces displacement: a 2022 paper by Casey J Dawkins demonstrates that restrictive land use regulations increase eviction filing rates, and strong housing supply reduces them. In fact, it is hard to find a quantitative study into displacement that does not highlight the importance of new housing supply in preventing or reducing the problem.
Of course, permitting this sort of development in order to alleviate displacement is all well and good, except that this development is often what opponents of gentrification are most against. It is often said that all existing residents in every situation are NIMBYs – people who oppose development in their metaphorical backyard, usually because of the disruption or change. But this is not always the case. Since 2018, London has been carrying out a radical experiment with local control: residents of social housing estates, owned by local government or by a housing association, have been given votes on whether regeneration of their estates should go ahead.
These estate ballots give current social housing tenants and owners of homes on the estate a choice of whether they would prefer things to stay as is, or whether they would prefer a brand-new – and usually larger – apartment or house on a new estate. The quid pro quo is potentially years of disruption while it is built, as well as hundreds or thousands of new private renters or homeowners moving in next door – the cost of the regeneration is funded through the development of new homes which are sold or let privately.
Estates can increase enormously in density: in Tower Hamlets, affordable housing provider One Housing is redeveloping 24 homes into 202, of which 50 percent will be rented at below-market rates. All 24 existing households will get new, higher-quality homes, and 84 percent of residents approved the proposals with 100 percent turnout. In Lambeth, 135 homes are being redeveloped by Riverside Group into 441 larger and higher-quality homes, of which almost half are submarket. This was approved by existing residents at 67 percent with an 87 percent turnout. The development comes with not only new homes, but new amenities like a gym, a community center, cycle parking spaces, and a new communal outdoor space.
In practice, when residents are given the power to choose for themselves, they tend to opt in favor of development, even when it means hundreds or thousands of new residents moving in. In fact, residents of social housing estates have overwhelmingly voted in favor of the estates being redeveloped nearly every single time when given the option via estate ballots. Here, no one is displaced, and the benefits of development are recycled into improved homes and amenities for the existing residents – plus, locals have control. And the system appears to work with flying colors. And this is despite potentially large numbers of new residents, typically on much higher incomes, moving into the newly constructed private homes.
This idea could be used to redevelop public housing in high-value locations around the world, in order to deal with the dual problems of insufficient housing density in high-demand areas and repairing and renewing public housing estates that desperately need it. Ballots such as this put residents in control of what happens to their neighborhoods and allow them to share in the upsides of gentrification without experiencing the risk of displacement. They are a win-win for existing and new residents alike.
Outside of the public housing context, more basic and traditional mechanisms to increase the supply of housing might work. Local governments can be more or less favorable to development; part of this is in response to pressure from local residents, but also part is in response to how the development impacts them. Places that receive additional taxes when they permit more homes have a direct and obvious reason to permit more of them: it allows them to achieve their other goals, whether that be increased funding to education or better public transport infrastructure.
But property taxation has been constrained and minimized in many places, reducing the incentive for local voters and local governments to welcome building. In the English context, local authorities that permit new housing can often face large infrastructure pressures, but get relatively little back in additional tax. The income they do receive is also typically heavily restricted by the central government in how they can use it. Contrast this with Switzerland, where cantons receive property and income taxes (not to mention other taxes) and therefore have a much stronger incentive to permit high-quality housing and attract new residents to their areas. American readers will also contrast the housing supply and resultant affordability differences between places like Houston that have high local property taxes and California, which constitutionally caps them.
National and regional governments might consider how and where they can change the set of taxation incentives that make their colleagues in local government reluctant to allow building – even if that requires ceding some powers or cash to them. Reform of property taxation in cities where it does not incentivize growth or allowing cities to keep a proportion of income tax could be paths toward doing this. Incentivizing local governments to permit more homes directly benefits people through lower rents and improved living standards, and the tax receipts can be used to improve local areas. National and regional governments should consider that where things have local costs, like gentrification, then they need to ensure benefits are also captured locally.
Making gentrification benefit locals
Of course, all of this assumes it will be difficult, or even impossible, to completely stop people from moving to existing neighborhoods with good locations but low average incomes. But just because the best defense is accommodation does not mean there isn’t more we could do to make inflows benefit people living there today.
As mentioned above, one way is making sure new residents contribute substantially to local government. As well as encouraging local governments to build the homes required to mitigate the pressures from higher housing demand, this also generates tax receipts that can pay for better public services for disadvantaged residents.
Across most of the United States, education is funded via local property taxation, so there is an upside to attracting higher-income residents who pay more, even for the children of residents who cannot. Schools are not just bolstered by new tax dollars, but also by new institutions like parent-teacher associations designed to improve schools. People who can afford it will often pay a premium to move to a ‘good school district’; this is a good school district moving to you. A study looking at the 100 largest metropolitan areas in the USA finds that the children of less-educated parents who stay in gentrifying neighborhoods are more likely to both attend and complete college, although it is unclear if this benefit occurs through improved school funding or through other neighborhood effects.
The benefits of more mixed-income communities can be very large for the most disadvantaged people there. In the 1990s, the US Department of Housing and Urban Development carried out a randomized experiment called Moving to Opportunity for Fair Housing. Volunteer families put themselves forward to randomly receive either a housing voucher that could be spent on any home or a voucher that got them a specific home in a specific neighborhood with lower levels of poverty than their own, or got only the normal social benefits available to them.
About 4,600 families were chosen and the effects on those who won the lottery versus those who volunteered but did not have been widely studied. In the interviews done at the beginning, a majority said they wanted to win ‘to get away from drugs and gangs’. Nearly all moved less than ten miles, and most moved to another neighborhood with a supermajority of non-white Americans.
One study found that adults who won this lottery saw substantial improvements to their mental health: they saw a 45 percent reduction in the relative risk of serious mental illness, an impact equivalent in magnitude to some of the most effective clinical and pharmacological treatments. Female youth in the study also saw substantial mental health benefits, including a 70 percent reduction in the relative risk of generalized anxiety symptoms, as well as improvements in educational outcomes; risky behavior like drug and alcohol use, and unprotected sex resulting in pregnancy; and physical health, although male youth appeared to experience adverse effects, which may reflect the different behaviors and social impacts within that group.
Other studies of the Moving to Opportunity data found similar things, but for crime and income. Children who moved because of these vouchers from demolished public housing to more mixed neighborhoods, with 21 percent lower poverty and 42 percent lower violent crime rates, ended up being arrested 14 percent less often for violent crimes than similar children who did not. They are nine percent more likely to be employed. And they earn $45,000 more in total (on average) over their lifetimes.
A positive-sum game
It’s understandable that people don’t like the rapid change that incomers can bring to their neighborhood. But the costs of gentrification are largest in the places where housing supply is least able to accommodate it. Especially in these places, much of the existing anti-gentrification toolkit is ineffective or counterproductive.
Though new housing developments are sometimes attacked as a vector of change, new housing can be the best way we have of making gentrification work for a neighborhood’s existing residents. It is unlikely that we can stop people from wanting to move around. But by making that migration less of a zero-sum competition for homes, plus making new house building benefit existing locals, we might be able to make it something that people welcome, rather than something to fight.
Anya Martin is an economist. You can follow her on Twitter here.
In this case, an area is defined as gentrifying if it sees an increase in the proportion of the population who are bachelor’s degree–holding adults aged 25 and up, although the authors report that the results are similar if you use incomes, rents, or house prices.
Although the notion that new builds only go to higher-income residents appears to be untrue: a Swedish study highlights that even amongst the first movers into new-build accommodation, almost a fifth are in the lowest income quartile, rising to a plurality by the second round of movers (see Figure 5).