You are the change you don't seek.

Photographer: Victor J. Blue, Getty Images

My Love-Hate Relationship With Gentrification

Megan McArdle is a Bloomberg View columnist. She wrote for the Daily Beast, Newsweek, the Atlantic and the Economist and founded the blog Asymmetrical Information. She is the author of "“The Up Side of Down: Why Failing Well Is the Key to Success.”
Read More.
a | A

In these modern days, with the gay marriage movement racking up more and more victories across these United States, homosexuality is (thankfully) no longer "the love that dare not speak its name." Allow me, then, to suggest a new candidate for the title: gentrification.

QuickTake Gentrification

Virtually no one has a good word for gentrification. It is lamented in tones from angry to mournful, by political commentators across the spectrum, possibly including me. Yet many of those same people are ... renting or buying homes in "up and coming" neighborhood, which they prize for their proximity to other young(ish), progressive, creative-class people much like themselves. Which is to say that they are gentrifiers. In a neat inversion of the old activist slogan, they are "being the change they don't want to see in the world".  

Their location puts them in the paradoxical situation of wishing gentrification wouldn't happen, while avidly rooting for all the stuff that gentrification brings, from farmer's markets to dog parks. If they are homeowners, too, they are not unhappy about the local price appreciation (their financial plan may indeed require it), however much they may regret its effects in the abstract. As a practical matter, this is something like declaring that you hate the Yankees, but have $5,000 on them to win the World Series. Your loyalties are bound to be divided.

I kid because I love, of course. I'm sure there are people who are actively rooting for gentrifiers to displace all the old residents so they can make money on their homes, and who cares what happens to them? But most gentrifiers don't want anyone to suffer the inevitable losses that come when a neighborhood is gentrified -- the people forced to leave homes they love, the smashing rich local social networks that provide friendship and help in time of needs. They just want to live in a close-in walkable neighborhood that makes it convenient to zip around their city, and they want that neighborhood to have stuff for them to consume. Since they can't afford to live in one of the already-gentrified precincts (at least, not in anything large enough for a family), they buy where they can afford, which is to say, in a neighborhood on the earlier end of the gentrification curve. They may be looking forward to price appreciation when they spend a little more than they can really afford to buy that house, but it's not like they're meaninguflly increasing the amount of gentrification; if they hadn't bought that house, one of the other 28 bidders would have done so, and everything would have gone on much the same.

This is the paradox of gentrification, of course. Individually, none of us is causing it; collectively, we all are. So you say it's a shame, and then you join the herd that's stampeding your neighbors straight out of the neighborhood.

Is there any way square this circle? One way is to impotently rage against gentrification while indulging in pointless but poignant longing for neighborhoods that can be preserved forever in that golden hour just before the first Starbucks moves in. Though this is not actually a complaint about "gentrification," however much its authors insist that they are really concerned about the plight of their poorer neighbors; it is a complaint about "gentrification by people who are richer than I am," which is known to have terrible side effects such as higher property taxes and L'Occitane en Provence stores. This is good for hours of fun at neighborhood dinner parties, but unfortunately, not for much else.

Another way is to move to a gentrifying neighborhood and then demand a vast, region-wide affordable housing program that will allow you to move in, while making sure that your neighbors can afford to stay, too. Hence the apparently limitless market for articles declaring that we suck at affordable housing, and here's how to fix that. This list from Ryan Cooper is roughly typical:

How exactly to get there is an open question -- personally I favor massively expanded public housing, upzoning, especially in already-wealthy areas, and mandating developers to build "lower quality" units (meaning smaller apartments without the concierge service or rooftop pool). That might involve changing the political structure of city governments in favor of city-wide decisions and away from local obstruction.

These sorts of lists are more effective than laments about the March of the Excessively Well Heeled, but not wildly so. 

I hear some gasps in my audience. Wouldn't upzoning help? And rent control and inclusionary zoning and tenant protections ... 

Well, define "help". Would these provide below-market-rate housing for some people? Yes, they would. Will they keep housing prices from going up, and create the fabled stable mixed-income neighborhoods in a growing city? Well, New York City, my former hometown, has many of these things. The District of Columbia, my current home, also has many of them. What both places also have is a lot of gentrification.

But what if we had more of them, you ask? What if we made a concerted effort to preserve affordability in existing units and build public housing -- good public housing, not crime-ridden "towers in a park", but "scatter site" public housing that integrated the poor thoroughly into middle class neighborhoods?

I see the dream, I really do, but there are a few problems with this. The first few are practical, the last, political -- though the practical considerations have political implications.

How Many Units Can We Build?

Consider my current city. Start with just how much housing DC would need to build in order to keep all the old residents  while making room for the newcomers. In 2002, according to the Current Population Survey, the District of Columbia had 284,000 households, 50,000 of which enjoyed household income of more than $100,000 a year -- approximately the same number as had household incomes less than $15,000 a year. An additional 75,000 households made between $15,000 and $35,000 a year; 39,000 made between $35,000 and $50,000, and the remaining 68,000 households took home between $50,000 and $100,000 per annum .

Fast forward to 2013. The District of Columbia had 310,000 households, of which 97,000 were in the over-$100,000 bracket. This may surprise some, but the number of people making less than $15,000 had stayed approximately the same -- I would imagine because so many of them were, one way or another, getting their housing paid for by the government, and were therefore insulated from the rapid increase in rents and home prices. But there were only 47,000 households making between $15,000 and $35,000 and 31,000 making between $35,000 and $50,000, which leaves us with a residual of 82,000 households with incomes between $50,000 and $100,000.

So over an 11 year period , the district gained a net of 26,000 households. But that net figure conceals considerable migration: 61,000 middle and upper-middle-to-wealthy households moved in, while 35,000 working poor and lower-middle-class families departed.

Now, obviously, not all of that represents a net transfer of housing from struggling families to the well off. Homes were built, and abandoned properties were rehabilitated. But to a first approximation, it's reasonable to say that to keep those families in the area, the market or the government would have had to provide 35,000 units of housing which rented for less than $1,500 a month -- in the case of households making less than $35,000 a year, considerably less than $1,500 a month.

For comparison purposes, inclusionary zoning, the latest rage in affordable housing circles, produced 15 units in 2013 (not 15,000, but 15) and though there are supposed to be more in the pipeline, collectively, they don't add up to more than a trickle per year -- and that's counting projects started during the height of a construction boom which is now ebbing. There were also "affordable dwelling units," which don't seem to have a count, but adds some number, not a huge one, to the total.

Meanwhile, according to the database of the U.S. Department of Housing and Urban Development, which is supposed to be fairly comprehensive, the number of total units subsidized by the federal government -- public housing, vouchers, etc. -- was 37,704 in 2013, up from 36,073 in 2009. So we're producing federally subsidized housing at a rate of about 300 a year, and a smattering of units in other programs. To keep all the departed households in the city, we'd have needed to add around 10 times as many vouchers, public housing units, and tax-subsidized affordable housing as we actually did. That would be really expensive.

How expensive? The most recent data show that the federal government provided rental assistance for 30,000 households in DC. On just one program, housing choice vouchers, which account for about a third of the rental assistance households, the Feds spent $144 million annually. In addition, they gave us $25 million in block grants for community development. The local government spends roughly an additional $150 million on housing programs for DC residents . As you can see, most of that money is not going to build additional units, but to the cheaper task of paying rent on buildings that have already been constructed. I can't guess at the multiple of these funds that would be needed to construct and operate tens of thousands of new units for people making less than $50,000 a year, but that multiple would be significant.  

Especially because we can't just start throwing up high rise housing projects like we did in the old days. Not all of those middle income families decided to leave simply because there were no affordable apartments to be had for love or money. There are rentals to be had in this city for $1,500 a month, or even less. But they tend to be either small, or located in less desirable neighborhoods plagued by crime, bad schools, and a lack of transporation or amenities. Working families faced with that choice often decide to leave for suburbs where they can have a car and some space, at the price of a long commute and less contact with their existing social network.

Traditional housing projects, which are relatively cheap to build if you can find the land, would enable us to house more of the near-destitute, who don't have many alternatives -- but who also, by and large, have stayed in the city. Unfortunately, traditional housing projects are pretty awful, and people don't like to live in them if they can afford to live somewhere else. Projects have some advantages believe it or not -- they make it very efficient to deliver social services, for example -- but they also concentrate poverty into a tangle of dysfunction and leave kids growing up in an environment where a middle-class, self-sufficient life is a ludicrously distant dream rather than a realistic aspiration.

The high cost of building new affordable housing is a big political problem by itself. But as we'll discuss tomorrow, it will become an even bigger political problem when we try to build a lot more affordable units.

Voucher Dreams

Enter the idea of vouchers, which are supposed to decrease crime and create more opportunity for the residents of projects by allowing them to find housing elsewhere. There is some question as to whether this method actually does decrease crime, or merely scatters it around. But the general consensus is that it's probably better than brutalist "towers in a park" which became synonymous with urban decay in the 1970s.

The problem with vouchers as a means of fighting gentrification (rather than a means to ensure that desperate people can find housing somewhere) is that private landlords generally are not going to prefer Section 8 voucher holders over affluent renters with good credit histories. Landlords may well prefer vouchers to market-rate renters who are only slightly more affluent than the voucher holders, because Uncle Sam guarantees that (most of) the rent will be paid on time every month, whereas your $12 an hour laborer may get laid off or have some sort of financial crisis. But if your choice is between renting to a middle class non-profit worker, and renting to someone who needs to have most of their rent paid by the government, you're probably going to choose the bourgeoisie. This is why, in recent years, the city has taken to begging landlords to pass up market-rate tenants in favor of voucher-holders, with somewhat limited success.  

As long as there is an influx of affluent people competing for housing with your voucher holders, the voucher holders are probably going to lose. So unless DC gets a lot more housing, vouchers will not solve the problem -- indeed, if gentrification continues to spread, I'd expect to see a net loss of voucher holders in the District proper, as fewer and fewer landlords are willing to take them. There is a reason that voucher recipients tend to end up in neighborhoods only marginally less poor than the ones they left; those are the neighborhoods that their voucher rent will cover, and where vouchers are attractive to landlords compared to other potential tenants. Unfortunately, the benefits of moving people from projects to those neighborhoods do not seem to be as large as voucher proponents had hoped.

Inclusionary Zoning

Okay, if we can't voucherize the problem away, why not build it away? Enter the idea of inclusionary zoning. The idea is that you require developers to build a certain number of affordable units in each new development. DC gives developers "bonus density" -- basically, they get to build taller than the zoning permits -- in order to mitigate the economic cost of doing so. 

Here's the biggest problem: most of our housing stock isn't new. For all intents and purposes, DC is barely building new single-family homes, except on areas very close to the borders of the city, which aren't convenient to public transportation -- and since single-family homes are generally sold, not rented, it's hard to make inclusionary zoning work for them . Meanwhile, 90.3 percent of the multifamily housing stock in the District of Columbia was built before the year 2000, meaning that if the inclusionary zoning law had been active for the entire time since then, and also had covered all new multifamily construction (it doesn't, because the math doesn't work for smaller buildings), we would have increased the total multifamily housing stock by less than 1 percent. Since the 2014 pipeline had about 6,000 rental units, call it 600 units a year. And that is with some heroic assumptions -- no four- or six-unit buildings, and every developer decides to take advantage of the bonus density.

That's not a realistic assumption. The other problem with inclusionary zoning is that it raises the cost of construction to the developer, which means that first, less housing gets built, and second, developers have incentives to sidestep the requirements -- by building the IZ units in a poorer neighborhood where land is cheaper, for example, or by building fewer, larger units that keep the project under the IZ threshold. Both of these will have the effect of raising home prices overall, and concentrating what building does happen into the luxury sector, with boutique projects that are too small to qualify for inclusionary zoning, or high-end high-rises with prices high enough to be profitable despite the affordability requirements. Paradoxically, the more you lean on developers to produce large amounts of affordable housing, the less affordable housing you may get.  

DC has also ended up using fairly high income thresholds and the rents that qualify as affordable, in order to make sure that it's still economic to build buildings in this city. Many of the "affordable" units have rents close to $2,000 a month, and use an 80%-of-area-median-income standard that opens them up to, say, a family of three with an income of nearly $80,000 a year. This is not targeting the low income working families that we started out trying to help; it is instead keeping some middle income families from moving to Maryland. But if you try to force developers to house more low income tenants, you raise the cost of the inclusionary zoning ordinances, and also, unfortunately, make those buildings less attractive to market rate tenants. Result: less housing gets built, and gentrification spreads further.

The Limits of Force

Enter the idea that we should force developers to build affordable units -- as Cooper says, cheaper units without all the amenities. The problem is that you can't exactly force developers to build affordable apartments; what you can do is give them a choice between building those kinds of apartments, and not building any apartments at all. If they choose the latter, then gentrification spreads even faster, as the affluent bid on more and more of the existing housing stock in poorer neighborhoods.

This is also the difficulty with rent control and similar schemes, which is why such programs have generally been retroactive, exempting anyone who builds new apartments. If you try to put new apartments under rent control, developers will just say "Thanks anyway, I'll take my money across the city line." Now, you can wait a while, get some apartments built, and then enact a new rent control law, this being what New York did under Mayor Lindsay in 1969. But this is not a trick you can pull too often, or too many times, since developers are not Charlie Brown, and will not keep lining up to kick the football. You can try to entice people to build rent controlled units, as New York has done with various tax breaks, but as rents and the price of land go up, this gets harder to pull off.

It's probably also worth stopping here to note that rent control, which is frequently mentioned as a solution to the problem of affordable housing, creates more problems than it solves. It does protect current residents, but creates a two-tier market that drives market-rate rents higher, as renters are forced to bid on the limited number of units that are actually available. Meanwhile, rent controlled units tend to stay off the market, because tenants could not get a similarly good deal if they moved. Some people end up ludicrously underhoused -- a family of three living in a small one bedroom apartment -- while widows whose children left decades ago cling to their four bedroom apartments. The bigger the disparity between the market rent and the controlled rent, the larger this problem becomes.

Worse still, for our purposes, this is not even targeted at the poor. Since the control is based on the rent and the age of the unit, not the income of the tenant, it is not well-targeted to lower-income families -- indeed, my experience in New York City was that landlords of these hotly desired properties actively preferred higher income tenants, who were unlikely to miss a rent payment, and whose income was near the range that might cause the unit to exit control . Or they go to insiders who have a connection to the landlord -- which again, means that they are likely to be more affluent than normal. Moreover, over time, the quality of the units tends to degrade, and eventually you need to give the developers higher rents so that they can renovate the facilities.

Okay, so traditional housing projects are terrible, voucher holders can't compete with gentrifiers over the long term, inclusionary zoning doesn't build enough units to halt gentrification before we all die of old age, and rent control not only doesn't work, but creates terrible rigidities in the market while only haphazardly helping lower-income families. What's left? I'll address those options -- and the problems associated with them -- tomorrow.

  1. Or people of similar demographics, since every population is naturally going to have some inflow and outflow as folks die, get married and move across country, or decide to join the armed forces and see the world

  2. Yes, I am aware that these categories are not the same size--the data set from the CPS only goes up to $100,000, so they can't be. I'm using these categories as a proxy for "destitute", "working poor", "lower-middle-class", "middle class", and "affluent", which are roughly the groups we are concerned with when we talk about housing policy. We can argue about where to draw the lines, of course, but someone has to draw them, and I guess that someone would be me, since I'm writing the post.

  3. An awkward length, I'm afraid, but that's the size of the data set

  4. Due to differences in the availability of data, I'm having to mash up different years. Sorry.

  5. To make a long story short: thanks to land and regulatory costs, even a cheap single-family home built in the District of Columbia is going to be a financial stretch for someone who needs considerable housing support.

  6. I know what you're thinking: means test rent control! Okay, let's think this through. There are two ways you can do this: force higher income tenants to vacate controlled units if they get a raise, or decontrol the units. New York does the latter, and this means that you lose the rent controlled units. The former is nuts, though not as nuts as the alternative I've heard proposed, where you try to decontrol the apartment only for the current tenant, using some sort of byzantine shadow-rent system. Obviously, this gives the landlord heavy incentives never to rent to anyone with an income below the means-tested level again--exactly the opposite of what you are trying to achieve.

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.

To contact the author on this story:
Megan McArdle at

To contact the editor on this story:
Cameron Abadi at