Neoliberal desegregation

On bunch of housing-related issues, the United States is badly paralyzed. In aggregate, “we” would like things to change. We would like our housing (and with it access to public goods like high quality schools) to be less stratified by race and class. Many of us (although perhaps fewer mid-pandemic) think that on environmental, social, and economic grounds, we’d be better off if we built and lived more densely, sunsetting postwar US suburban sprawl for something more like the megacities of East Asia.

The pro-density agenda is controversial. But at least in theory, the desegregation agenda is not. Even in this moment, with its boogaloo warriors and the Fox-News-o-sphere shouting from a reactionary crouch as protests rage over racial injustice, pretty much no one in the United States overtly favors residential segregation. Yet almost all white Americans — not just hateful bigots on the right, but liberals, lefties, wishy-washy social democrats — tacitly engage in practices that reinforce that segregation. Bourgeois liberals apologize for the practice, we are not unaware, but once we have kids, we insist they grow up in “nice” neighborhoods with “good” schools, knowing and quietly exploiting correlations between both race and affluence and our scare-quoted notions of quality. One way people try to address this is to encourage (or shame) families into not insisting upon access to segregated public goods, or at least not relying on the correlations embedded in segregation to make their choices. I don’t think this is likely to be fruitful. Structural racism requires structural remedies. To use a much-too-dry, economist-ish analogy, families with the option of purchasing segregation as a visible proxy for high quality public goods are in a situation very much like a depositor in a (pre-FDIC) bank she believes to be sound, but that is facing an incipient run. She might be right about the fundamentals, but a bank run will destroy even a solvent bank. Participating in the run, withdrawing her funds, is antisocial. But failing to participate (by withdrawing early, while she still can) will cost her life’s savings, and won’t save the bank. The individual incentives to behave antisocially are too strong for it to be credible that everyone will spontaneously agree to do the right thing. And if (nearly) everybody will not do the right thing, doing it on your own yields little social benefit and a lot of self harm.

Obviously, the analogy here is not perfect. Our bourgeois liberal’s child may derive real benefits, not just costs, from being educated in a racially and economically diverse setting. (The most affluent families purchase slots in private schools that bus in a carefully selected, “safe”, diversity.) A bank either goes bust or not, but when a well-resourced family does choose to place their child in a less affluent, less lily white, school, there may be incremental benefits to other children and families. If so, each family’s individual choice to acquiesce to segregation as a proxy for public goods imposes an opportunity cost on kids and families who cannot make that same choice. On moral grounds, the case for condemning parents who seek predominantly white and asian neighborhoods as a proxy for safety and good schools is undoubtedly stronger than the case for condemning a person who gets out early during a bank collapse. But given just how starkly public goods like safety and educational outcomes are in fact segregated in our society, and given parents’ unusual solicitude for the welfare of their own children over other moral goods, I don’t think trying to remedy segregation by encouraging or shaming people one-by-one will do a lot of good.

We’ve tried some more structural approaches, most notably bussing and school assignment lotteries, both of which try to decouple residential segregation patterns from school quality. These approaches have not proved sustainable. For traditional beneficiaries of segregated public schools, these practices impose two kinds of costs. Their kids have to bear the cost policymakers intend to impose — sharing more broadly the public schools, and so bearing more of the costs and risks associated with the education of less privileged populations of students. But they also impose deadweight costs, like long bus routes and the quotidian challenges of pick-up and drop-off when home, school, and work may all be in entirely different neighborhoods. These deadweight costs create beautiful personal and political pretexts for rolling back the policies without feeling evil. Of course we support equal opportunity and desegregated schools. It just doesn’t make sense, though, that my kid should have to go to a school halfway across the city, when the school just a block from our home is much better! To sustainably remedy the segregation of public goods like education and public safety, I don’t think anything other than remedying the geographical and social segregation of humans will suffice.

Here I propose a very neoliberal approach to that. Affluent and especially white Americans segregate themselves because it is in their and their families perceived interest to do so. Individually, we face incentives to continue a centuries-old, self-reinforcing dynamic. But one thing neoliberal social engineering is very good at is flipping incentives. What if we literally paid people to integrate their neighborhoods, or taxed people who insist upon, or fail to remedy, segregation? We already have a property tax system. Suppose that for each point of residential geography, we computed the overall demographics within a circle extending for, say, 50 miles. Then we compute a metric summarizing the divergence between the demographics of the (very) immediate neighborhood and the overall regional demography. We provide a property tax refund in an amount that decreases with demographic divergence. Property taxes would become much lower in neighborhoods that are well integrated. Neighborhoods that are segregated would face higher property taxes. All of a sudden, the segregated themselves would face significant financial incentives to figure out how to integrate their own neighborhoods. In effect, we as a polity would be trying to purchase integration with tax credits paid directly to homeowners.

If you take the ugly patterns of contemporary America to be eternal and immutable, you might argue that this wouldn’t work. The public goods that affluent, especially white, Americans associate with segregation are extraordinarily valuable, so rich people would just pay up to keep the status quo unless the tax benefit for integration was implausibly large. I think this misses two important points. First, “affluent” America is not uniformly that comfortable or affluent. In most “nice places”, substantial fractions of residents have leveraged themselves to the hilt to buy their smallish-for-the-neighborhood home in that great school’s catchment area. Affluent America is now full of $1M plus homes. At current-ish tax rates, without going refundable, a full property tax credit could amount to more than $10K in savings every year, a pretty big deal for many people who struggle to afford “nice”. If we wanted to keep the scheme revenue neutral, we’d increase the base property tax rate to cover the cost of the integration credits, widening further the range of credits that could be offered. (And nothing prevents making the tax credits refundable, effectively offering a negative property tax for the best integrated neighborhoods.) There’s a real incentive here.

And for most of these people, I think there is not a real trade-off they would pay up for. Perhaps my glasses are rose-colored, but I think most families who participate in the dynamic that sustains and reinforces existing segregation understand the unfortunately accurate correlation between neighborhood demographics and public goods quality, but do not mistake those correlations for causality. That is they understand that, for bitter historical reasons, if you want to predict which neighborhoods are likely to be safe and have good schools in the United States, racial demographics are informative. But they do not believe that “whiteness”, for example, causes public goods quality, or that “blackness” diminishes it. Therefore, it should be possible to integrate ones immediate neighborhood without a cost in the quality of neighborhood-based public goods. Whenever a house in the neighborhood goes up for sale, existing residents would have a financial incentive to actively recruit diverse newcomers who they’d welcome as neighbors.

There are obviously things that are problematic about this. This recruitment would be quite similar to the way upscale private schools recruit their diversity. Race might not be causal of local public goods quality, but class plausibly is. The PTAs of “public” schools raise a lot of money from parents for “enrichment”. Students whose families lack social and financial resources sometimes require schools to do extra work to compensate, creating a burden. Affluent white neighborhoods would end up competing for the “nicest” (meaning richest, most bourgeois) black families to integrate themselves with. That’s tokenistic and ugly. But still better than the racially segregated status quo.

And of course, the financial incentives would not apply solely to affluent neighborhoods. In most of the United States, public goods quality is middling to poor regardless of racial demographics, but segregation is sustained by some combination of people’s modestly higher comfort level with their own groups and (hopefully modest) discrimination against outgroups. Direct homeowner financial incentives that favor integration could make a lot of headway in these neighborhoods.

It’s not only households of the segregated majority that could use financial incentives to integrate their neighborhoods. “Pioneer” families moving into previously homogeneous neighborhoods face challenges ranging from culture shock and unintended “microaggressions” to overt racial hostility. Under this proposal, pioneer families would effectively receive financial compensation for helping integrate a neighborhood, relative to purchasing a home of similar value in an own-ethnicity segregated neighborhood. And the proposals would give new families and their neighbors a shared, common, financial interest in making things work out.

Under the status quo, racist behavior by white homeowners is encouraged by a plain economic incentive: So long as affluent Americans use segregation as a proxy for pubic goods quality, any integration of a neighborhood reduces perceived public goods quality, and therefore home values as well. Conventional American homeownership is in financial terms a huge, undiversified, leveraged speculation. Homeownership overshadows all other investment for most families, and is for them the basis upon which any financial security rests. Under these conditions, it is not right, but it is also not surprising, that people with expensive houses prioritize maintaining home values above what should be more important social goods, like not being racist. A reduced property tax burden for integrated neighborhoods could offset this financial incentive. Property tax burdens get impounded into home values too. A property tax advantage for integrated neighborhoods would push home prices upwards, offsetting any price effect of integration on perceived public goods, especially over time as correlations between segregation and public goods quality (hopefully) diminish. Realizing the lower tax burden and higher home values would be a shared project for both new, integrating, families and existing residents of formerly segregated neighborhoods.

I’ve mostly discussed the incentives of homeowners, but if landlords experienced the same integration-dependent property tax schedule, they would also have a financial incentive to integrate their buildings. There are devils in details. Would those incentives lead landlords to violate equal housing laws? Should we modify equal housing law to permit practices that would diminish segregation according to our metric, so that landlords can seek to capture the subsidy, perhaps by sharing it with the pioneer families they’d seek to attract? There’s more to think about here.

Integrating disproportionately white, especially upscale, neighborhoods sounds like a worthy project to liberal American ears, but what about integrating disproportionately black or latino neighborhoods? Would an incentive designed to move every neighborhood towards a region’s average demography increase the threat vulnerable communities already face from displacement and gentrification? We value ethnic enclaves like Chinatowns, both as tourist attractions and living communities. Should we design incentives in a way that excludes these neighborhoods, because they reflect a form of segregation we’d prefer to retain? If so, should the same exclusion apply to Little Italys or traditionally Polish neighborhoods? Maybe it is upscale communities that most urgently, and least problematically, should face pressures to integrate. We could design a tax credit that only offsets property taxes above some threshold, so that poorer communities would not face financial incentives to integrate. Perhaps we’d simultaneously make the property tax progressive, so that owners of lower-value homes share some benefit. One advantage of a neoliberal, technocratic, thought experiment is it forces one to think pretty explicitly about values and tradeoffs, in order to translate them to formulas. (But a key disadvantage of neoliberal, technocratic policy is that often formulas are proposed and enacted while the values and tradeoffs they embody are largely inscrutable to the general public, enabling onerous values to get entrenched into law.)

We could use the same technique to purchase densification, if we as a polity agree that densification is something we want to buy. Right now, existing “homevoters” tend to favor neighborhood preservation over densification. Localities enact thickets of zoning regulations whose effect and purpose is to give neighbors veto power over new development. People like Elizabeth Warren have proposed offering Federal grants conditioned on localities reducing land-use regulation, on the theory this will create an incentive to permit densification. But that is not a great approach. It creates incentives for local officials to game the system in order to meet homevoters’ dueling preferences for more amenities (what Federal grants can buy) and continued restriction of new development. The likely effect would be a shift from status quo land-use restrictions to more tacit practices that still enable neighbors to veto projects but aren’t disqualifications for receiving grants. A cat and mouse game between density proponents and preservationist homeowners would ensue. A much simpler approach would be to define a property tax credit for neighborhood density. This would “flip the incentives” of (some) homevoters, from opposing new development to welcoming it. Local officials could then just do what these voters want. By adjusting the scale of the tax credit, and perhaps setting a threshold beneath which it wouldn’t apply, we could try to target densification towards more upscale communities so that it is less of an engine for displacement and gentrification.


16 Responses to “Neoliberal desegregation”

  1. Oliver writes:

    Although I really like the mathematical approach to creating incentives, I think there’s a system much easier to implement, and one that is less likely to be changed year over year as a political football, like the Federal tax system. The complexity seems opaque.

    Skipping over the details that I know and some of my incomplete knowledge: Here in Germany, the schools are all funded at the state (Bundesländer) level. All schools get funded (number of teachers, capital expenditure, and so on) by the state based on the number of students. Local property taxes play almost no role in the local quality of education. Teachers in one city in a state are not paid better than another city in the state, and the quality of the learning materials doesn’t diverge dramatically between schools. The teachers are not employed by local governments,* so they aren’t a local political football. There is some local funding for the building’s infrastructure, but it shared between the local and state governments.

    The end result is a much more homogeneous school system with none of the local school district pressure on housing prices.

    *As I understand it, this was done back in the early years specifically to reduce the influence of the church and the local politicians on the education system, which at the time of creation of Germany were the parallel competing powers.

  2. Sergej writes:

    @Oliver, and the places in schools are distributed, in the first place, based on the zip code of residence vs school, correct?

    That is the system we have here (Austria, not much different from Germany I guess) and which in my opinion re-enforces the argument. We did exactly that to get our daughter into one of the better schools in the city, i.e. we moved into the district so that the proximity argument could not have been used against her. We went to the director to have a chat and make sure the director sees us and understands who we are and bla-bla.

    One can have a system with equal funding etc, yet it does not help. The current system of geographical segregation is so deep in minds of people that a Meldezettel with XYZ zip code literally triggers less questions from banks or authorities compared to zip code ZXY. The same as the teacher’s employment record from a school from district XYZ makes their professional life, current and future, easier.

    The incentives are not financial (in the end schools are public here), and yet there are still major incentives.

  3. blog writes:

    Hmm it looks like your website ate my first comment (it was super long) so I guess I’ll just sum it up what I had
    written and say, I’m thoroughly enjoying your blog.

    I as well am an aspiring blog writer but I’m still new to everything.
    Do you have any suggestions for newbie blog writers? I’d genuinely appreciate

  4. Roger Fox writes:

    I completely agree that the implementation of a lower tax burden and a higher cost of housing will be a common project for both new, integrated families and existing residents of previously segregated areas.

  5. Tom writes:

    The bank run analogy is helpful, and clarifying – thanks for that.

    The question I had as I read was addressed in the penultimate paragraph. It seems there’s a risk that those who oppose integrating wealthier white neighborhoods would make a bad faith argument about “protecting” the neighborhoods segregated in the other direction. (Sort of like opposition to estate taxes couched in terms of protecting farmers.)

  6. quanticle writes:

    I’m not sure I agree with the premise of the post. The premise is that there is a broad consensus in favor of “desegregation”, but there is a collective action problem that ensures that the benefits of desegregation accrue broadly to the community as a whole while the costs are borne by the individuals who choose to desegregate. In other words, you’re presuming that everyone (or rather, a large enough proportion of people) thinks desegregation is a good thing, and is a thing that’s worth fighting for over and above other priorities.

    I don’t think that’s true. Where you see a broad (but weak) consensus in favor of desegregation, I see widespread preference falsification. I think there are a large number of people who say they’re in favor of “desegregation” because they’re told that’s what all right-thinking people support. When confronted with actual desegregation (i.e. actual black people moving into their neighborhoods), however, they find it extremely discomfiting and act in ways that allow them to preserve their segregated communities while paying lip service to liberal values.

    In other words, while I agree that creating a tax policy like this would solve segregation, I don’t understand how you’d build the political consensus to enact this tax structure in the first place. My strong suspicion is that if you actually proposed this tax system, it would be shot down with plausible sounding excuses by the very people who say they’re against segregation, but choose to live in highly segregated neighborhoods. And on the other hand, if people were sufficiently non-prejudiced that they were willing to pass this tax policy, then I suspect that we’d live in a world where segregated neighborhoods would be a problem in the first place.

    I find that this solution has the same flaw that plagues so many other “neoliberal” “technocratic” solutions. It has this core assumption that, if we just make things complicated enough, people will be fooled nudged into behaving in a manner opposing their beliefs. While that tends to work temporarily, what happens in the long run is that people find workarounds that allow them to act in accordance with their beliefs in the context of the new incentive structure. Meanwhile every new system of neoliberal nudges adds to the ever expanding thicket of regulations and bureaucratic complexity that makes real adaptation more difficult in the face of a crisis.

  7. Eli writes:

    But one thing neoliberal social engineering is very good at is flipping incentives.

    LOL. How can you write this with a straight face? Neoliberal social engineering is very good at building complicated policies centered around flipping incentives which make sense on paper (to those suitably indoctrinated in the cult of incentives) but which, in practice, achieve few to none of their stated goals, but enrich a parasitic class of technocrats who ensure that these policies remain entrenched for all time. See: 401(k)s replacing pensions, Obama-era healthcare and police “reform,” basically anything to do with “affordable housing,” etc. etc.

  8. Detroit Dan writes:

    I’ve got to agree with the last 2 comments. While initially sympathetic to neoliberal tinkering with incentives, the last several decades have demonstrated that the simplistic provision of public services is generally preferable. Also, class based approaches work better than race based approaches, in my opinion.

  9. Zach writes:

    I’m more in line with the last three comments, the problems with any solution such as this is at best it ends up favoring those who can deal with the complication and any benefit to those who don’t have the resources to deal with the complexity is partially skimmed by someone looking to “help”. At worst it’s gamed by those able to deal with complexity to the detriment of those who aren’t.

    That I can’t come up with a way it can be gamed means I’m likely in the camp who will end up paying for it and not benefiting.

    With all that said, I’d still probably vote for it.

  10. Chris writes:

    Steve, I usually agree with you about everything, and in the rare case where I see any error in your reasoning it is usually a judgment call way down in the details. But here you are reasoning from a completely false premise.

    We do not have segregation accidentally or as a result of failure to tweak a few marginal incentives – we have it as a result of deliberate policy choices made by generations of voters and politicians. Even in liberal Boston, forced busing was a social and political disaster – and not because of the long bus rides (Boston is not a very large city), but because those who had options voted with their feet by moving out of the city limits (a huge boon to home prices in Brookline, Newton, etc.) or switching to private schools (a huge boon to tuition at the numerous second-tier private schools in the area), and everyone else hated the system that was left behind. People will happily put their kids on a hour-long bus ride for a school they like – you see it in NYC for the magnet schools.

    There is already an incentive available for those who want to integrate segregated urban neighborhoods with troubled school districts – affordable home prices, easy development opportunities with extensive subsidies, and low property taxes (usually – there may some exceptions, but the ones I know mostly finance with income and sales taxes). And yet when a handful of people take up these incentives, they are mostly bashed both by commentators and by local residents as “gentrifiers.” What makes you think that slightly adjusted incentives would make this work on a large scale?

    To Zach’s point – this would be gamed in a few ways.

    If we use an arbitrary or pre-existing set of neighborhood lines, we get the same problem that plagues the “opportunity zone” tax credit – most of the benefits accrue to random census tracts that happen to have the right statistical attributes and investors who jump on these opportunities to extract the tax benefit while making minimal changes to the real neighborhood.

    More likely, there would be a huge push to gerrymander “neighborhood” lines to pull in the right mix to get a credit, without creating functional integration. The line can cross a highway, railroad track, river, etc. Or you can pair a gated community/luxury high rise/single family subdivision with the public housing project or the affordable apartment building along the secondary highway/”bad block” nearby. Then you just need to find a way to avoid any integration. Maybe the wealthy residents tend towards younger and older folks who don’t need the school system. Maybe they continue to choose private schools like the wealthy residents of already nominally integrated neighborhoods in Manhattan. Maybe the neighborhood has two public schools and the “good one” develops direct or subtle ways to exclude students who don’t fit in (test scores, academic standards, tighter discipline, required parental engagement, etc.).

  11. Benign Brodwicz writes:

    I’m sorry, but this is “nudging” taken to a pointless extreme.

    Imagine “paying people to [vote Democratic] …”.

    Equal per-pupil federal funding of neighborhood schools as the Germans do would be the cleanest first-approximation solution, probably. The rich would still hire tutors and fund extra-curricular activities or send their kids to private schools, but the poor neighborhood schools would get *a lot* more money!

  12. Effem writes:

    I have never been so discouraged about the prospect for enacting solutions. To me this is reasonably simple: there are too many groups with unacceptably bad outcomes in this country. The two most notable are urban blacks and rust-belt whites. And most of their issues (and therefore solutions) overlap. Yet as long as we continue to make everything about race they will NEVER cooperate as a voting bloc. All this outrage wasted as we move further from solutions, not closer. Oh well.

  13. Bob Dickman writes:

    Lol what a fucking joke, fuck this blog

    Please explain the impact of this inarticulate proposal on poor communities of color. Blocked

  14. Darius J Wilkins writes:

    1) This is actually a nice little introduction as to why means testing is an important policy tool. Just because it’s usually done really badly does not mean it’s worthless.

    2) One thing people fail to grapple with in post and comments is that it’s not so much that white people do not want to integrate. It goes further than that. White people, as a whole, and a strong, loud, pushy minority in particular, actively resents Black (and other POC) people who gets *any* sort of quality goods or services–from cars to Harvard. There is no way to desegregate without effectively *making* white people desegregate.

    3) For short stretches of history at a time, people in the US and in other places have made integration work well enough. These times are short because they involve two things anathema to the powers that be–it usually requires that Black people have lots of genuine political/social/gun power and it usually requires working coalition politics. Recipe for high taxes, and high labor costs (Oh nos, we’ll have to invest in labor saving devices!!)

    You can really see a lot of this at work now in Mexico and Venezuela–where you see lots of Anglo commentators pouring on the hate, but to less effect than they are used to, because both Mexico and Venezuela has much stronger non-white sub-societies than in most other places in Latin America. Venezuela never could do what Argentina did in eliminating black people, and Mexico cannot be nearly as systematic as Guatemala in organized state slaughter/repression of Native Americans–just much bigger, much more socially cohesive non-white groups in various parts of the country.

    So you sort of see how American liberalism responded to the challenges of the ’50s and ’60s, by breaking up community solidarity through allowing just a little bit of integration–Nixon’s cynical Affirmative Action/Afro-Capitalism policies that soak ambitious and capable minorities away from pushing for *real* changes.

    Therefore the way to improve the quality of educational goods probably would require Afronationalist sentiments, 1840s, not 1920s,30s sort of nationalist, and negotiate with other social block like a union would, against THE MAN.

  15. albatross writes:

    I think there’s an important distinction you’re glossing over: some people probably explicitly want to exclude nonwhites (or maybe just blacks) from their kids’ schools and from their neighborhoods, because they don’t like nonwhites (or just blacks). But for many, many people, they (we) want to exclude dysfunction from their kids’schools and from their neighborhoods.

    I care a lot about whether my kids’ school has fistfights in the halls every day, or disruptive students who shut down learning in some of their classes and can’t be controlled by the teacher or principal. I care a lot about whether my neighborhood has people tossing trash in their yard and having loud parties every night. I don’t actually care at all about the race or social class of the people doing these things. I suspect this is also the sentiment of most of the well-intentioned liberals who pay extra for a house in a nice school district.

    Dysfunction is inversely correlated with income/social class, so by buying a house in an expensive neighborhood, you buy less dysfunction, and by buying a house that is expensive because of its school district, you buy a place for your kids in a school with less dysfunction. This is not a great situation, but it’s the best one available for most people who don’t want their kids to be in a dysfunctional school or to live next to a drug dealer or someone with a car up on blocks in their front yard. Since income/social class correlates with race, you also get racial segregation there. (One way to see this is that the racial segregation in housing we see now doesn’t seem to involve excluding Asians or Jews. I think the older racial segregation based on restrictive covenants routinely excluded both groups. But Asians and Jews both do okay economically, so we don’t see the segregation pattern there.)

    IMO, the only good solution here is to find ways to avoid dysfunction that don’t make horrible distortions on the housing markets.

    The best general solution I can think of for schooling is vouchers, since that just completely unlinks your home address from what school your kids attend. Make sure the schools can select for kids and can kick out kids who are disruptive, and you probably get something pretty workable.

    For general quality of life in the neighborhood, I think the best answer is probably HOAs with some teeth. Though that has plenty of downsides in terms of annoying rules (you can’t paint your mailboz that color without special permission!).

  16. asdf writes:

    And for most of these people, I think there is not a real trade-off they would pay up for. Perhaps my glasses are rose-colored, but I think most families who participate in the dynamic that sustains and reinforces existing segregation understand the unfortunately accurate correlation between neighborhood demographics and public goods quality, but do not mistake those correlations for causality. That is they understand that, for bitter historical reasons, if you want to predict which neighborhoods are likely to be safe and have good schools in the United States, racial demographics are informative. But they do not believe that “whiteness”, for example, causes public goods quality, or that “blackness” diminishes it. Therefore, it should be possible to integrate ones immediate neighborhood without a cost in the quality of neighborhood-based public goods. Whenever a house in the neighborhood goes up for sale, existing residents would have a financial incentive to actively recruit diverse newcomers who they’d welcome as neighbors.

    It’s causal. See The Bell Curve, etc. Everybody knows it and acts like it’s true no matter what they say in public. Revealed vs Stated Preferences, even if such knowledge is subconscious rather than conscious.

    You’ve been trying this nonsense for five decades and it never works. The whites/asians just keep moving around to avoid the blacks/browns. All you’ve ever managed to accomplish is making some poor last ones out of a neighborhood lose their life savings.