The government does not believe that the New Homes Bonus is an effective incentive. It often rewards authorities for housebuilding that would have occurred in the absence of any incentive, and its efficacy is further blunted by the interactions with the remainder of the Settlement…Given its limited impact on housebuilding, removing the New Homes Bonus from the Settlement does not run counter to the government’s ambition to build 1.5 million homes in this Parliament, to tackle the housing crisis and kickstart economic growth.

This was a pretty big incentive. My LLM estimates that it gave local councils an average of £3,900 per net new home over the life of the program. But despite the relative generosity, none of the evaluations I was able to find suggested it was significantly changing the local calculus. And it certainly didn’t create a sea change toward new housing, which is extremely difficult to build in England.

Perhaps a more rigorous evaluation would find statistically significant effects and allow a proper cost-benefit analysis? But it’s not an encouraging piece of news and nudges my priors downward for what we should expect from US state and federal schemes to incentivize growth.

Thanks to Lyman Stone for putting this one on my radar.

Highlights

Some of the best papers in my notes resisted categorization:

Lu Han (presenting), Rachel Ngai and Kevin Sheedy presented a nice model that helps explain the Current Moment in housing cost. When the Fed increased interest rates sharply to curb inflation in 2022, that made homebuying relatively less attractive than renting. That change decreases prices but increases rents. Here’s a slide showing the change from an earlier presentation:

But the authors find that a model with few frictions can’t explain this change. Instead, they need rental search frictions, which allows the lessor to capture more of the surplus when there are more apartment-seekers chasing each vacancy. For renters, when it rains, it pours.

Buy-to-rent investors play a mitigating role in this framework, shifting some properties from ownership to rental. Too bad federal and state governments are trying to shut them down.

Yurou He presented a clever job-market paper with good theoretical and empirical contributions. Land assembly for redevelopment creates a surplus that each seller would like to extract. In particular, she used the layout of the parcels to identify “pivotal” parcels, which are indispensable to the geometry of the assembly. A very curious fact, which she didn’t spend time on, is that assemblies with either 0 or 2 pivotal parcels are very common; assemblies with 1 are very uncommon. I’m not sure if that’s a boring artifact of parcel geometry or a fascinating artifact of multiplayer bargaining games – it would be good to know.

As theory predicts, the sellers of the pivotal parcels (only) are able to command above-market prices during assembly.

(The one aspect of the project where I remained unsure was whether she had reliably captured cases where the assembler hides behind various shadow buyers to prevent bargaining, which seems like the obvious move.)

Micah Baum won the student prize with his paper on the social consequences of Houston’s minimum lot size reduction. He looked at the outcomes for children who lived in neighborhoods where small-lot homes were allowed (he errantly calls this area “downtown”; it’s geographically larger than the entirety of DC, Boston, SF, or Manhattan). To do this impressive study, he uses individual-level Census Data.

One thing this allows him to do is directly characterize the 20,000 households that moved into the new small-lot houses:

In the affected neighborhoods, children of incumbent homeowners came out way ahead: they stayed in place, experienced improving neighborhoods, stayed in school, and benefited from higher parental housing wealth. Renters’ children, by contrast, were more likely to move out of central Houston. Although their new neighborhoods were no worse, they suffered some costs of disruption and had worse life outcomes.

Although the net benefits outweigh the costs, Baum finds, there is no easy way for the winners to compensate the losers. But even if that’s true, it’s not unique. Social life in any system (capitalist, socialist, corporatist, whatever) is shot through with non-Paretian tradeoffs. Developed-world cities and nations today use progressive taxation and public goods provision as a means to roughly net out the millions of little pros and cons. The degree may be too low or too high, and surely some individuals come out way ahead, but it’s not as though non-Paretian efficiency gains are unique to upzoning.

A quick trip around the world

The most surprising paper of the conference was from Yuki Kanayama (presenting) and Taisuke Sadayuki, who look at the causes and consequences of housing dereliction in Toshima City, Tokyo. Kanayama confirmed that even today, when 30-year-old houses have relatively high quality, Japanese buyers are uninterested in used housing. So there’s a steady supply of old houses going to seed. The authors use decades-old road plans as an instrument to define treatment and control groups: if owners think that upzoning or eminent domain might someday come along, they’re more willing to sit on derelict properties. So far, no surprises.

More surprising: the presence of a derelict house really drags down local rents. A single derelict lowers rent within 80m by 1.6%.

But here’s the real surprise: The rent cost of a derelict house is much larger (more like -5%) for rentals that are “far” from a police koban or fire station.

How far is far? That’s the biggest surprise: the median distances from police and fire are 300m and 500m. If this is true, Japanese people are really uncomfortable living more than 5 minutes’ walk from police and fire service.

Chatting after the presentation, Kanayama said that friends’ anecdotes support this view: they check the distance from emergency services before renting. I don’t think I’ve ever heard an American friend mention fire and police proximity as a pro or con.

Filip Sangild Beck, David Jinkins (presenting), and Ismir Mulalic look at the role of public housing in a rapidly gentrifying city: Copenhagen. True to stereotype, public housing in Denmark is abundant (20% of the stock), desirable, and not particularly targeted toward the poor. The American presenter told us he’d signed up his Danish-born child on the waiting list.

The authors’ question is an ideal case for using a QSM: how would Copenhagen be different if it changed its public housing policies? They find that

• Reserving public housing for lower earners (= non college grads in their model) would increase rents in the urban core (which is preferred by college grads).
• Selling off public housing in core locations and replacing it with units in cheaper, further-out areas would decrease core rents
• Privatizing all public housing would lower aggregate welfare.

The bigger point is that we now have the tools to evaluate the equilibrium effects of complicated policy changes, including consequences like relocation of workplaces.

John Mondragon presented work with Schuyler Louie, Rami Najjar, and Johannes Wieland – but not the controversial paper you’re thinking of. This is a newer investigation that attempts the difficult task of harmonizing housing costs and consumption across countries. He presented 5 “stylized facts” about housing across the development cycle:

1. Dwellings per capita increase with GDP per capita, but then they level off. (In other words, household sizes shrink as incomes rise, but stop shrinking after some point).
2. The number of dwellings rises with the population in both rich and poor countries. (This is obvious at some level, but I think it’s the closest to being evidence against regulation mattering).
3. Real house prices increase with GDP.
4. Real house prices are not correlated with the change in population.
5. Housing expenditure shares increase with GDP/capita. Almost all of this increase is accounted for by the number of dwellings per capita, at least until a high level of income.

To my reading, there’s nothing in these facts that is serious evidence for or against an important role for regulation.

They interpret the facts, though, as in their earlier paper, as evidence that, above some level of income, demand for the number of housing units is satiated, but demand for quality keeps rising. The problem with making this the central story about housing cost is that, within developed countries, people in the most expensive cities live in the smallest, crappiest dwellings.

One minor point: they really need to switch from using people in the denominator to adults.

Hans Manner, Ferdinand Rauch, and Martin Schlesinger (presenting) had an alluring title: the Industrialization of the Alps. It is a great question, indeed, why the northern slopes of the Alps are currently the highest-income region in Europe, across multiple national and linguistic borders. These areas were marginal to the Roman Empire, difficult to reach by water transport, and mostly irrelevant in the Middle Ages.

Sadly, this paper didn’t tell us any of that. Instead, it was a within-region comparison of a single Austrian province (Styria). Places with coal industrialized first. Early electrification was somewhat random but generated permanent gains.

QSMs rule everything around me

Nate Baum-Snow (presenting) and Daniel Hartley (according to my notes, but not according to the conference program) want to study how “gentrification” affects neighborhood change. It’s a great application of the newly-popular Quantitative Spatial Model (QSM) revolution. But it was not clear to me at all that they were studying something recognizable as “gentrification”, because I didn’t hear anything about limiting the area of study to places that were initially poor.

Rather, what they clearly did study was what happens to a neighborhood when residents’ commuter market access to jobs requiring a college degree improves:

• The college-educated share of residents rises
• Rents and home values rise
• Outmigration falls
• The future earnings of neighborhood children rises
• Local amenities (as estimated by the model) rise roughly one-to-one with the share of workers who are college-educated.

This is a good time for a shameless plug of my own contribution to the conference: I handed out flyers advertising the first free, national, easy to use dataset of US commuter market access, the Mercatus Commuter Market Access Dataset of Congested Auto Travel (McMADCAT).

It was well-timed: it seemed like the conference featured dozens of papers with QSMs, and the first keynote lecture, by Stephen Redding, covered the basics of the QSM. Many of the other papers reviewed in this post use a QSM even if I don’t mention it.

Amy Min Zhang explores the effects of regulation on pollution in general equilibrium. Even nationwide regulatory changes can have bigger impacts in higher-regulation areas. The 1991 Clean Air Act amendments caused a shift in heavy industry from the Rust Belt to the less-regulated South. Using a quantitative spatial model, she finds that the regulation cost 0.2 million jobs nationwide, but reallocated 0.4 million jobs from high- to low-regulation regions. (What was less clear in the presentation was whether firms in the low-regulation areas could somehow evade the federal requirements or if the issue was the interaction between federal and state regulation. She also did not incorporate an international dimension, but it seems very likely that jobs also leaked overseas.)

Frederic Robert-Nicoud, Pierre-Philippe Combes (presenting), Gilles Duranton, and Laurent Gobillon offered one of the only theory papers of the conference, a new workhorse model (they hope) that improves on the popular QSM framework by introducing non-homothetic housing demand.

The fundamental tradeoff of urban economics is between agglomeration benefits and crowding-and-commuting costs. But if people can sufficiently economize on housing by living in smaller and smaller dwellings, then the models predict that every city will either grow infinitely or shrink to zero.

The authors’ contribution is to make housing a necessity while keeping the model tractable. Now, when a city’s size increases and both prices and wages rise, the necessity effect forces people to spend a larger share of income on housing. That is, the costs eventually outweigh the benefits. In a system-of-cities model, this allows every city to continue existing at some level of population.

Planners, checkers, and politicians

Tom Cui (presenting) and Beau Bressler look at the effects of a major urban planning effort from the 1960s. “Section 701” brought federal dollars to a large percentage of small cities – but those above 50,000 in population were ineligible. The authors have a tricky time generating precise estimates because the total number of cities is modest – so they can’t just use the obvious empirical strategy of comparing those just above & below the boundary. But they use variation in state enthusiasm for the program within Census regions (Alabama was eager, Texas was indifferent) to generate triple-difference estimates.

They find that receiving planning assistance generated a huge, negative effect on housing growth in the 1970s and 1980s, continuing (but with less precision) into the 2000s.

And this wasn’t because the planners were promoting smart growth: the drop in permitting is concentrated in multifamily housing. It also increased the use of value capture policies. The authors interpret the findings as local adoption of a “regulatory innovation” that more efficiently converted local NIMBY preferences into policy.

Jacob Herbstman tries to suss out the effects of Chicago’s aldermanic privilege on housing supply. Aldermen exercise extreme discretionary regulatory power within their wards. Until Lori Lightfoot was elected, each alder received a nightly (!) email with all the building applications in his ward.

Herbstman uses 2015 redistricting as a boundary/event study, using city blocks that were redistricted as the treatment group. He finds that shifting to a more-strict alder decreases permitting in those blocks by 20%. However, standard errors are large for all his estimates.

Overall, I’m not completely convinced by the experiment. The districts are racial gerrymanders, and the alder’s race may be correlated with strictness. Blocks that are redistricted are more likely to be those that have undergone racial change or to be a mismatch with one alder’s race, which probably impacts the supply or demand for permitting in other ways. This is a case where the empirics are less valuable than gaining a rich qualitative understanding of the underlying phenomenon. Even if aldermanic privilege was proven to have zero net effect on permitting, it would be objectionable on the grounds of fairness and corruption.

Desmond O’Shaughnessy presented one of the papers by The American Zoning Project (headed by Alex Bartik and Arpit Gupta). The team has collected 3,600 historical zoning codes from 2,500 cities, and is in the process of digging up more. They find that codes have become more flexible over time with respect to use, but stricter with respect to parking and procedure and more extractive in terms of value capture. One historic episode they note is a 1950s push by industries to replace rigid zoning lists with measurable performance standards for industrial zones.

They have committed to making their data fully public.

Steph Kestelman revealed negative research about a zoning program from Los Angeles: although the Transit Oriented Communities caused a big increase in applications, it did not lead to an increase in completions. It did nudge some production to neighborhoods where it is more difficult to build (in the upside-down world of LA, that means lower-income neighborhoods, where anti-gentrification sentiment makes it difficult to build apartments).

The surge in applications was largely explained by developers filing incomplete, quickie applications that were placeholders for later consideration. The lack of a building response, according to developer interviews, is that the density bonus offered by TOC was not enough to offset the deep affordability requirements.

It will be interesting to see if new zoning forced by SB 79 finally breaks open LA’s stingy zoning near transit stations.

Suvy Qin studies how city personnel impact approval processes for building plans, also in Los Angeles. Plan review in Los Angeles averages 300 days (and the standard deviation is also 300 days), almost all of the time spent in back-and-forth between checker and applicant.

On average, the city can hire better plan checkers during economic downturns, allowing Qin to exploit timing as plausibly exogenous variation. Those better plan checkers sign up for more overtime, complete more reviews, and have equal quality outcomes to the control group. For the city, it would be economical to offer higher wages in order to attract better plan checkers. My own view: it would be even more economical to reduce the stringency of plan checking.

Connections and boundaries

Another theme of the conference is the enormous quantity and quality  of detailed data that researchers can throw at questions big and small. Even Ph.D. students are now writing papers with propriety datasets. Where a quantitative case study was considered pretty good a decade ago, a part of proving your chops as a student is to assemble the most comprehensive data on your topic.

She’s now a professor at Cal, but the most unlikely student-data story might be from Laura Weiwu, who apparently discovered data on a neglected Census Bureau server allowing her to link 57 million children  born in the 1960s and 70s to their parents in tax records. Given the era, she asks how the expansion of highway systems affected income mobility. Using a QSM, she finds that new interstate highways benefit upward mobility: doubling the commuter market access in parents’ locations increases children’s future incomes by 30%. But the benefits were not shared by Black families, because they were excluded from many of the suburbs that became accessible via interstate construction. She uses a formal model to show that, in the absence of segregationist barriers, Black families would also have benefited.

Michael Neubauer (presenting) and Gabrielle Grafton ask what happened to those left behind during the Great Migration? It’s a great question, and they have a clear answer: wages rose, visible racism decreased, and some farms closed. A 1 percentage point increase in Black emigration led to a 4% increase in wages for Black stayers and a 2% increase in wages for White stayers.

The challenge for studying something like this is that the selection effect is intense. Who migrates and who stays depends on their personal characteristics, and the rate of migration depends on both push and pull factors. They try to deal with individual selection by limiting their study to people with characteristics that make them very unlikely to migrate (e.g., older people). And they try to deal with the endogeneity of the migration rate by isolating pull factors in places with strong migration links.

However, as some attendees pointed out, it’s likely that places with strong migration links are also likely to have other economic links. So the authors have more work to do to convince readers that their approach is really isolating the effect of outmigration.

Kaan Cankat (presenting) and Allison Green try to understand how important was the ability of central cities to annex land during the 20^th century suburban boom. They used a staggered diff-in-diff and instrumental variables, both of which I found unconvincing. If you believe their results, they get a decrease in the per-capita costs of some city services (like firefighting) while other costs (like sanitation) stay constant or rise slightly. Overall, the effective property tax rate declines 2.5%. If these results are right, the suburbs (especially when brand-new) were subsidizing the city. There’s also some role for economies of scale: fewer fire stations, but with a higher ratio of engines (expensive) to trucks (cheaper).

Speaking of city lines, Colin Williams, a runner-up for the student prize,documents that HUD Code manufactured homes are much more common outside municipal boundaries than inside. He uses the entire USA – except New England, which has few unincorporated areas.

Subsidized housing

Evan Soltas (presenting) and Allan Lee find a catastrophic decline in the efficiency of Low Income Housing Tax Credits (LIHTC): “real construction costs rose 80 percent from 2013 to 2024, holding fixed project size and location.” They can’t fully explain why; rising complexity is one candidate.

Beau Bressler built a database of all US public housing, including resident characteristics at the building level. With this level of detail, you’d hope to uncover invisible insights. But his conclusions are mostly in line with the conventional wisdom on public housing:

• It was built much more in poor, Black neighborhoods. Evidence from Philadelphia shows that abandoned proposals for public housing were in neighborhoods with a near-zero share of Black residents.
• It worsened the trajectories of the surrounding neighborhood, especially when built in poor, Black neighborhoods (maybe that’s a modest surprise).
• Raj Chetty’s Opportunity Atlas shows that public housing tracts have very bad upward mobility, but that effect is entirely explained by observable characteristics like poverty and single-motherhood.

Dionissi Aliprantis (presenting) and Stefanie DeLuca (and maybe a third author named Park?) try to understand why the Baltimore Regional Housing Partnership (BRHP) was so much more successful than the similar Moving To Opportunity (MTO) program. Both are premised on giving housing vouchers to people who are poor and mostly Black and nudging them to move to better neighborhoods. MTO gets praised a lot for having benefits for younger children, but its outcomes for older children and adults are mediocre and the MTO moves often took people to only slightly-better neighborhoods. BRHP, on the other hand, resulted in a huge change in neighborhood type. The typical BRHP resident went from a 3^rd-percentile census tract to a 71^st percentile tract! By contrast, receiving a housing choice voucher typically moves people from the 4^th to the 21^st percentile.

The authors identified two reasons why BRHP was so successful. The first is that the worst Baltimore neighborhoods are in extremely bad shape. The bottom 10% of regional neighborhoods (they constructed a composite measure of neighborhood socioeconomic status) contain 30% of the region’s Black population:

With disadvantage and segregation so extremely aligned, it’s not hard to find improvements.

The second reason, and the one that dominates statistical analysis, is that BRHP was regional in design and intent – after all, it was the result of a court order to address segregationist policies in suburban Baltimore County. Fully 2/3 of the improvement in neighborhood quality is accounted for by cross-border moves: people leaving Baltimore City for the county. This finding implicitly damns the non-portability of Housing Choice Vouchers.

Aliprantis recommended the book Waiting for Gautreaux, a first-hand account of the similar highly successful, court-ordered voucher program in Chicago. Another book that’s great for understanding the Baltimore story is Lawrence Lanahan’s The Lines Between Us.

Homelessness and crime

Elior Cohen looked at what happens in Los Angeles when a new homeless shelter or rehousing site is opened. From 2014 to 2024, homelessness rose by 33% while the number of beds rose by 54% – to the point that there are more shelter + rehousing beds in the system than homeless people in the PIT Count. In the most recent PIT count, 83% of LA census tracts had at least one unsheltered person.

A surprising feature of his data was how often sites closed; my assumption was that once a site came online, it lasted for decades.

Using an event study framework, he found very different effects of shelters versus permanent beds.

• There’s strong spatial sorting: almost all facilities open in places with a lot of unsheltered homelessness at baseline
• A new shelter increases the number of sheltered homeless in the Census Tract (obviously) without decreasing the number of unsheltered.
• Permanent housing increases the number of unsheltered homeless in the tract.

There’s no great takeaway here unless you’re looking for evidence to oppose housing for homeless people in your neighborhood.

Emily Merola also studied homelessness, but in Seattle. She looked at encampment closures using a diff-in-diff with matched comparison group designed to avoid spillovers. Again, it’s all depressing. About half of closures are repeats (the city clearing a location it had cleared not long before…the median time between repeats is 136 days). She has promising data – cell phone pings to measure foot traffic, police and city service calls, but ultimately finds muddy evidence. Public service requests at the cleared site decline, but they rise by even more in nearby areas. Crime declines at the site, but then rebounds.

Jens Ludwig, author of Unforgiving Places, gave a keynote lecture on US gun violence. For a conference occurring in the City of Big Shoulders, it was a worthy homage to the influence of Chicago School sociology on economics. His decades of work with Chicago neighborhoods and police have convinced him psychological work is the key to reducing gun violence: “78% of gun homicides are arguments settled with guns.” Methodologically, he sees “too much Becker and Friedman and too little Richard Thaler” in economists’ treatment of crime.

While I have no critique of the specific interventions he likes (such as training adolescents to deal with conflict), the macro story left me unconvinced. I think one questioner pointed out that there’s no time series in “System I” versus “System II” thinking. And while gun-use may normally occur in a moment of passion, gun-acquisition requires some forethought.

Ludwig agrees that magically removing all guns would stop most homicides, but points out how politically impossible that is. It seems to me that convincing every poor, disaffected adolescent to submit to intensive conflict-management training is almost as impossible. Nevertheless, many cities have experienced large increases or declines in murder rates. There are “easy” sources of variation well within the political norms.

The City of Chicago was also a highlight. The day before the conference was my own Ferris Bueller’s Day Off, complete with a Cubs game (the Cubbies clinched a playoff berth and Pete Crow-Armstrong entered the 40-40 club), beer gardens, and gracious local hosts who took time out of their days to show me some of their favorite aspects of the city.

Peak Chaos (45% to 55%)

At the midpoint of AV road share, the system no longer risks mere strain, but failure. When the roads contain a complex mix of fully human-driven vehicles, ADAS-supervised vehicles, conditionally automated vehicles, and geo-fenced robotaxis—none dominating—no single behavioural logic will govern traffic.

I think this is not quite right on two counts:

1. The peak is likely to be fairly flat, since the randomness of specific interactions will make 40% saturation feel a lot like 60% saturation. Elsewhere, Andrew says that with fewer than 30% AV’s, it will be “the calm before” the storm. It can’t both be true that 50% is a disaster and 25% is basically fine. All the same types of situations will emerge at both saturation levels, just at different frequencies.
2. I suspect that peak chaos is going to occur at a much lower level of saturation.

Mental modeling

Full disclosure: this post began as an email to Andrew (whom I like and respect!) that I decided to formalize instead.

If I were to start modeling road chaos with the introduction of a new type of driver, I would want to include three key ingredients:

1. How frequent are the opportunities for confusion
2. How quickly do human drivers adjust to machine drivers
3. Whether machine drivers are ‘smoother’ than human drivers in the final equilibrium

Machine drivers have to adjust to human drivers very early in the process or they won’t be allowed to stay on public roads; we can take that as given.

Analogy on two wheels

From long experience as a cyclist, my guess is that the crossover point – where further saturation starts reducing chaos – happens at a pretty low rate of saturation, more like 5% than 50%. When I cycle in suburban areas where cyclists are rare (especially back in the 90s), drivers can be genuinely confused with how to deal with me. They might proceed very slowly behind me, or pass dangerously, or honk.

But at a modest saturation rate (maybe 1%), most drivers adjust and learn how to pass a cyclist, or handle a cyclist passing them in the red light queue. But at the same time, the number of opportunities for error is 50 times higher at 5% cyclists than 0.1% cyclists. Do drivers get 50x better with regular exposure to cyclists? No. No they do not.

In particular, there are a small number of drivers who have no clue how to drive around bikes. As bike saturation rises, that small number is forced by circumstance to acclimate. At the extreme you have the Netherlands, where it’s inconceivable that a driver not know how to operate around bikes. In DC or Boston, the 5% or less bike share causes more disruption than does the 50% bike share (or whatever) in the Netherlands.

Back to robots

To my mind, disruptive human-AV interactions will occur when (a) there’s a bunch of vehicles involved and (b) at least one of them hasn’t adjusted. Given that it’s humans, not machines, who still have to learn to deal with the other, conditions (a) and (b) are going to be satisfied most commonly when there is still a large human majority on the road, but some of those humans haven’t yet learned how to interact with a machine driver. 

Long before we reach 50% saturation, humans will internalize lessons like “if I double-park, AVs will just sit there, cause a huge backup and people will be angry at me” or “AVs are going to follow yield-to-the-right rules strictly”. Once the frequency of un-accultured human drivers starts to shrink to a small minority, then the worst is over. Peak chaos might occur when 5% of vehicles are AVs and 20% of humans are still clueless. These are just vibes numbers, of course, and cluelessness is not a binary variable. But I’ll bet Andrew a beer that the number of human-AV collisions peaks (in a given geography) before AV saturation reaches 25%.

Deeper differences

So far, I’ve written this agnostically, as if humans and machines have random differences and we could swap their places in the model. But it’s not random. Machines are far more predictable and uniform in behavior than humans. So it will be pretty easy to learn their tendencies. 

The part that I don’t know is whether machine drivers will ultimately be smoother than humans. On foot, humans are very good at negotiating each others’ space with instinctual balance and nuanced body language. We’re less fluent in vehicles, but still pretty darn impressive given that we didn’t have any evolutionary hardware for solving spatial problems at 100 kph. Will AVs be able to copy all of that, or develop their own smoothness? Or will they be held back by the necessarily-high degree of conservatism in programming? If AVs are just plain clunky and slow, then city driving will become slower and clunkier as AVs take market share even without any misunderstanding between modes.

The Build Now Act uses an existing stream of federal funding for larger local governments, Community Development Block Grants, to reward localities that improve their net housing growth rate over the course of a decade, and penalize those that worsen their growth rate. The amounts of money in play are modest: the median grant is less than $1 million per year, and the penalty is 10%.

A big part of the legislative process for the Build Now Act was to exempt most places from eligibility, which means they can neither gain nor lose funding.

• Any place too small to receive direct CDBG funding (e.g. cities below 50,000 people) is implicitly ineligible and doesn’t show up on the map below.
• Any place with low housing costs is ineligible.
• Any place with a high rental housing vacancy rate is ineligible. Unfortunately, most of the variation in rental vacancy rates seems to be driven by mis-categorization of short-term rentals. This makes San Francisco ineligible.
• Any place where a federally-declared emergency or disaster, such as a major winter storm, occurred in the past 3 years.
• Any county that “lacks the legal authority to enact or update zoning and permitting ordinances.” It remains wide open to see how HUD will interpret this. In the version mapped here, I made conservative estimates.

Throughout the legislative debate over Build Now, I maintained a dataset that helped me loosely estimate how the law might work in practice. My data are not perfect and have some known errors. And, of course, they are backward-looking. Build Now will not take effect until FY 2029, I believe, at which point the evaluation decade will have flipped and the map of disasters will have shifted.

In the map embedded below (and hosted by Datawrapper here), I show which Entitlement Communities would be ineligible and which would gain or lose funding based on the data I have available. NOTE: The amount of data seems to be straining the Datawrapper server. Reload this page several times and eventually the data will show up.

The tooltips show each community’s key stats, including reasons for ineligibility or funding gain/loss. Here’s the Bay Area, where high rental vacancy rates (i.e. lots of short-term rentals for business travelers) make most municipalities ineligible:

And here’s detail from Utah, where South Jordan would lose funding despite having the 25th-highest housing growth in the country from 2020-2024, because its housing growth rate from 2015-2020 was even higher, so it has worsened its performance.

Don’t take this map so much as an indication of which places will actually gain or lose funding when the law comes into effect. Rather, it’s an indication of what a typical year might look like. The biggest takeaway is that most places are ineligible. The second takeaway should be that the law ought to be amended to better match what proponents say it’s meant to do: provide more federal funding to support places that are growing, even if those are in cheaper parts of the country or experienced a nasty snowstorm 2 years ago.

UPDATE: What would I do instead?

If you think the Build Now Act is unnecessarily complicated and poorly targeted to support or incentivize housing production, then you’re right. How can it be fixed? One way, which I & others proposed several years ago, is to build a complex but capable formula.

The other way, which I’ve come around to, is brutally simple:

• Define eligibility in a way that includes at least 2/3 of jurisdictions. Disaster exemptions should be only for disasters that destroy homes, like the Altadena fires.
• Take 10% of CDBG funding away from every eligible community
• Rebate that funding pot back to every eligible community based on how many net homes they built in the previous year.

More homes, more funding. Any politician can explain that to his constituents.

Another type of restriction that deserves attention is deed restrictions. Although El Paso, Texas, has fairly permissive zoning and new state preemptions that legalize residential in commercial zones and small-lot subdivisions of 5-acre+ parcels, those permissions cannot be used on the huge swath of commercial and residential land restricted by deed.

Alex Hoffman, an urban planner at the City of El Paso, mapped out deed restrictions in the city – yielding a map of rigidities that looks suspiciously like a zoning map. Alex even found 5.5 acres restricted to quadplexes only: no triplexes or five-plexes, thank you very much!

Even in neighborhoods where single-family, multifamily, commercial, and public uses exist alongside each other, most of those uses are locked in by deed.

There’s a range of reform options for deed restrictions, including some combination of the following:

• Cancel parts entirely
• Terminate deed restrictions after some period of time
• Allow owners to “buy out” their deed restrictions

Personally, I favor putting time limits on deed restrictions. This comports with the longstanding jurisprudential “rule against perpetuities”: no generation should be allowed to lock up the future. And unlike entirely canceling or banning deed restrictions, it allows most of their positive features – especially predictability in the development process – to work while eliminating most of their negative features (locking in land uses that have become sub-optimal).

This argument made me wonder: what is actually being built out there? The Census has datasets! If you go here and click the “Housing Units Completed By Purpose and Design” link you will find data going back to the 1970s. Single-family houses peaked in 2006, at 1.6 million. After nosediving during the Great Recession, house completion recovered partially- but even in 2025 just over 1 million were completed, almost a 40 percent drop. About 18 percent of these 1 million are attached, a slight increase from 2006 (when just over 14 percent were).

By contrast, the conventional apartment complex (by which I mean units in buildings with 20 or more units) has recovered from the recession. 173,000 such units were completed in 2006, and 420,000 units were built today- even more than the 1974 level of just over 350,000. (But before you scream “Aha! We are building enough apartments!” remember that this is only a 20 percent increase, in a country where population has increased by about 60 percent).

What about “missing middle” housing? These categories of housing have nosedived over the past 20 years, to an even greater extent than single-family houses: from 41,000 to 16,000 in the 2-4 unit category, from 42,000 to 11,000 in the 5-9 unit category, and from 57,000 to 37,000 in the 10-19 unit category. For all three categories, the 2006 construction level was less than half the 1974 level, so on balance less multifamily housing was built in 2025 than in 1974.*

In particular, condo construction has simply collapsed. Americans completed 192,000 multifamily for-sale units in 1974, 127,000 in 2006, and 23,000 in 2025. By contrast, multifamily rental completions are only about 25 percent below their 1974 level (461,000 in 2025, 596,000 in 1974).

So to sum up: single-family construction, condo construction, and “missing middle” construction have not recovered from the recession. Larger-scale apartment construction has recovered from the recession but is still not at 1970s levels. This suggests to me that there isn’t really enough housing built to satisfy either family demand OR nonfamily demand.

*Some pro-natalists might think that not building apartments is a fine thing, because without it renters will be forced into supposedly more family-friendly single-family or missing middle housing. So let’s look at the five states with over 60 live births per 1000 women age 15-44: Alaska, Kentucky, Nebraska and the Dakotas. If fertility was related to apartment buildings, these would be states where the overwhelming majority of renters lived in houses and duplexes. This is certainly true in Alaska and Kentucky (where only 18 and 22 percent of renter-occupied units are in apartment buildings with over 10 units). But in South Dakota, 40 percent of renter-occupied housing units are in such apartment buildings- more than in Maine, Massachusetts, New Hampshire, Oregon or Vermont (the lowest-fertility states). The comparable numbers for Nebraska and North Dakota are 40 percent and 57 percent. So there isn’t really a strong correlation here.

But a recent paper by some engineering scholars suggests that this argument is oversimplified. They suggest that emissions from “operational impacts” (for example, day-to-day electricity and natural gas use) are more significant than construction impacts. As a result, the article concludes that in Toronto, annual greenhouse gas emissions for new high-rise multifamily buildings are on average 24 percent less than low-rise multifamily buildings.

I note, however, that the study is inconclusive in some ways- most notably because it includes a fairly small number of Toronto buildings, and (perhaps more importantly) because its high-rise category includes all buildings in the sample that are over five stories. Thus, it does not address whether there is a difference between 5-10 story midrises and much larger buildings.*

*Readers might be wondering why the authors did not find enough buildings to make such distinctions. First, the authors chose to focus on post-2009 buildings, thus limiting sample sizes. Second, the authors’ data was based on an earlier study with a limited building sample.

I discovered pretty quickly that the four most unequal large metros (measured by GIni coefficient) are Miami, New York, Los Angeles, and San Francisco: all high-cost metros. So at first glance, it thus appears that there is a strong correlation between inequality and housing prices.

But then I looked at the fifth through tenth most unequal metros, and a different picture emerged. These were, in order of Gini coefficient: Memphis, Houston, Boston, Tampa, Cleveland, and Chicago. All of these except Boston are low and medium cost metros; for example, the median home price in Memphis is under $300,000. And the differences between these metros and the high-cost metros are not enormous: San Francisco’s coefficient is .489 and Memphis’s is .486.

The least unequal metros with over a million people were mostly low and medium-cost metros,* but with a few exceptions: in Washington, Riverside and Denver the median home price was just over $600,000.

So what distinguishes the high-inequality/high-cost metros from the other high-inequality metros? As one might guess from conventional economic theory, income. The highest-cost high-inequality metros (Miami, New York, Los Angeles and San Francisco) have median household incomes over $90,000. So do Washington and Denver (although Riverside barely misses the cut at $87,000). By contrast, Memphis’s median income is just under $65,000, Cleveland’s is just under $68,000 and Houston’s is just under $80,000. In other words, high inequality plus high income = high prices. High inequality + low income = low prices.

*The bottom 10 included: Salt Lake City, Washington, Virginia Beach, Riverside, Grand Rapids, Kansas City, Minneapolis, Indianapolis, Denver, and Raleigh.