He reached this conclusion by asking a sample of 200 homeless people: “Where are you from?”  53 percent were from outside Los Angeles County, and “[n]early 40 percent” were from outside California.   Evidently, Rufo believes if you were not born in Los Angeles, you came there to gorge yourself on fentanyl or indulge yourself with social services.  The problem with this claim is that most NON-homeless people in Los Angeles County are also not born locally. According to table B05002 in the U.S. Census Bureau’s American Community Survey, only 5.05 million of the county’s 9.75 million residents were born anywhere in California, or 52 percent.   Because Rufo admits that 60 percent of Los Angeles homeless are from California, this means that homeless people are actually more likely to be from California than the citizenry as a whole.   (On the other hand, I am assuming that “from California” and “born in California” are the same thing, which may not be true.)

Moreover, as a logical matter, not being “from California” is not the same as “moving to California to use drugs without being arrested.”  If you moved to California after high school, was a stable, sober worker at first, became an alcoholic, lost your job, and then became homeless, obviously you a) are not from California yet b) did not move to California to use drugs with impunity.

Rufo is on firmer ground when he cites a Rand Corporation study showing that 36% of homeless people were not residents of Los Angeles County when they became homeless (p. 24).  Anyone who became homeless and then moved to Los Angeles for social services or warm weather would be in this category.  However, a large minority of these 36% were from other parts of California (presumably including nearby counties): only 22% of Los Angeles homeless became homeless out of state.

I have no idea how many of these 22% moved to California for the reasons suggested by Rufo.  But even if all of them did, Los Angeles still has an astoundingly high level of native homelessness.  According to a Brookings Institution estimate, Los Angeles has the fifth highest rate of homelessness among American cities (697 per 100,000 residents).  If we subtract the 22 percent, Los Angeles’s homeless rate is 544 per 100,000 residents- about 80 percent higher than the 49-city average, still the eleventh highest among 49 cities surveyed, and higher than all but five non-California cities.   Los Angeles has the third highest level of unsheltered homelessness (472 per 100,000 residents).  If you shave off 22 percent of that number, Los Angeles would have 369 per 100,000 residents, three times the 49-city average, the eighth highest rate among 49 cities, and higher than every non-California city but one (Portland).  

In other words: even if a significant percentage of Los Angeles homeless were the sort of social service tourists described by Rufo, Los Angeles would still have far more locally-grown homeless residents than less expensive cities- which supports the idea that yes, homelessness is a housing problem after all.

Instead, Hughes hypothesizes that underlying NIMBYism is in fact stronger in Europe – strong enough that it is rarely challenged and aroused. There are a lot of empirical claims here, and the story probably differs substantially country to country as it does region to region in the U.S.

It’s long past time to retire the idea that high housing costs are an Anglosphere problem. More research, commentary, and advocacy, especially from Europeans, would be a welcome addition to the global YIMBY discourse.

He acknowledges that orthodox economists believe that the law of supply and demand affects housing, and that thus more housing should equal lower prices.  But in Broken City, he argues that new housing (or maybe just new infill housing) means higher land prices and thus to higher housing costs.  His focuses heavily on Vancouver, British Columbia: he writes that “Vancouver, despite tripling the number of housing units within city limits since 1960, has been rewarded for its effort with the third highest housing prices in the world.” (p. 65).  

But Vancouver is just one example: Condon doesn’t compare it with cheaper cities.  Canada has other larg-ish cities, including Edmonton, Calgary, Montreal and Toronto.  Did those cities add more or less housing than Vancouver?  Did housing prices rise more rapidly or more slowly than in Vancouver?  Condon doesn’t address these issues.  He does mention in passing that housing prices have risen in Houston (which was allowed more housing than most North American cities),** but I am not sure he answers the question that motivates most YIMBYs: why is Houston so much cheaper than New York or Vancouver?***

Moreover, Condon’s description of Vancouver overlooks the fact that housing markets are regional and not bound by municipal lines.  What was going on in Vancouver’s suburbs? Were they building at a Vancouver-like pace, or did growth control policies limit suburban pricing and thus create a regionwide housing shortage?  Condon doesn’t address this issue.

His theoretical logic is that “When the city authorizes a doubling of market density.. the residual land price goes up in response… [thus there is] no substantial decrease in the cost per square foot of new housing.” (p. 196).  But these imaginary numbers seem arbitrary.  If density increases by 100 percent, why would land price increase by 100 percent instead of 20 percent or 50 percent?  I am not sure I understand his argument here.

And if land prices were a function of more housing construction, we would find that land prices exploded in places with lots of housing construction, but were stable or declining in places where “Not In My Back Yard” sentiment impeded housing construction.  But even a casual look at the American Enterprise Institute land cost database suggests otherwise: over the past decade or two, land costs has risen in most metro areas: not just in places with lots of new housing such as Austin, but in less permissive areas (such as San Francisco and its suburbs).

Unlike less sophisticated opponents of YIMBYism, he doesn’t argue that new housing is all bad, and he is all for housing owned by government and nonprofits.  But he also argues in favor of programs that make housing less profitable, such as impact fees, rent controls, and inclusionary zoning.  Some (if not most) economists believe that these policies discourage housing construction- but Condon argues that they make housing cheaper by reducing land costs, suggesting that if an extra tax makes a project unworkable, “the price for the land will be renegotiated down” (p. 163).  He points out that California cities “have been less timid than most states” in imposing impact fees (p. 162).  But since land and housing costs in California have exploded in recent years, this fact doesn’t really support his argument.

 He also praises inclusionary zoning because it encourages the creation of nonmarket housing and “lower[s] the prices of developable parcels” (p. 165)- in other words, reduces the demand for land.   In particular, he praised Portland, Oregon’s expansion of inclusionary zoning.  But after Portland passed its inclusionary zoning ordinance in 2016, land costs in Multnomah County (which includes Portland) increased from $1.06 million per acre in 2016 to $1.56 million in 2022.     Meanwhile, building activity crawled to a halt.  In 2017, almost 6000 permits were issued for Portland buildings with five or more units.  In 2024, fewer than 1000 such permits were issued.

He also praises inclusionary zoning in Cambridge, Mass.- where inclusionary zoning has apparently been far less harmful. But Cambridge’s program is far less restrictive, because inclusionary zoning is part of an overlay district- which means that landlords have the option of proceeding either under inclusionary zoning (which allows more units in exchange for more “affordable” units) or proceeding under the prior zoning code.

Similarly, he argues for rent control on the ground that it “quell[s] the appetite of international [investors] for a city’s land” (p. 142)- in other words, makes people less interested in being landlords, which supposedly will reduce land prices.  In New York, where I live, the state government implemented vacancy control, where rents don’t go up even when a tenant leaves.  As a result, landlords have warehoused thousands of rent-stabilized apartments, apparently because they would have to spend money on renovations to comply with housing costs, and are not allowed to raise rents enough to pay for the renovations.  Rather than declining, the median asking rent in each of New York’s five boroughs has increased by 50 percent or more between 2019 and 2026.

And as a broader theoretical matter, his relationship with the law of supply and demand seems inconsistent: on the one hand, he argues that housing supply doesn’t make housing cheaper, but on the other hand he argues that extra taxes reduce the cost of land by reducing demand for land.   

*For those of you who have not read this blog, YIMBY means “Yes In My Back Yard”, a slogan adopted by those of us who support expanded housing supply.   

**One might ask: if YIMBYism works, why have housing prices risen in Houston? But this question is easily answered: Houston has gained population far more rapidly than high-cost cities, leading to a demand explosion.     But even with this demand explosion, rents have risen more slowly there than in cities that have allowed less housing.  Nominal rents in Houston are about 3.29 times their 1980s level,  while rents in metro New York are five times their 1980s level.  

***Condon hints at an answer in his historical discussion: he points out in Chapter 4 that as American densities nosedived in the mid-20^th century, land costs went down.  So perhaps he believes that sprawl equals low land costs and density equals high land costs.  But Condon admits that densities in the English-speaking world are still very low compared to middle-income countries (p. 93)- so why hasn’t low density continued to bring land costs down?

Person A- infill is cheaper than sprawl because no one has to build new sewer lines!

Person B- Sprawl is cheaper because its more expensive to build in the city!

Person B’s post made me wonder: is suburbia always cheaper?

I decided to do a rough comparison: the cheapest new city houses vs. the cheapest new suburban houses of the same size. I went on Zillow and searched for houses that were a) built after 2020 and b) 2000-2500 square feet.

In some metros (even fairly inexpensive ones), suburbia clearly did seem to be cheaper. For example, the cheapest new house in Chicago started at around $600k, while similar options in Lake and McHenry Counties started at around $360-370k. In St. Louis, the cheapest new home cost around $450k, about a third more than the cheapest new houses in St. Louis and St. Charles Counties.

On the other hand, I found counterexamples. The cheapest new house in Atlanta cost $290k, about 15 percent below its counterparts in Cobb and Gwinnett Counties. The cheapest new house in Houston cost $225k, about the same as its counterpart in suburban Fort Bend County. The cheapest in Cleveland cost $288k, about the same as its counterpart in the suburban part of Cuyahoga County, slightly less than its counterpart in outer-suburban Lake County, and about 25 percent less than its counterpart in outer-suburban Geauga County. In Philadelphia, the cheapest new house is about $350k, about 10 percent lower than its counterpart in outer-suburban Chester County.

All of this is subject to two qualifications. First, this is not an academic “everything being equal” comparison: all I am trying to show is that it seems possible to build urban and suburban houses for similar prices. Second, it may be that at the true outer edges of suburbia, houses are cheaper than in the suburbs I have mentioned.

In a new video, Aevy TV presents CEPT University research with an impressive mastery of the technicalities. A key point – and this connects to Kaur & Anand’s work – is that the large setback rules imposed by Indian regulations make land use extremely inefficient. Instead of public unbuilt space (for roads, parks, and plazas), Indian cities have shocking amounts of private unbuilt space, which by all accounts is poorly used.

The content bounces around a bit (the rent control and government management tangents seem out of place), but it’s a great place to start if you want a deeper understanding of Indian cities than you’d get from movies or tourism.

Returns to Political Contributions in Local Housing Markets
by Rui Yu
Review of Economics and Statistics, forthcoming

Abstract:
This paper examines how politically connected firms shape housing supply in U.S. cities. Using new data on campaign donations to U.S. mayors and a regression discontinuity design, I present three findings. First, developers connected to the mayor sell more new housing units. Second, more sales of new housing by connected developers coincide with higher local housing supply: cities where mayors received more developer donations issue nearly 70 percent more permits for new housing units. Third, differences in mayors’ pre-existing policy stances — rather than connections to developers — is a quantitatively larger determinant of local housing supply.

This is a reasonable critique and I’ve never used Paris as a model for this reason. Being allowed to build six-story buildings isn’t going to make your city Paris any more than having free speech is going to make you Marcel Proust.

But let’s flip the lens around: What about those buildings Collison chose to exemplify “what is actually on offer”? That’s the Batik Apartments in Seattle, completed in 2018. Strong YIMBY policy have made Seattle a renter’s market:

I agree that the Batik is pretty ugly. I count at least 7 distinct exterior materials interspersed even within a very small area. Why? It’s not the cheapest option, which would be to stick with one, maybe two finishes for simplicity. Nor are the various small protrusions, each of which adds cost and reduces resilience.

So why was it built like this? It could be that customers want this and they just don’t share my preference for simplicity. It could reflect the artistic values of the architects, who seem to be very proud of it.

A third possibility is that the Batik looks unique in exactly the same way every other new building is unique because it had to clear Seattle’s famously strict design review (which has since been pared back by YIMBY-led state laws). Here’s the developer’s 81-page response to the city’s Early Design Guidance (EDG) Report feedback. (The city has a helpful website outlining the 22 steps to submit plans for EDG, which of course is just one of many approvals before construction may begin.)

The EDG response has many renderings of the structure and lovingly presents this awful palette of colors and textures.

Could Seattle architects have legally built classically-styled apartments? I can’t find a single modern example. There’s a good chance that many neighborhood Design Review Boards would have rejected a Haussmann-style building (or even a copy of the Cobb Building) as “pastiche”. In the silly world of contemporary architecture, it’s dishonest to pretend your building is older than it really is, but mandatory to “break up the massing” and pretend your building is several different buildings. The obsession with articulation, multiple materials, and “doodads” yields a style I call “design review rococo.”

American building without design review

So what would American architects build if it was just between them and the market? Surprisingly few big cities allow us to test this idea. Even fast-growing Sunbelt cities like Miami, Austin, and Phoenix usually have design review. Thankfully, there’s Houston.

And…it looks pretty similar. The colors are perhaps a bit more muted on average, but that probably reflects the different climate. Houston builders presumably aren’t wasting their money on architects’ whims. The people with skin in the game believe that breaking up the massing and using mixed materials, although it adds costs, attracts more renters.

Personally, I love the exception among the Houston photos – the Marquis Enclave. It’s an industrial loft imitation complex which covers seven entire blocks and has created an excellent shady streetscape at the edge of the Fourth Ward:

But if you look at the Marquis Enclave’s advertising, it nowhere mentions the architecture or public realm. Instead, the word “modern” shows up four times on the recruitment page.

Aesthetic NIMBYs

If this casual analysis is correct, then Americans don’t care very much about exteriors. And to the extent that they care, they favor midrise buildings that Collison and I find ugly. I would very much like to be wrong about this!

Practically speaking, what it means is that Collison’s core point is probably wrong: offering Americans a product that he and I like better isn’t going to make them favor upzoning.

[Following up on a neighbor’s complaint,] Building Commissioner Dennis Manuwal reviewed the property and confirmed the shed, which sits just two feet from the property line, was in violation of city codes. However, while inspecting the Emmons’ property, Manuwal drove through the rest of the subdivision and observed numerous similar infractions.

During a Board of Zoning Appeals (BZA) meeting on Monday, Plymouth Plan Director Ty Adley reported the scope of the issue. Of the 31 sheds located within the South Pointe Subdivision, only 13 currently meet setback requirements or have previously obtained a variance. The remaining 18 sheds are situated too close to the side yard, rear yard, or both.

The subdivision in question has just 41 houses. Almost half are out of compliance. The obvious solution would have been for Building Commissioner Manuwal to keep his observations to himself. A more sophisticated solution would be for the city to reduce its onerous 20-foot rear setback for accessory structures, which obviously don’t correspond to the normal practice.

The city chose neither of those. Instead:

Homeowners must now either physically relocate their sheds to comply with the 10-foot side and 20-foot rear setback requirements, or they must officially apply for a “variance of development standard.”

To obtain a variance, residents will need to present their case to the board and explain specifically why their shed cannot be relocated to meet the standard setbacks. Adley emphasized that there is no blanket solution, and the board will consider each variance application individually on its own merits.

There is a blanket solution – relax the zoning – but the city officials are too committed to their codes to be responsive to their citizens.