The Case Against The Case Against YIMBYism

Last month, I attended YIMBYtown, a semi-annual gathering of YIMBY advocates and activists from all over the country.1 I came away floored by the movement’s maturation: heavy-hitters from both parties addressed the conference, and YIMBYs from a number of states came bearing news of major victories. One panel was all about Montana’s recipe for YIMBY success; another focused on Washington State’s slew of pro-housing legislation. I’m proud of what YIMBYs have accomplished in California, but other states are giving us a serious run for our money.

Others were less impressed. Case in point, writer Michael Friedrich, who used YIMBYtown 2024 as the news peg for his “The Case Against YIMBYism” in The New Republic (where I am also an occasional contributor).

What is the case against YIMBYism? In Friedrich’s view, it’s that there is no evidence that YIMBY policies will actually correct the housing shortage2 and lead to broad-based housing affordability. “Ultimately, the solution to the housing crisis is political and economic, not a matter of technocratic tweaks,” Friedrich writes. “Zoning reform is neither necessary nor sufficient to fix it.” Instead of making it legal for private actors to construct more homes, he proposes that our energies would be better spent pushing for public housing development at an unprecedented scale.

Friedrich did not attend YIMBYtown in person, but he did tune into the livestream and interviewed some of the attendees, including myself. And to his credit, he gave me ample time to respond to his criticisms of YIMBYism in our two interviews. Not much of what we discussed is reflected in the article, but that’s how it goes when one has a deadline and a limited word count.

One of the advantages of writing a free newsletter is I am unconstrained by either deadlines or (perhaps to my detriment) word counts. So I’d like to take this opportunity to address Friedrich’s argument at length. Needless to say, I think his assessment of the evidence for YIMBY policies, his political analysis of U.S. land use politics, and his proposed alternative to YIMBYism are all deeply flawed.

The Uses and Misuses of Academic Research

Let’s start with Friedrich’s review of the evidence for and against land use reform as a mechanism for bringing down prices. Friedrich argues that there is little evidence upzoning leads to more private homebuilding, and, further, that there is little evidence more private homebuilding will drive down housing costs. But his case is based on a selective reading of the relevant literature that leads him to misinterpret some findings while downplaying others.

For example, Friedrich correctly notes that there is limited research on the long-term impacts of major YIMBY reform. While some of the early indicators from Minneapolis and Auckland have been promising, we simply don’t know for sure whether land use liberalization has put those cities on a path to housing abundance.

That is only to be expected. Even if the Minneapolis and Auckland reforms “worked” — which is to say, even if no further reform is needed — it will be some years before homebuilding in both cities catches up with demand for housing. And it will probably take even longer before the academic literature catches up to real-world conditions and successfully isolates the impact of land use liberalization in each city.

That isn’t an argument against YIMBY reform. Instead, it is a reminder that the academy can only do so much to guide policy. We simply don’t have time to wait around for conclusive findings from Auckland and Minneapolis before we tackle the housing crisis in other cities. And if such findings were a precondition for action, then Auckland and Minneapolis would never have done anything in the first place, and we wouldn’t even have the initial results from their experimentation with YIMBY policies.3

Where the academic research on a particular course of action is incomplete or ambiguous, policymakers need to make use of what evidence is available. And the evidence in favor of land use reform is considerable: while we’re still waiting on verifiable examples of total victory in cities that have implemented YIMBY policies, there is plenty of research pointing to a relationship between land use reform and improvements in housing affordability. More on this evidence a bit later.

Further, we have a rich array of anecdotal evidence to go off of: the U.S. cities with the worst housing crises, such as San Francisco, Los Angeles, and New York City, tend to be places that significantly downzoned in the latter half of the 20th century and obliterated a massive amount of their low-cost housing stock through urban renewal. If tighter land use controls and widespread housing destruction preceded the modern housing crisis, it stands to reason that looser land use controls and widespread housing construction will reverse the housing crisis.

Then there are the real-world examples: Houston is a fast-growing city that lacks a zoning code and has largely managed to avoid the cost crunch of other urban areas with expanding populations. Casting our gaze further abroad, we find that Tokyo’s flexible zoning rules have allowed it to consistently expand its housing supply and keep pace with a growing population.

As for the academic literature, while it may have gaps, the overall thrust of the research is clear: removing statutory bans on homebuilding tends to lead to more homebuilding (no surprise there), and more homebuilding tends to result in lower prices. Friedrich cites a couple of studies to this effect but complains that they find only very modest results. Referencing Xiaodi Li’s 2019 paper on apartment construction in New York City, he notes that it found rents decreased only 1 percent for every 10 percent increase in housing supply.

But Friedrich misunderstands the significance of the findings. Li’s study focuses on the effect of new market-rate construction on rents within 500 feet. To assess neighborhood-level effects, you need to isolate them from the dynamics of the larger metropolitan housing market. That’s in the nature of the type of research that Li is doing here. But there’s no such thing as a housing market contained to a 500-foot radius. Instead, housing markets, like job markets, and like urban economies writ large, tend to cover entire metropolitan areas.

To put things in more concrete terms: Li’s findings imply that building a new apartment complex in Bushwick will modestly reduce rents within a city block radius. If you build another apartment building a few blocks away, that will also probably ease demand pressures on your block of interest — but Li’s research does not observe that effect. And her paper certainly doesn’t tell us how building an apartment building in Williamsburg effects rents on our block in Bushwick. But if at least some of the demand pressure in Bushwick comes from people getting priced out of Williamsburg, then presumably homebuilding in Williamsburg can drive down rents on our block.4

As for Li’s measured effect size, Friedrich over-interprets the results. “Even if developers somehow built 50 percent more housing in New York City, the median one-bedroom unit would still rent for $3,548 per month (if applying the study’s findings to today’s market,” he claims. That’s wrong on two counts: First, it extrapolates a study on neighborhood-level effects to the city level, which you can’t actually do for the reasons I’ve already discussed. Second, it takes the measured one percent effect size too literally. Li’s research, while valuable, did not uncover some hidden law of Newtonian physics, where a ten percent increase in housing stock will always lead to a one percent reduction in rents. The most useful components of her findings — as in all similar research — are the directionality and the significance, not the effect size. The strongest claim we can make based on her work is that there is a statistically significant relationship between homebuilding and rent decreases within a 500 foot radius, and it appears to be causal. That’s pretty good!

Friedrich makes a similar error in his gloss on a 2023 Urban Institute paper examining the relationship between land use reforms and housing costs. He writes: “The one major study of land-use reforms across the United States shows that during the three to nine years after a city enacts upzoning and other density-increasing measures, new housing supply increases by less than 1 percent.” Again, the important thing is less the effect size than the identified relationship (land use reforms lead to more housing supply) and the confidence interval (statistically significant at p ≤ 0.01).

Similarly, Friedrich does not note the limitations to this study, which are considerable. The challenges start with the paper’s scope of observation: the authors aim to observe the effect of land use reforms over no fewer than eight metropolitan areas, including the regions around Los Angeles, Chicago, Dallas, Philadelphia, and Miami. These are all very different regions, with wildly disparate histories, demographics, housing markets, regulatory regimes, geographies, and built environments.

Land use is simply going to mean very different things in different contexts: a city that already has very permissive zoning might not see much change in housing supply if you upzone further, whereas a city with very strict zoning might see an immediate supply response following even modest tweaks to the zoning code. Similarly, developers might not build much new housing in an area with a collapsing population even if you make it easier for them to do so. Trying to average out the effect of land use reforms over these jurisdictions can only tell us so much.

Then there’s the paper’s variable of interest: land use reform, broadly defined. In order to get a statistically significant result, the authors lumped together a number of different reform types, ranging from raising height limits to encouraging accessory dwelling unit development. As they themselves write: “We also do not differentiate between relative impacts of different changes, and we do not have the power to assess the varying impacts of reform types, like ADU or height-limit policy.”

So a change in the height limit of, say, ten feet gets treated the same as a twenty foot increase. Similarly, a bill legalizing mixed-use development in certain areas gets treated the same whether or not it gets loaded down with poison pills that render the policy unworkable. At such a high level of abstraction, it’s simply impossible to say much about the impact of land use reform, except the following: there is evidence that land use reform, broadly speaking, is associated with more homebuilding, broadly speaking.5

While Friedrich downplays the evidence that pro-housing policy improves housing affordability, he gives more credence to the argument that market rate development actually makes housing less affordable for low-income renters by driving up rents on the lower tier of the market in neighboring buildings. He cites this paper out of Minneapolis, but doesn’t note that its findings are out of step with the overall body of research on this topic. As Shane Phillips, Michael Manville and Michael Lens point out in this handy literature review, that may be because the paper’s authors don’t adjust their rent data for inflation — an “unusual decision,” as Phillips et al dryly note. Once one adjusts for inflation, the rent increase on the lower end of the market basically disappears.

The Phillips et al report is worth reading in full because it summarizes a number of other influential papers on the relationship between market rate development and housing costs at the neighborhood level, including a few that escape Friedrich’s notice. Phillips, Manville, and Lens focus on the neighborhood level because, as they observe, “whether new housing can reduce housing prices overall … isn’t really in doubt.”

Dubious Alternatives

What of Friedrich’s counterproposal? After mischaracterizing the evidence on market-rate construction and housing costs, he suggests that housing advocates should instead focuse exclusively on non-market solutions: massive federal investment in public housing, rent control, and other tenant protections.

It is puzzling that Friedrich frames market and non-market policies as being fundamentally in tension when most YIMBYs support a mix of the two. Indeed, the organization I work for, California YIMBY, has sponsored social housing legislation and an amendment to the state constitution that would make it easier to build subsidized housing in exclusionary jurisdictions. We’ve also supported statewide rent stabilization and just cause protections, as well as another constitutional amendment that would make it easier to raise public revenue for subsidized housing.6

While market-rate housing can adequately serve most residents in a healthy housing market, Friedrich is correct that it won’t necessarily serve people with extremely low incomes. That’s where public subsidies come in: even as increased homebuilding causes rents to fall, we should target very low-income households with a combination of rental vouchers and publicly subsidized units. Land use reform is a necessary complement to the subsidies, because it allows us to extract the maximum value out of each dollar spent housing the needy.

Think of it this way: in City A, which has plentiful apartments that rent at $1,000 per month, you can pull someone out of homelessness, give them a rental voucher, and get them comfortably housed for $12,000 annually. But in City B, where the average market-rate studio apartment rents at $3,000, the cost of housing that same individual commensurately rises threefold. If you have any sort of budgetary constraints on your rental voucher program, that means you can only serve one-third as many residents in City B as you can in City A.7

Similarly, public funds for homebuilding don’t get anywhere near far enough in high-cost cities with restrictive land use regimes. Partially thanks to high land costs and complicated permitting rules, it can cost up to $1 million to develop a single unit of low-income housing in California. Again, the more it costs to build a single one of these units, the fewer people you’re actually able to get housed.

Friedrich would no doubt reply that the federal government should simply spend more money and build more housing on its own. After all, unlike cities and states, the feds are able to run a deficit. His model is Vienna, where more than half of residents live in subsidized housing.

It is interesting to note that Friedrich does not subject the Vienna model to the sort of scrutiny he reserves for the research on private housing development. What’s more, he doesn’t leave us with any understanding of how the United States can get from here to there. I would not recommend a one-to-one reenactment of the process that led to Vienna’s large social housing sector, which indirectly profited from the mass wartime destruction of existing housing stock and the expulsion or murder of many Viennese residents.

Fortunately, Friedrich does not think we need to level American cities or ship people like me off to the death camps in order to achieve a Vienna-style result. But his own strategy is not particularly compelling. He writes:

I am all for repealing the Faircloth Amendment and building more public housing. I am not, however, at all convinced that “agitating to repeal the Faircloth Amendment” is going to yield results anytime in the foreseeable future, given the state of our national politics. Nor does it get any easier from there. There has been no point in the history of the United States where more than a small fraction of our low-income population has resided in below-market housing, let alone public housing. But now we’re going to convert the majority of all urban housing to public ownership? How long is this supposed to take, and what are the intermediate steps supposed to be?

By all means, we should support efforts to repeal the Faircloth Amendment and build more social housing in the United States. But I remain skeptical that social housing as a model will ever scale up to represent a large share of all U.S. housing — certainly not within anything approaching a reasonable time frame. And even if I’m wrong, that doesn’t imply we should abandon efforts to get more private housing built in the meantime. After all, we already have a political formula for doing that; we know which political levers to pull; we have a large and growing cross-class, bipartisan coalition behind us. And we can see it already working. Giving up on all that in favor of Friedrich’s sketchy, speculative strategy would be not just imprudent but actually irresponsible.

Skipping Class

Hiding behind Friedrich’s argument is a reductive, Manichaean class analysis. In the moral universe implied by his article, there are basically two camps: On the YIMBY side, you have for-profit developers, the real estate industry, tech bros, and right-wing think tanks.8 On the anti-YIMBY side, you have “obstructionist homeowners, anti-gentrification politicians, and low-income tenant organizers.”

Because a lot of the people who identify with Friedrich’s argument are self-described socialists, it’s somewhat ironic that this schema doesn’t make sense when you start digging into the interests of the relevant actors. I’m just a bourgeois shitlib, so far be it from me to accuse people of being insufficiently Marxist; but I do think the conversation around class and YIMBYism could use fewer vibes and a little more dialectical materialism.

Taking material interests more seriously would lead one to acknowledge that “real estate” is not a coherent actor with a single set of motivations. Realtors, private equity firms, large landlords, small landlords, large developers, and small developers all have divergent interests that sometimes clash; even that list of players is a simplification. Thinking in these terms gives us a richer understanding of how class politics intersect with housing policy.

Consider the difference between a company that primarily earns its revenue from collecting rents on existing properties and another company that makes money by constructing new rental units. The corporate landlord and the corporate developer in this scenario actually have diametrically opposed interests: the more apartments the corporate developer builds in a particular neighborhood, the less the corporate landlord can charge for rent. Artificial housing scarcity is a de facto subsidy to predatory landlords. If you look at filings from major property management firms, they will sometimes straight up admit this.9

Similarly, different types of market-rate developers will have different interests. The big players in a city like San Francisco or Los Angeles are those who know how to play the game according to the existing rules: they have the experience, the money, and the political influence to make a decent mint in cities governed by NIMBY land use rules. Zoning and permitting reform is actually against the interests of these developers, because it changes the rules to let a whole bunch of smaller, more flexible and creative players into the market.10 Big firms in an oligopolistic market don’t like to generate competition for themselves.

I could go on, and maybe at some point I’ll write a longer piece about the class dynamics of American land use. (You can read my first treatment of the issue at The Nation.) For now, suffice it to say the premise that “real estate” is a monolithic, pro-YIMBY actor doesn’t really pass the smell test.

Conclusion

Friedrich quotes one of his fellow anti-YIMBYs as saying that YIMBYism has always been “a promise that we didn’t need to redistribute anything.” As should now be clear, this is incorrect. More homebuilding is inherently redistributive: by driving down rents, it effectively moves wealth from rentiers to renters.

Very few YIMBYs would argue we should stop there. We have a public track record of pushing for other redistributive policies — not just posting about them, but putting real organizational muscle behind turning them into law. We just tend to think that land use liberalization is one of the main areas where we can make progress to greater housing affordability, and that other approaches won’t work especially well in its absence. The evidence is on our side here.

The purpose of this post isn’t to beat up on Friedrich. Despite his omissions and elisions, I think he’s trying to approach this issue in good faith; he certainly spent a lot of time hearing me out before he ran his piece, which is more than I can say for various other authors working in his genre. I clearly failed to convince him the first time around, and I doubt he will find this post any more convincing. Regardless, I felt it was important to correct the record.