Should we bribe the NIMBYs?

Why some YIMBYs tolerate a tax on housing

10 years ago, Bill de Blasio’s New York City adopted new inclusionary zoning requirements. Turns out rent is still expensive there. (Photo by Andy Katz/Pacific Press/LightRocket via Getty Images)

Everyone hates luxury development. But we need fancy new apartment buildings to fix the housing crisis: When rich people can move into new units, then non-rich people can move into the older homes that the rich move out of.¹

Despite the truth of this, voters just do not believe that more building alone will make housing more affordable. And many politicians want the least fortunate Americans to see more obvious benefits from homebuilding.

One solution is to force developers of fancy new apartment buildings to set aside some of those homes for low- or moderate-income people. This policy is called inclusionary zoning.

Starting with Montgomery County, Maryland, in the 1970s, inclusionary zoning policies have been adopted by most of America’s major cities: New York, Boston, Chicago, San Francisco, Seattle, Denver, Minneapolis, and Washington, D.C., all have some version of the policy in place.

Inclusionary zoning’s logic is simple enough: Some people will always make too little to afford market rents, and we don’t want to lock poor people — or their children — out of the benefits of living in cities and towns that have the best economic opportunities.

But economists and real estate developers counter that inclusionary zoning is functionally a tax on development, because it lowers the profit margins on building more units. That means fewer units get built overall. And, while a small number of lucky tenants get new affordable housing, overall housing costs are pushed higher than they otherwise would be, absent the inclusionary zoning requirement.

In other words, if we’re going to subsidize housing for the poor, inclusionary zoning is a really inefficient way to do it.

Just how much housing does inclusionary zoning exclude?

Figuring out exactly how much inclusionary zoning reduces housing supply growth is harder than you might think. After all, cities that impose inclusionary zoning mandates may differ from those that do not in ways that also affect the production of new housing.

Perhaps cities with inclusionary zoning are also wealthier, or more liberal, or have a higher cost of living. That makes it difficult to compare like with like and tease out the effect of inclusionary zoning alone.

But a new paper — “Inclusionary Zoning and Housing Supply: Evidence from California’s Palmer Fix” by Noah Kouchekinia, an economics Ph.D. student at the University of California, Irvine — has found a way around this challenge.

In 2009, a California court ruled that local inclusionary zoning ordinances in Los Angeles were a form of rent control rendered illegal by a 1995 state law. This made all local inclusionary zoning ordinances on rental units in the state unenforceable. In 2017, the California legislature passed a “Palmer Fix” law, which allowed localities to reimpose inclusionary zoning ordinances that were still on the books.

This allowed Kouchekinia to isolate the effects of inclusionary zoning ordinances using a difference-in-differences technique.²

The 2009 court ruling and the subsequent “Palmer Fix” provide as good a setting for difference-in-differences as you’re likely to find in the real world: Prior to 2009, some cities in California had inclusionary zoning and others did not. From 2009 to 2017, no cities in California had inclusionary zoning for rental units. Then, from 2017 onward, inclusionary zoning was “turned back on” in the cities whose ordinances came back into effect.

The next step is to find a way to measure the stringency of the inclusionary zoning ordinances. Usually, inclusionary zoning rules give developers a menu of options. For example, a developer may have a choice between setting aside 10% of units for tenants who make up to 60% of the area median income or setting aside 15% of units for tenants who make up to 80% of the area median income. Either way, the developer must forgo some share of the rent they would earn if all units were rented at market rates.

Kouchekinia assumed that developers are profit-maximizing, and he therefore measured the stringency of inclusionary zoning ordinances as the minimum percentage of market rents that they would have to forgo to be in compliance. He then estimated the effect of an increase in stringency on the supply of new housing.

He found that “for every percentage point increase in the stringency of an inclusionary zoning ordinance, the flow of new housing supply decreases by 7.6%.” Since the typical ordinance “requires a developer to forgo about 4.2% of rent,” he estimated that the typical inclusionary zoning ordinance reduces annual housing production by 31.8%.

Back-of-the-napkin math³ suggests that, from 2018 to 2024, that amounts to just over 86,000 fewer units built in the state of California because of inclusionary zoning.

Kouchekinia emphasized in an email that his paper was “a work in progress” and that, while he had presented it at conferences for peer review, he has not yet released a polished version as part of a working paper series.

But his results align with prior literature on inclusionary zoning, which employs a more theoretical approach. A 2024 UC Berkeley Terner Center report simulated the effects of inclusionary zoning in Los Angeles over a 10-year time frame. It found that, as the inclusionary zoning share increases from 0% to 1%, the number of total units built decreases sharply. Then it falls more gradually and almost linearly at rates beyond that.

Another paper from Vincent Rollet at MIT built a model of supply and demand for floor space in New York City and simulated the effect of a requirement that 20% of floor space in new buildings in New York be rented at below-market prices. Rollet found that, from 2020 to 2060, the inclusionary zoning mandate would induce “the construction of 73 million sq. ft. of affordable floorspace, but lowers the total amount of floorspace in the city by 400 million sq. ft. In other words, for each sq. ft of affordable floorspace created by the policy, the overall available floorspace drops by 5.5 sq. ft.”

What happens when you secure new affordable housing units by deliberately forgoing a much larger amount of housing at market rates? Rents increase. If demand is unchanged and supply contracts, then markets clear at higher prices.

The anti-market delusion at the heart of the housing crisis

“Kouchekinia’s paper is complementary” to the prior literature, according to Chris Elmendorf, a professor at the UC Davis School of Law who studies housing policy. While “we should expect mixed effects depending on the policy specifics,” he added, the common finding is that inclusionary zoning reduces the supply of housing available to consumers overall. “There’s no such thing as a free lunch.”

We have real-world examples that appear to support the more theoretical literature as well. Portland, Maine, expanded its inclusionary zoning mandate to 25% of units in new buildings in 2020; housing production has since collapsed, and the city council is looking into rolling it back.

Portland, Oregon, adopted a similar mandate in 2017. Construction plunged until the city council voted in 2024 to expand local tax exemptions and increase financial subsidies for projects subject to the law.

Getting voters to yes (in my backyard)

Recently, Cambridge, Massachusetts (where I grew up), and New Haven, Connecticut (where I went to college), passed massive upzoning bills — a recognition in these deep-blue jurisdictions that capitalism is going to play a major role in easing their housing woes.

Last year, the Cambridge City Council passed a bill legalizing units up to six stories by right (meaning without the special permits previously required) on large-enough lots in areas with inclusionary zoning. This means that you can now put an apartment building up to 74 feet tall in places where previously only standalone low-density units were allowed. In New Haven, the Board of Alders recently passed a bill allowing for much larger and denser housing construction downtown.

Both cities have inclusionary zoning ordinances on the books. In Cambridge, 20% of floor space in a new development of 10 units or more (or at least 10,000 total square feet) must be sold or rented at below-market rates. In New Haven, at least 15% of units in new downtown developments of 10 or more units must be sold or rented at below-market rates.⁴

Both Burhan Azeem and Eli Sabin, the Cambridge city council member and New Haven former alder who championed their cities’ respective reforms, said that they wouldn’t have been able to build the political coalitions for their upzoning laws without those inclusionary zoning ordinances on the books.

“Without inclusionary zoning of anything, the economics would be a lot better, and we’d have a lot more housing development in total,” Azeem told me over the phone. “But that’s not the world we live in. We live in almost the opposite world, where new housing production is incredibly unpopular, where 85% of the existing buildings were noncompliant, where Massachusetts and Cambridge in particular were zoned for less density than already existed.”

“In a world where five or six votes on a nine-member council is what you need, telling someone that, in this new project, there’s going to be at least one unit that’s affordable to a teacher was a really persuasive argument to get them to see the overall benefits of the upzoning bill,” Azeem added.

Sabin noted that, while he feels that skepticism of the short-term benefits of new market-rate housing is justified, inclusionary zoning might not be the most efficient way for a city to provide housing to low-income residents. That’s because if fewer units are built overall, as the literature suggests, the city government receives less in tax revenue.

“There is a counterfactual that should be considered, which is how much revenue the city would have if there hadn’t been the inclusionary zoning tax, and whether you would be able to spend that amount more efficiently to help families afford to live in the city,” he told me. “I just think that more transparency about what we’re spending and having a more honest picture of what choices are hidden within the inclusionary zoning — that would be valuable.”

Efficiency promotes equity

So far, we’ve been talking about a particular kind of inclusionary zoning ordinance: unfunded inclusionary zoning. An unfunded mandate is when the government forces you to do something but doesn’t provide any funds with which to do it. In this case, the government is demanding affordable units and expecting landlords to pay the difference between market and affordable rents.

With funded inclusionary zoning, landlords receive some sort of financial compensation for the rent they forgo on the affordable units in a building. That compensation could take many forms: In principle, the government could just write landlords a check for the difference between the market rent for a given unit and the amount tenants are paying.

In practice, the compensation usually takes the form of reduced development fees or property tax abatements — something along the lines of how Portland, Oregon, tried to fix things when its inclusionary zoning backfired.

In 2025, Washington state passed HB 1491, which set new statewide minimum density standards near transit stops. The law:

• Imposed an inclusionary zoning requirement on new developments.

• Required that cities with development impact fees reduce them by 50% for qualifying projects.

• Provided a 20-year property tax abatement for new developments with affordable units.

Inclusionary zoning policies can be made less harmful, but “less bad” isn’t the same as “good idea.” Inclusionary zoning is a very kludgy and inefficient way to provide housing to those who cannot afford to pay the market price for it. A more efficient way would be to just lower the market price by flooding the city with new units.

The best way to do this would be to have no inclusionary zoning ordinances, drastically reduce zoning regulations in general, and let builders build up to the point at which housing markets clear. Then, the government would use some of the increased property tax revenues to pay for rental subsidies for low-income families — perhaps with larger subsidies allotted if a qualifying family chose to live in a high-opportunity (and commensurately higher-cost) neighborhood.

Unfortunately, we live in a second-best (or third-best or fourth-best) world. To the extent that voters and interest groups may need to be bought off to get new housing built, (funded) inclusionary zoning may be worth considering. It’s better to pad new construction with what are effectively bribes than to do no new building at all.

But we should be clear-eyed about what inclusionary zoning is: It’s a tax on new housing. And when you tax something, you end up with less of it.

Recommended reading:

1. You might even say that the fancy new units “trickle down” to non-rich renters.

2. Ideally, the way we would assess the effect of a policy is to randomly assign the policy to different localities. Of course, in the real world, we don’t do a random assignment of policies.

The idea behind difference-in-differences is that we can approximate randomization by comparing otherwise similar cities with and without inclusionary zoning ordinances. We can do this by confirming that the cities showed “parallel trends” on our variables of interest before one adopted the policy. That way, we can isolate any subsequent difference in outcomes as a result of the policy and not other factors.

3. Using the permitting data from Table 2 and the coefficients from Table 4, column 2.

4. Of that 15%, 10% is set aside for people earning at or below 50% the local median wage and 5% is set aside for Section 8 recipients.