Infrastructure

Leveraging LIHTC for Housing Abundance

Realigning federal subsidies will increase housing supply in America’s big cities
January 21st 2025

Introduction

Americans are riled up about the cost of housing. For the first time in living memory, housing affordability figured prominently in a presidential campaign — and for good reason. Research shows that housing supply in U.S. metro areas has become progressively less elastic over time. This means that when demand for housing grows — as occurred during the pandemic, when people started working from home and wanted more living space — it causes higher housing prices rather than the development of more housing.

Presidential candidates and members of Congress have floated various ideas for increasing the supply of housing, but to date, none of their proposals have addressed the local political barriers that stand in the way. In big, high-demand cities, a proliferation of interest groups seeks payoffs and in-kind benefits whenever new housing is proposed. Their demands raise the cost of development and privilege politically connected developers over firms that could produce new housing at lower cost. 

This white paper suggests a fix for the dysfunctional urban politics of housing: Congress should modify the Low Income Housing Tax Credits (LIHTC) to make projects in big, expensive cities ineligible for LIHTC subsidies unless the city opts into a pro-housing regulatory framework.1 To retain LIHTC eligibility, big cities would have to accept limits on fees and price controls on new multifamily housing; review housing development proposals “ministerially,” that is, just for compliance with objective standards; and allow reasonably dense housing to be built on commercial corridors. These rules would apply to all housing development in the city, not just to LIHTC-subsidized projects.

The new conditions on LIHTC eligibility would change the urban politics of housing. Affordable housing developers and advocates for poor people would lobby cities to opt in, even as other groups that benefit from the status quo push back. For the typically progressive politicians who serve on big-city councils, it would be tough to say “No thanks” to LIHTC. Council members who voted to turn down the money rather than accept pro-housing policies would be attacked for throwing poor people and affordable housing developers under the bus — an inviting line for challengers in heavily Democratic cities. Though large, expensive cities are overwhelmingly Democratic, our goal is nonpartisan: LIHTC-based housing supply incentives can only influence the political economy of housing supply in a jurisdiction that values access to LIHTC.

The big-city LIHTC conditions we propose are not radical. They draw on recent legislation enacted on a bipartisan basis in states as different as Florida, California, Montana, Colorado, and Massachusetts. And they give effect to the “sense of Congress” expressed in the pending Affordable Housing Credit Improvement Act, which calls on the House and Senate to “develop incentives within the affordable housing credit program to… reform burdensome land use and zoning regulations.”2 

Cities that opt into the new regime will become much less supply-constrained. This will benefit not only the city’s current tenants and would-be homeowners, but also people and firms throughout the entire metropolitan region and beyond. In a free society, housing markets are all connected: people who move into a new dwelling usually move out of another, which then becomes available for other people to buy or rent, and so on. Fixing the big-city politics of housing will make housing more affordable for everyone, everywhere.

To be clear, our proposal is offered as a fix for bad housing policy in big cities, not as a comprehensive LIHTC reform. The LIHTC program is notoriously inefficient. It crowds out other investment in multifamily housing.3 Per federal dollar expended, it provides far less value to low-income tenants than does the housing-voucher program.4 There’s a strong argument for jettisoning the LIHTC program in favor of a refundable renter tax credit outside of supply-constrained markets. In constrained markets, however, new subsidies for renters would mainly result in higher rents. To help renters and would-be homeowners in such markets, Congress must address the barriers to supply. Incentivizing big cities to opt into federal pro-housing rules is one way to do it, and LIHTC developers are a constituency that can be rallied to the cause.5

The problem 

The rising cost of housing in the United States is not an economic mystery. People who cluster together into large metro areas become more productive, less vulnerable to exploitation by employers, and less exposed to risk from adverse economic and environmental shocks. They also enjoy the cultural amenities and diversity of services that urban agglomerations make possible. This is why cities exist.6

When demand for housing in an urban area grows, the market used to respond by delivering a lot more of it. Builders could tear down older houses and erect apartment buildings, or subdivide open land on the urban perimeter and mass produce suburban homes. The redevelopment of old houses into apartment buildings was common before zoning took hold in the first half of the 20th century.7 In the years after World War II, newcomers were accommodated in booming suburbs.8 Then something changed. A new concern for environmental quality in the 1970s led to restrictions on pell-mell suburban expansion, but no commensurate effort was made to liberalize land-use restrictions and lower the cost of building in existing urban and suburban areas.9

As a result, developers struggle to meet the demand for housing. This is evident from a number of papers estimating what economists call the “price elasticity” of housing supply within geographic units such as census tracts, counties, or metropolitan areas.10 Where housing supply is elastic, an increase in the demand for housing induces a lot more housing production, which in turn stabilizes prices. Where supply is inelastic, a similar demand shock results in little new production and registers instead as higher prices.

The recent papers on supply elasticities use different methods but reach similar conclusions: in most places, housing supply has become much less elastic over time. Part of the reason is that as a metropolitan area grows, there is progressively less undeveloped land within commuting distance of the central city. Since 1980, “[t]he net change in housing units [in census tracts that were developed by 1980] has been close to zero.”11 Increasing the housing stock by adding density in neighborhoods where people already live is not inherently infeasible, however. Rather, local governments have chosen to make it unprofitable — or to ban it outright. 

A new study analyzing the land-use codes of thousands of local governments documents two patterns of regulation.12 In high-demand cities, the dominant approach is “complexity”: the municipal code allows considerable density and bulk, at least in some districts, but to get a housing project approved, the developer must run a gauntlet of public meetings, provide deed-restricted units for low-income households, and satisfy various other demands. Conversely, in suburban jurisdictions, the dominant approach is flat-out exclusion: minimum lot sizes, density caps, and bulk and setback rules prevent single-family homes from being redeveloped except as bigger, more expensive single-family homes.

In combination, these modes of regulation create a metro-wide “zoning tax,” which the economists Joe Gyourko and Jacob Krimmel have translated into the amount by which the price of a typical, suburban single-family home lot is bid up due to restrictions on development throughout the region.13 Even prior to the COVID demand shock, the zoning tax added $100,000 – $200,000 to the price of an ordinary family home in the Los Angeles, New York, San Jose, and Seattle metro areas, and about $400,000 in the San Francisco area (in 2018 dollars).14

There are three basic strategies for increasing the supply of housing and thereby reducing the zoning tax: 

  1. Allow developers to convert more outlying farms and open space into suburban housing; 
  2. Allow developers to build denser housing, such as townhomes, duplexes, and small apartment buildings, in existing suburbs; and 
  3. Allow developers to build more and larger apartment and condo buildings in the central city. 

Each strategy comes with distinctive challenges. The first solution — outward suburban growth — is practical when a metro region is small. But as a metro grows large, the costs of commuting from ever more distant suburbs become onerous. This is why much of the housing development today in famously sprawl-tolerant Houston, Texas consists of infill housing, rather than greenfield subdivisions at the urban fringe.15

The second solution — intensification of land-use in existing suburbs — risks public backlash. People who live in suburban neighborhoods are much more opposed to dense development than people who live in big cities or near big buildings.16 And people everywhere are more supportive of building apartments along major streets and in districts with a mix of uses than in single-family-home areas.17 A further complication is that suburban homeowners are increasingly swing voters in national politics. After Governor Kathy Hochul, a Democrat, released a draft bill with ambitious housing-growth targets for local governments in the New York City region, suburban lawmakers rebelled and defeated it.18 Long Island Republicans overperformed in the next congressional election, and Hochul suspended the state’s next big urbanist initiative, a congestion-pricing plan for Manhattan.19

This is not to say that intensification of land use in existing suburbs is politically impossible. California has made impressive headway with state laws that give homeowners the right to add one or two accessory dwelling units, although bills to allow lot splits and duplexes in the suburbs were trimmed to the point of ineffectiveness.20 In New Zealand, legislation authorizing small multi-unit buildings in all residential neighborhoods generated a boom in housing production and brought down prices, though it also triggered a NIMBY backlash.21 Houston, Texas has tried to thread the needle with a citywide ordinance that allows townhomes and detached houses to be built on very small lots, while inviting blocks and neighborhoods to opt out if that’s what a majority of the homeowners want.22 Promising experiments with “gentle” intensification of density in single-family home neighborhoods are also underway in cities as varied as Nashville, TN,23 San Diego, CA,24 and Spokane, WA.25 But given the political delicacy of these reforms, and the role of suburban swing voters in national elections, it’s probably unrealistic to expect Congress to play anything more than a small supporting role, such as by authorizing HUD to provide grant funding or technical assistance. 

The third solution — remove the barriers to more, bigger apartment and condo buildings in central cities — presents a political challenge of a different order: overcoming the medley of urban interest groups that compete to “capture value” from proposed developments. 

In principle, it should be much easier to grow the housing stock in central cities than in the suburbs. In addition to urban residents’ affinity for density, there often exist industrial or commercial districts with large lots, which reduces the cost of land assembly26 and allows development to proceed with minimal impact on existing residential neighborhoods. The conventional wisdom used to hold that cities are very open to development27 or even dominated by “growth machines.”28 In 1991, when a commission of luminaries convened by then-Secretary of Housing and Urban Development Jack Kemp released a sweeping study of federal, state and local barriers to housing development, land-use regulations in the urban core were barely mentioned.29 

Around the same time, a “return to the city” began.30 Young people flocked to formerly downtrodden neighborhoods as crime rates fell and new cultural amenities beckoned. Suddenly there was money to be made in urban housing development — and, on some sites, a “surplus” of potential value. (The surplus is the difference between a project’s financial value and the cost of building it, including the opportunity cost of forgoing the existing use of the site.) Developers who wanted to build for the newcomers were met not with open arms, but by a proliferation of urban interest groups that sought to capture value from proposed projects. From the scaffolding of 1970s-era good government reforms — community review boards, liberal access to courts, environmental study requirements, hard-look judicial review, and single-member district elections — a new convention emerged: the community-benefit agreement or CBA.31 Big housing projects would go forward only if the developer inked a deal with third-party groups, committing, for example, to hire union workers, provide deed-restricted affordable housing units, set aside land for parks, pay for transit improvements, give cash to community organizations, and on and on. 

In theory, if city councils know the true size of project surpluses and could quickly impose a feasible benefits deal, the CBA convention need not significantly constrain development.32 But city councils have imperfect information about the size of a project’s surplus, the CBA negotiations take time, and, ex ante, developers thinking about what to bid for a site don’t know what package of benefits will be required to get a deal done.33 More fundamentally, a city council can’t force the various groups to acquiesce in its preferred deal. A dissenter who wants more of the pie can always sue or threaten to sue, raising claims under environmental review and other laws that regulate the processes by which cities review housing proposals.34 Even if the legal claim is flimsy, filing a lawsuit can delay a project for years, causing the developer to incur large holding costs and mechanically reducing the project’s internal rate of return.35 The upshot is a collective action problem among would-be value capturers.36 Each local group has an incentive to hold out for a bigger piece of the pie. This dynamic can squander the surplus — and it disincentivizes developers from proposing projects in the first instance.

In California, the poster state for constrained supply and high demand, the legislature increasingly recognizes that the local politics of discretionary project approvals and environmental litigation are thwarting urban infill development. Starting with a proposal from Governor Jerry Brown in 2016, the state has considered numerous measures that would require infill housing developments to be reviewed ministerially — that is, only for compliance with objective standards — and that would exempt such projects from the California Environmental Quality Act. The labor unions and other lawyer-laden groups that exploit the status quo insist, however, that any streamlining bill cover only those projects that pay union-negotiated wages, provide deed-restricted affordable housing, and would not be built on a site that may have been occupied by a tenant or a rent-controlled37 housing unit. Ezra Klein dubs it the “everything bagel”38 approach. The state’s proliferation of cost-elevating requirements makes new housing unprofitable to build even in very expensive places,39 and as a result we get very little of it. So far, the scores of housing bills that California has passed have generated no discernable uptick in housing production relative to other states, with one salient exception: accessory dwelling units (ADUs).40 

In the case of ADUs, California law preempts local zoning (and HOA restrictions) insofar as they prevent a homeowner from building a freestanding, 800-square-foot ADU in their backyard and a “junior ADU” carved out of the existing home. State law also prevents local governments from charging expensive fees for ADUs, demanding off-street parking for ADUs, or imposing discretionary conditions of approval on an ADU project. And there is no requirement that ADU developers use union labor, pay into affordable housing funds, or impose deed restrictions that prevent ADUs from being rented or sold at their fair-market value. As a result, ADU production has boomed, going from a few hundred units per year to tens of thousands.41

For big, high-demand cities to realize their potential as a source of new housing, states need to do for apartment and condo development what California did for ADUs: require code-compliant projects to be approved ministerially, prohibit high fees and other exactions, and preempt big cities’ zoning and development standards insofar as they preclude new projects of reasonable size. The enactment of these reforms need not prevent cities from capturing value created by the removal of land-use restrictions. But cities would have to capture value in a more transparent manner, such as by taxing land or auctioning the right to build.42 Unlike impact fees, inclusionary zoning, and ad-hoc community-benefit demands, these mechanisms for value capture could not be used to set an infeasibly high price on the right to build.43

The LIHTC solution

3.1. Overview

The Low Income Housing Tax Credit is the major federal source of funding for affordable housing development, providing subsidies of about $14 billion dollars per year.44 Each state receives a LIHTC allocation proportionate to its population; states then distribute their tax credits to projects in accordance with their “Qualified Allocation Plan” (QAP). Federal law establishes certain factors that states are required to consider through their QAP, but states have broad discretion to add others and decide how much weight to give each factor. 

We propose that Congress establish new, categorical restrictions on LIHTC funding for projects in big, expensive cities. In cities above a threshold for size (population) and housing costs (median fair-market rent), projects would be LIHTC-eligible only if the city opts into a pro-housing regulatory framework.45 The pro-housing rules would apply to all housing projects, not just projects built with federal subsidies. Specifically, the city or its parent state would have to agree:

  1. To review housing development projects ministerially and approve them if a reasonable person could conclude that the project complies with applicable zoning and development standards.
  1. To cap impact fees at $10,000 per unit (roughly the national average for impact fees on multifamily housing).46
  1. To waive affordability mandates and other exactions insofar as a reasonable person could conclude that they render the project economically infeasible.
  1. To waive height, bulk and density restrictions insofar as they prevent development of apartment and condo buildings of reasonable size in commercial districts. (The federal government would specify building heights and volumes that big cities must allow.)

As explained more fully below, these criteria are drawn from recent, bipartisan reforms adopted in both Republican-led and Democrat-led states. If applied concurrently to the same sites, they would dramatically increase the economic feasibility of apartment, condo, and townhome development. 

Congress would have constitutional authority under the Commerce Clause to impose the above requirements, but politically that might be a bridge too far. Instead, we suggest that Congress give big cities a choice: either opt into the pro-housing regulatory framework, or forego federal funding for affordable housing.47

Presenting cities with this choice would change the local politics of housing. Progressive council members would suddenly be torn between the interest groups that exploit the status quo (lawyered-up labor and community groups), and affordable housing developers and lower-income constituents, who want continued access to the tax credits. And whereas housing politics today play out in backroom negotiations between developers, “groups,” and council members, city councils would have to take a public, high-stakes vote on whether to accede to the federal pro-housing framework or forgo many millions of dollars in affordable housing funding. Council members who vote to turn down the money would come under attack for throwing poor people and affordable housing developers under the bus. Such attacks probably wouldn’t resonate in an exclusive suburb, but they’d be catnip for challenger candidates in heavily Democratic cities.

We are not the first to argue that Congress should condition affordable housing funds on local governments’ removal of land-use restrictions. A generation ago, Jack Kemp’s blue-ribbon commission proposed that states’ LIHTC allocations and private-issuance bond authority “be contingent upon the State having [a HUD-]approved barrier-removal plan.”48 The commission also “strongly recommend[ed that] Congress amend the National Affordable Housing Act of 1990 to authorize HUD to condition [other] assistance to State and local governments based upon their barrier-removal strategies.”49 The commission’s call fell on deaf ears. Thirty-three years later, the National Affordable Housing Act still prohibits HUD from “establish[ing] any criteria for allocating or denying funds…based on the adoption, continuation, or discontinuation by a jurisdiction of any public policy, regulation, or law.”50

The Kemp commission’s ambitions were Sisyphean, however, and its unanswered call hardly dooms our proposal today. The commission put restrictive suburban zoning squarely in its crosshairs, whereas our proposal targets only large, expensive cities, whose residents generally support dense housing. And the commission would have stripped funding from states that fail to adopt a HUD-approved barrier-removal plan, whereas our proposal leaves states’ LIHTC allocations intact.

The Kemp commission was also writing at a time when the states had little experience trying to control local barriers to housing development. Implementing the commission’s recommendations would have required Congress and HUD to fashion an entirely new and far-reaching set of federal rules and administrative practices about the legal adequacy of state barrier-removal plans. Our proposal borrows instead from existing state law, piggybacking on the leading bipartisan reforms of the YIMBY era, as explained below. Federal courts and agency administrators fleshing out our proposed requirements would have a body of state law and state-court decisions to draw upon. 

3.2. Opt-in field preemption or opt-in waivers?

The central challenge for our proposal is ensuring that the pro-housing conditions on big cities’ LIHTC eligibility meaningfully constrain the cities that opt in. To generate the hoped-for burst of housing development, the conditions must concurrently (1) ensure an ample supply of sites on which the applicable zoning and development regulations allow reasonably large apartment and condo buildings; (2) prevent cities from imposing big pecuniary or in-kind taxes on development of the sites; and (3) prevent cities and third parties from imposing large costs through permitting delays. 

All three legs are necessary for the pro-housing stool to support development. Knock one out and the old ways will come right back. For example, if Congress conditioned LIHTC funding on generous zoning and speedy permitting, but left fees untouched, cities could establish “impact fees” of $100,000 per unit or more and invite developers to apply for discretionary fee waivers in exchange for CBAs. Or if Congress required speedy, predictable permitting and capped fees, but left zoning untouched, cities could shrink-wrap their zoning to the existing built environment, forcing landowners who want to redevelop their site to apply for a discretionary “spot” upzoning and pay off the interest groups that oppose it.51

The history of California’s ADU law is instructive.52 Way back in 1982, the legislature decreed that local governments may disallow ADUs within residential zones only if the locality makes a finding of “specific adverse impacts on the public health, safety, and welfare.” Many local governments responded by requiring ADU applicants to obtain onerous, discretionary permits. Concerned that local governments were abusing their discretion, the state legislature, in 2002, directed local governments to permit ADUs ministerially; enacted a template to which local ADU ordinances must conform; and required local governments to submit their ADU ordinances to the state housing agency for review. The 2002 bill did not, however, displace “height, setback, lot coverage, architectural review, site plan review, fees, charges, and other zoning requirements generally applicable to residential construction in the zone in which the property is located.”53

Studying the response to this statute, law professors Margaret Brinig and Nicole Garnett collected the zoning ordinances of every California municipality with more than 50,000 people, as well as public-meeting minutes and news stories.54 They found that most California cities effectively thwarted the new mandate with a “thousand paper cuts.” Cities stymied ADU construction with design review, costly building-material mandates, rental restrictions, owner-occupancy requirements, minimum lot sizes, conditional use permits, permit-filing fees, impact fees, and tight allowances for the permissible size of an ADU. Some of these requirements probably violated state law, but anti-ADU local governments had few compunctions about pushing the envelope of their reserved authority. 

Frustrated by local intransigence, California enacted additional ADU bills in 2016, 2017, and 2019. The 2016 statute further constrained local requirements for parking, unit size, fire sprinklers, utility-connection fees, and lot-line setbacks. Additional tweaks were made in 2017, and in 2019 the legislature created an essentially unqualified right for every homeowner in the state to add a freestanding backyard ADU of up to 800 square feet, plus a “junior ADU” of up to 500 square feet within the envelope of an existing structure. The 2019 legislation also preempted local impact fees on ADUs of up to 750 square feet. At this point, ADU production took off, and ADUs now comprise a substantial portion of all new housing in California’s major metro areas.55

Congress cannot be expected to be similarly vigilant about big cities’ land-use practices. There is too much else on the federal legislative agenda and too much partisan polarization, to say nothing of procedural obstacles like the Senate filibuster.

Under these circumstances, we see two workable paths forward. The first, which we’ll call opt-in field preemption, is for Congress to authorize HUD or another federal agency to promulgate model codes that would entirely supplant local zoning and development standards for the cities that opt-in.56 Adam Ozimek and John Lettieri urge this approach, proposing that “the federal government… establish a standardized zoning and building code drawn from best practices nationwide and designed to allow builders to meet local housing demand without having to navigate onerous bureaucratic hurdles.”57 Cities would not be able to vitiate the federal codes with novel local restrictions because their decision to opt in would preclude them from applying any zoning and development standards of their own.58 

The other approach, which we’ll call opt-in waivers, would require participating cities to waive local requirements on an as-applied basis and accept congressionally prescribed remedies for violating the federal pro-housing rules. Congress would issue the initial set of rules. Participating cities would remain free to apply municipal zoning and development standards except insofar as they conflict with specific federal pro-housing rules as applied to a given project. A federal agency would be authorized to update the initial pro-housing rules, and organizations which receive fair-housing enforcement funding could be invited to enforce them.59

The opt-in field preemption model has some attractive properties. As Ozimek and Lettieri emphasize, it would help to standardize zoning, which in turn would create stronger competition and economies of scale in the development industry. Over time, this is likely to increase productivity in the construction sector and bring down the cost of housing.60

But opt-in field preemption has significant downsides too. To induce cities to give up zoning entirely in targeted areas, the federal government would probably have to put a lot of money on the table. Ozimek and Lettieri suggest that the federal government offer a generous per-unit bounty for each new home in districts where the federal rules apply. That could be transformative — and a huge improvement over Congress’s practice of doling out funds using population-based formulas, earmarks, and discretionary grants distributed by bureaucrats who have little capacity to monitor the implementation or the efficacy of the grantees’ programs. But the federal government’s current fiscal outlook is not bright, and it’s doubtful that Congress will soon have the appetite for a large new spending program. Opt-in field preemption would also take years to implement: a federal agency would have to hire staff and develop expertise in zoning and building codes; then undertake notice-and-comment rulemaking; and then fight through the inevitable legal challenges to the new federal codes.61 Finally, members of Congress who are wary about too much national control over land use may be reluctant to provide big financial inducements for local governments to opt into fully fledged federal zoning and building codes.

The opt-in waiver approach would be easier to implement, less costly, and, we suspect, less controversial. Congress could write the initial rules into the statute, borrowing from state law, as explained below. The statute could be implemented (at least initially) just by posting a list of cities whose housing prices and population meet the thresholds for “high” and “big,” denoting which cities have opted in. As for cost, we think, though we cannot be sure, that the threat of losing LIHTC funds would prompt many big cities to opt in, and so the federal pro-housing policies would not require fiscal sweeteners tied to future housing production in the participating cities.62 The politics are likely to be easier too, because our plan would only affect a small number of big cities (where the public is tolerant of dense development); because it would piggyback on bipartisan state-level reforms; and because participating cities could still apply most of their zoning and development standards in most instances. 

The big downside of our approach is that participating cities may find loopholes and abuse their residual land-use authority, much as California cities thwarted state ADU law for years. We think this risk can be mitigated, however, by adapting another state innovation: tracking city-level housing production over time, benchmarking cities against peer jurisdictions, and eventually decertifying the poor performers regardless of whether they had opted in. Oregon is building out this approach.63 A federal agency like HUD could draw on Oregon’s work, as well as recent academic research on the microgeography of housing supply.64 Peer benchmarking and subsequent decisions about whether to decertify a city or negotiate a package of reforms would require considerable administrative capacity. But, critically, that capacity doesn’t have to exist when Congress enacts LIHTC conditionality. It can be developed over time, and meanwhile, the threat of future decertification may discourage cities that opted in from abusing their residual authority over land use. 

We emphasize that our opt-in waiver approach in no way precludes Ozimek and Lettieri’s field preemption model should fiscal and political windows for the latter open up. Field preemption may also prove to be a reasonable alternative to decertification and loss of LIHTC for the handful of cities that remain very bad performers even after opting into the pro-housing rules (see section 3.3.4, below).

3.3. Borrowing from the states

The last section asserted that Congress could construct the federal pro-housing rules by borrowing from bipartisan reforms in the states. This section makes that claim concrete, with examples. We address each leg of the pro-housing stool: limits on permitting discretion, limits on fees and exactions, and allowances for the physical size and density of new housing developments. Then we discuss a new Oregon model for remedial interventions in cases where a pro-housing plan doesn’t yield the intended results.

3.3.1. Controlling permitting

A number of states have recently passed laws that check or eliminate local political discretion over the approval of housing projects that comply with applicable objective standards. There is some variation from one state to the next, but, in general, these laws (1) prevent cities from denying, reducing the density, or conditioning the approval of a project on the basis of subjective standards, and (2) provide judicial or third-party-permitting remedies in cases where a city wrongfully denies a project application or delays it excessively. Some states also require that zoning-compliance determinations be made by an administrative actor rather than an elected body.

Though a national audience may be surprised to hear this, California has spent the better part of the last decade bolstering controls on local permitting discretion. California’s strongest streamlining laws would be a good model for Congress and other states — if the cost-elevating “bagel toppings” were stripped away. Here are the salient features of the new California permitting framework:

  • Anti-retroactivity. Developers can lock in the rules that apply to their project with a simple filing called a “preliminary application.”65 
  • Prompt, written notice of applicable standards. Within 30-60 days of receiving a complete project application, cities must provide written notice of any objective zoning and development standards with which the project does not comply. Projects are deemed to comply as a matter of law with standards the city doesn’t flag as violated.66
  • Ambiguities resolved in favor of housing. If a reasonable person could deem a housing development project compliant with applicable standards, then the project does comply as a matter of law.67
  • No subjective conditions of approval. Under the strongest of the California streamlining laws, cities may not impose any conditions of approval on a project unless the conditions are necessary to make the project compliant.68 Under the state’s weaker streamlining law, cities may impose discretionary conditions of approval so long as the conditions don’t reduce the project’s density.69 The weaker version is substantially less effective because the existence of municipal authority to impose subjective conditions triggers review under the California Environmental Quality Act, leading to delays, costs, and third-party litigation risk.70
  • Fallback judicial permitting. If a city takes too long reviewing a project application, the developer can go to court and the court will order the project approved if a reasonable person could deem it compliant with the applicable standards.71 Courts can also order approval of projects that were denied on frivolous grounds.72
  • Well-calibrated litigation incentives. If a court finds that a city has violated state streamlining law, the city has to pay the plaintiff’s attorney’s fees and the court retains jurisdiction over the project.73 If the city appeals the court’s judgment, the city has to post a bond and compensate the developer if it loses the appeal.74 Conversely, if a third party successfully challenges the city’s approval of a dense housing project on an infill site, the third party gets attorneys fees only if the court finds that the city approved the project in bad faith.75

Other states, including Texas, Tennessee, Florida, and New Jersey, have addressed the permitting problem by authorizing developers to obtain permits from a licensed third party, such as a structural engineer.76 Manufactured housing built to the national HUD code is also permitted by HUD-approved third-party inspectors.77 Third-party permitting is also recognized in Japan.78

We are not aware of any studies that assess the relative efficacy of California-style and Texas-style controls on local permitting discretion. Unless or until one approach is shown to be demonstrably superior, it would be reasonable for Congress to invite big cities to choose between the “California solution” and the “Texas solution” for permitting delays.

3.3.2. Controlling local fees, extractions, and price controls

States have adopted widely varying restrictions on local governments’ use of impact fees, in-kind exactions, and price controls to regulate development. Many states require impact fees to be justified by a “nexus” study, which nominally calibrates the amount of a fee to harms caused by a project or infrastructure needs occasioned by it. But nexus studies can be reverse-engineered to justify almost any fee,79 as indicated by the fact that impact fees in California (where the studies are required) can reach $100,000 per dwelling or more.80

Some states ban cities from imposing rent control or “inclusionary” zoning, i.e., requiring developers to set aside a minimum share of the units in a project as price-restricted housing for low- or moderate-income households.81 Other states actively encourage inclusionary zoning, with laws that require cities to plan for price-restricted housing development at varying levels of affordability.82

Any sensible policy response to the problem of excessive fees, exactions, and price controls needs to wrestle with three issues. First, the observed, official schedule of fees and exactions may comprise only a small portion of the actual set of fees and exactions negotiated in the shadow of third-party litigation threats or a city council’s deliberate slow-walking of a project. Put differently, limits on exactions that are not accompanied by strong controls on permitting may prove to be more illusory than real.83

Second, fees, exactions, and even inclusionary-housing mandates aren’t unambiguously bad for housing development. True, they raise the cost of building and thereby reduce housing production relative to a counterfactual world in which everything is the same but for the high fees. That counterfactual is implausible, however, to the extent that fees, exactions, and the like buy local political goodwill.84 Increasing local political acceptance of development surely also tends to result in more development, other things equal. Getting the fee level “right” is a tricky balancing act, and the best answer may depend on whether other, less economically distortive devices for capturing the site value created by land-use liberalization are legally and administratively feasible.85 (If the federal fiscal window were to open wider in the future, Congress could solve the impact-fee problem a la Ozimek and Lettieri, by offering a payment to cities for each new dwelling unit in exchange for the city agreeing to waive fees and exactions. This would be a better use of an expanded federal checkbook than, say, down payment assistance or non-targeted tax credits.)

Third, fees, exactions, and price controls can function as an equally exclusionary substitute for classical forms of exclusionary zoning such as minimum lot sizes and density caps. If a higher level of government requires a city to allow smaller lots and more units on each lot, the city may respond by imposing prohibitively costly exactions or by requiring new units to be sold or rented at prices so low as to render development infeasible. This is not a fanciful concern. The mayor of Huntington Beach, California, a virulent opponent of state housing mandates, recently said that though he personally opposes rent control, he would vigorously exercise the authority conferred by a proposed ballot measure giving cities absolute control over the price of rental housing. Why? To “defend [his] communit[y]” against the state’s pro-housing mandates (by setting rents so low that developers stop building apartments).86

Bearing these tensions in mind, we recommend the Congress establish the following rules for cities that wish to retain LIHTC eligibility:

  • Fees totalling less than $10,000 per unit (adjusted for inflation) are permissible and require no further justification.87 Ten thousand dollars is roughly the average total impact fee on a new unit of multifamily housing, nationally.88
  • Fees above $10,000, in-kind exactions, and price controls and income-based occupancy limitations must be waived if the developer requests a waiver and a reasonable person could conclude, from the evidence in the record, that the project would be infeasible without the waiver.

The second bullet borrows from California’s “builder’s remedy” law, which similarly checks inclusionary zoning by cities without a state-approved housing plan.89 Rather than preempt every local inclusionary ordinance, the law allows cities to require up to 20% of the units in a project to be affordable so long as this doesn’t render the project infeasible. Crucially, cities get no deference on feasibility. They must demonstrate it by a preponderance of the evidence, and if a reasonable person could conclude that the local inclusionary mandate renders a project infeasible, the mandate must be waived for that project.90

Purely on the policy merits, it would be reasonable — and simpler — for Congress just to ban price controls on any housing that does not receive public subsidy, rather than requiring case-by-case waivers. A recent literature review concludes that rent control usually increases rents on uncontrolled units, reduces housing construction and rental housing supply, reduces household mobility, increases misallocation of units (e.g., an old, single person living in a large apartment that could be occupied by a family), and reduces the quality of housing.91 But rent control and inclusionary zoning mandates are quite popular, especially among Democrats.92 We think our proposed LIHTC reforms are more likely to obtain bipartisan support in Congress and to induce participation by big cities if they don’t completely prohibit cities from regulating prices. California’s new feasibility standard represents a middle path worth trying.

3.3.3. Controlling zoning

The third leg of the stool — ensuring an ample supply of sites zoned for reasonably dense housing — could be handled in a variety of ways without entirely supplanting local zoning, as shown by the following examples from Florida and California.

Density bonuses relative to a parcel’s zoning. California’s Density Bonus Law offers density increments (additional dwelling units) of 20% to 100% relative to local zoning for qualifying projects of five or more units.93 Developers seeking a density bonus must submit a hypothetical “base project” design, showing the number of units that could be built on their site in a zoning-compliant project. The bonus is applied to the base project’s unit count. The actual project doesn’t have to resemble the base project, but the number of units can’t exceed the base count plus bonus. 

Bonuses relative to other zoning. Florida’s Live Local Act goes a step further, authorizing housing development in commercial and industrial zones (where local zoning doesn’t allow it), and borrowing and “bonusing” certain development standards from other districts in the same jurisdiction. A city “may not restrict the density of a proposed [Live Local] development… below the highest currently allowed density on any land in the municipality.”94 Live Local projects are also entitled to “150 percent of the highest currently allowed floor area ratio on any land in the municipality,”95 and to heights of up to three stories or “the highest currently allowed height for a commercial or residential building located in [the] jurisdiction within 1 mile of the proposed development.”96 However, “if the proposed development is adjacent to… a parcel zoned for single-family residential use… within a single-family residential development with at least 25 contiguous single-family homes, the municipality may restrict the height of the proposed development to 150 percent of the tallest building on any property adjacent to the proposed development, the highest currently allowed height for the property provided in the municipality’s land development regulations, or 3 stories, whichever is higher.”97

Substituting another zoning district. Another strategy is to spell out certain project typologies or attributes that must be allowed in good-for-dense-development locations (e.g., on major streets, near places of commerce), and then to specify a “substitution rule” that transports to the site all of the zoning and development standards of another district where projects with those attributes are allowed. Thus, under California’s commercial-corridors upzoning law, local governments must allow (along big streets and near transit) certain housing projects with densities of 35-70 dwelling units per acre and heights of 35-65 feet.98 The zoning, subdivision, and design-review standards that apply to the project “shall be those for the zone that allows residential use at a greater density between (1) the existing zoning designation for the parcel [and (2) t]he zoning designation for the closest parcel that allows residential use at [the] density [to which the developer is entitled under state law].”99 Similarly, under California’s builder’s remedy, the zoning standards for the project are those that “would have applied [had the project] been proposed on a site [elsewhere in the city] that allow[s] the density and unit type proposed by the applicant.”100

Waivers. Complementing other strategies, a state may require waivers of otherwise-applicable zoning and development standards that would “physically preclude” a project at the density authorized by state law. California’s Density Bonus Law adopts this approach.101 One might worry that cities would try to force costly, value-destroying redesigns of projects without “physically precluding” them, but California courts have foreclosed conditions of approval that reduce project amenities.102 103 These judicial precedents, in combination with the potent remedies available under California’s streamlining laws, appear to have deterred most cities from trying to sink density-bonus projects. 

For purposes of conditioning big cities’ access to LIHTC funding, we would combine the best features of Florida’s and California’s upzoning laws, roughly as follows:

  • From California, borrow height and density allowances tied to street width and proximity to transit.
  • From Florida, borrow FAR allowances. 
  • From California, borrow waivers of local zoning and development standards that would physically preclude a project of the allowable density, inclusive of amenities.
  • From Florida, borrow step-down rules for projects that adjoin a cluster of single-family homes.
  • From both states, borrow limits on parking requirements for projects near transit.

Or if the above feels too daunting, Congress could simply:

  • Authorize multifamily housing development on parcels zoned for or proximate to commercial or industrial uses, in the spirit of Florida’s Live Local Act and California’s AB 2011.

3.3.4. Outcomes and fallbacks

To discourage gaming and reward good performance, we strongly recommend that Congress provide for outcome-based evaluations of cities and future decertification of the worst performers.104 

Oregon is developing a model that the federal government can build upon. The legislature has instructed Oregon’s housing department to produce a “dashboard” that tracks housing production at the city level and benchmarks each city against “other local governments with similar market types.”105 For each city, the department has proposed a set of comparator jurisdictions based on several attributes, including current population size, household incomes, and share of housing used as second or vacation homes.106

Oregon cities whose housing production is below the median of their market peers, over a four-year period, may have their land-use practices audited by the state’s housing department.107 If a low-performing city and the department fail to reach agreement on remedial measures by a statutory deadline, the city may be required to apply “model ordinances and procedures developed by the department to all residential development within the city.”108

Congress should authorize HUD (in consultation with the Council of Economic Advisors) to develop a housing-production dashboard that would track cities’ housing production and rate them against comparators. Once every several years, HUD should designate the worst performers (say, jurisdictions in the bottom tercile) over the previous period. Big cities that land in the worst-performer bin would lose their LIHTC eligibility regardless of whether they had opted into the federal pro-housing program — unless they agree to additional, demanding constraint-removal measures. For these cities, HUD might reasonably require opt-in field preemption (if authorized by Congress) as a condition of continued LIHTC eligibility, since by hypothesis the more limited federal pro-housing rules had not paid off.

HUD should be given broad discretion to develop the comparator methodology. Oregon’s approach is still a work in progress, and there may be better ways of adjusting for nongovernmental factors that affect housing supply and demand. For example, economists have shown that housing supply elasticities vary at the census-tract level as a function of characteristics such as density, topography, and share of undeveloped land.109 Rather than benchmarking cities against others of similar size, income, and desirability as a vacation destination (the Oregon model), it would probably make more sense to benchmark each census tract within a city against other tracts across the nation that have similar characteristics.110 Cities overall-performance score would be a weighted average of their tract-level performance.111

A significant advantage of this approach is that the comparator set could include tracts from many different kinds of cities, not just other big, expensive cities that are subject to LIHTC conditionality. The more comparators, the less likely it is that a city’s performance rating will be sensitive to the choice of whether to include or exclude a given comparator. Also, once the tract-based methodology is established, it would be straightforward for HUD to produce performance evaluations for all cities for which production data are available, not just the big, expensive cities subject to LIHTC conditionality. This information would be very useful to state policymakers and perhaps even to the cities themselves. 

If Congress adopts our recommendation for performance-based evaluations, it should also invest in better data about housing production. Today, there is no official, annual census of net housing production. The U.S. Census does conduct an annual building-permit survey, but it relies on voluntary self-reports from local governments with no incentive for participation or accuracy.112 The Building Permit Survey also excludes conversions and remodeling of existing structures — so it misses ADUs created out of garages, basements, or attics; large houses converted into apartments; and office-to-residential conversions. Moreover, the fact that a city issues a building permit does not mean that housing will be built on the site. In highly regulated markets, property owners often acquire building permits to increase the option value of their property for redevelopment, which is quite different from deciding to exercise the redevelopment option. Another U.S. Census product, the Survey of Construction, estimates housing starts and completions, but only at the national and Census Region geographies. This means it can’t be used to assess city-level performance. These federal data limitations are serious enough that California has built its own state-level permit tracking and construction completion system in order to enforce state housing law. 

Because of the weaknesses of the U.S. Census products, economists prefer to measure housing production using tax-assessor data aggregated by private firms, or counts of dwelling units from the decennial Census. Researchers studying a tractably small number of cities have also resorted to direct outreach to obtain more comprehensive data directly from municipal departments. 

It would be a little odd to make a $15 billion federal program depend entirely on data which the federal government does not own. And though once-a-decade feedback on the size of local housing stocks per the decennial Census would be better than nothing, more frequent performance evaluations would be much more valuable. A collaboration between the teams managing the Census Building Permit Survey, Survey of Construction, Master Address File, and the U.S. Postal Service Delivery Sequence File to develop a “single source of truth” on the number and location of active residences in the United States would be terrific but is beyond the scope of this paper to develop. The data problem is nonetheless of sufficient urgency that fixing it should be part of any significant new federal action on housing.113

3.4. Regulatory refinement

Congress should authorize HUD or Treasury to waive, refine, or supplement the pro-housing conditions by rulemaking. Some cities will probably find loopholes. A federal agency should have authority to close them. Some pro-housing conditions may prove to be pointlessly cumbersome. A federal agency should have authority to waive them. As states experiment with new devices for facilitating housing production, federal administrators will learn about what interventions work to induce housing production, and about what accommodations or compromises minimize political backlash. The federal pro-housing standards should evolve accordingly. 

To the extent that Congress worries about a rogue administration radicalizing the LIHTC eligibility conditions, Congress might require that regulations be patterned on pro-housing reforms that have been adopted on a bipartisan basis in the states, or that are otherwise justified as having substantial support among elected officials of both major political parties. 

3.5. Objections

This section anticipates and answers a number of objections to our proposal.

3.5.1. Target the suburbs, not big cities

Many commentators have argued that federal efforts to remove local barriers to housing supply should target “higher income, exclusionary communities,”114 not big cities. We do not think Congress should condition LIHTC funding on pro-housing policies outside of large, expensive cities. In suburban towns, the public is much less supportive of dense housing and so city councils would probably just turn down the money. Should they accept the money instead, the resulting development could trigger political attacks on candidates who supported the LIHTC bill and jeopardize the fragile bipartisan coalitions that have had some success pushing zoning reform in the states. A focus on the suburbs doomed not only Jack Kemp’s efforts a generation ago, but HUD Secretary George Romney’s fair housing efforts a generation before that.115 We propose here not to repeat 53 years of federal housing reform failures in the suburbs, but to learn from those failures. 

3.5.2. Don’t punish low-income residents for their city’s bad decisions 

A critic might say that a city’s low-income residents aren’t responsible for the decisions of their city council, so it’s unfair to deprive them of low-income housing that could be developed with LIHTC funds just because their city council chose not to accept the federal pro-housing rules. 

This objection overlooks two basic points. First, our proposal would not take away subsidized housing from anyone who already has it. The “punishment” imposed on a city’s low-income residents would be, at worst, a slightly lower probability of winning a unit in the city’s below-market-rate-housing lottery.116 Second, low-income people are already being punished — through high rents — on account of the anti-housing policies and practices of their city councils. Our policy will ameliorate that suffering. And it will ameliorate it not only in the big cities that opt in, but also in cities that remain on the sidelines. Housing markets are connected.117 A growing housing stock in the big, high-price cities that opt in will induce some in-migration, reducing demand-side pressure on housing markets in cities elsewhere. The resulting decline in rents will also enable the federal Housing Choice Voucher program to serve many more low-income applicants.118 Reforms enacted in response to our proposal will also increase the stock of legally permissible LIHTC construction sites, stretching LIHTC dollars into more subsidized affordable units in the long run.

3.5.3. Condition LIHTC eligibility on performance only

Given that no state has concocted a demonstrably effective formula for increasing the supply of housing in big cities, one might think that Congress should just condition LIHTC eligibility on cities’ housing-production performance, rather than on cities opting into a set of federal pro-housing rules. Because housing production varies with the business cycle, a performance standard would probably have to be relative, as in Oregon. Cities would be ranked against peer jurisdictions and the worst performers would lose LIHTC eligibility until the next evaluation date. 

We have three concerns about a purely performance-based approach. First, because it would invariably doom some cities (the worst performers) to loss of LIHTC eligibility, Congress would no doubt set a very low bar. Perhaps cities in the bottom decile or quintile would lose eligibility. This might put some pressure on cities near the cutoff to remove barriers to housing development, but middling and better performers could rest on their laurels. By contrast, our approach would require every city that wants continued LIHTC eligibility to improve their housing policies by opting into the federal pro-housing rules.

Our second and more fundamental concern is that the purely performance-based approach would do little to solve the underlying political-economy problem that makes housing development in big cities so expensive (see section 2). Instead of council members casting a single, high-profile vote through which they must either choose to forgo federal funding for affordable housing or accept a suite of federal pro-housing rules — rules that will sap the resources and political strength of NIMBY and “value capture” groups over time — council members would have to cast vote after vote after vote on discrete projects, rezonings, streamlining proposals, impact-fee measures, inclusionary-zoning reforms, and the like. Each of these votes may have a small, uncertain effect on the probability that the city loses (or regains) LIHTC eligibility in the future. Each of these votes will have a clear effect on the council-member’s future support from local interests demanding that the member vote “No.” If the city does eventually lose LIHTC eligibility, the public won’t be able to trace that loss to specific votes of specific council members. Future challenger candidates could blame incumbents for presiding over the city’s loss of affordable housing funding, but the challengers won’t be able to credibly promise to restore it (who knows what policy changes would sufficiently increase housing production?), and the incumbents could deflect blame by attributing the LIHTC loss to “market forces” or “the Fed’s interest rate policies” or “federal bureaucrats.” 

Under these circumstances, we think it’s pretty unlikely that the remote and uncertain effect of a council member’s anti-housing vote on the city’s future LIHTC eligibility (in a purely performance-based regime) would be enough to change their mind on any matter where an organized local faction wants the member to vote No.

Our third concern is measurement error. As noted above, there is no official, annual census of net housing production at the jurisdiction level. Outcomes could be measured with data from private vendors, but such dependence may lead to implementation difficulties or legitimacy concerns. 

While we are wary of basing LIHTC eligibility on performance alone, we would be fine with allowing the very best performers to opt out of the federal pro-housing rules without loss of LIHTC eligibility, even if performance is measured with significant error.119

3.5.4. Let cities choose from a pro-housing policies menu

Under our proposal, cities that opt in would have to accept a premade package of federal pro-housing rules about permitting, exactions, and zoned capacity. By contrast, several states have recently enacted “pro-housing designation” programs that reward cities for choosing several policies from a menu containing anywhere from 15 (Massachusetts) to 55 (California) items. Might this be a better approach?

We think not — at least not as the states are now doing it. As we have emphasized throughout, zoning, procedure, and costly regulations are substitutes for city councils that want to stop development or force it through a discretionary approval channel. To liberalize supply, governments must concurrently reform zoning, procedure, and cost-elevating requirements and apply the reforms to the same set of sites. The states’ menu-based approaches deny this reality. They let cities choose a handful of policies and apply them to different sites, with no regard for complementarities120. The state-promulgated menus may be useful as a way of drawing city councils’ attention to policy tools that council members hadn’t known about, but that’s about it.

That said, if Congress wants to create more flexibility for big cities, it might consider authorizing HUD or Treasury to create a small “menu of packages” for cities to choose among. Each package would combine permitting reform, fee/exaction reform, and zoning allowances, and concurrently apply the reforms to the same sites. For example, cities might be given the choice between Texas-style third-party permitting and California-style ministerial review with a judicial fallback, or between allowing more density near fixed transit stops and allowing more density on commercial corridors.

3.5.5. It’s unconstitutional

Opponents might lob a grab bag of claims against the constitutionality of our proposal. These claims are weak, though the constitutional foundations of administrative law are sufficiently in flux that we can’t be completely certain how a court would rule on all of them. 

First, opponents might contend that restrictions on LIHTC that apply only to large, expensive cities would violate Art. I § 8 cl. 1 (the Uniformity Clause), which requires federal taxes and tax credits to be uniform throughout the country. This claim is very unlikely to succeed because the Uniformity Clause merely requires that a tax with specific geographic application or effect operate uniformly in every place where it is in effect and that Congress have a legitimate, non-geographic rationale for its policy decision.121 Our proposal would operate uniformly vis-a-vis the cities that meet the size and price criteria. Those criteria find rational, non-geographic justification in the national interest in economic growth, in the greater demand for housing cities with higher prices, and in the greater public tolerance for dense development in big cities. There is also ample precedent for our proposal in the numerous provisions of the U.S. tax code that distinguish among places.122

Second, opponents might argue that our proposal’s LIHTC restrictions unconstitutionally “commandeer” or “coerce” states take action, in violation of the 10th Amendment. This argument is almost certain to fail as well. Our proposal does not commandeer states because it does not mandate any state action — it lets states decide whether they would like to receive LIHTC generally and it lets big cities and/or their parent state decide whether to opt into the federal pro-housing rules. Nor does our proposal unconstitutionally coerce state action through incentives which are so overwhelmingly strong as to amount to a “gun to the head.”123 LIHTC allocations are minuscule in proportion to annual state budgets, so states and localities have a real choice as to whether they will accept the pro-housing conditions.124 Moreover, in contrast to the Medicaid funds at issue in NFIB v. Sebelius, LIHTCs are federal tax credits allocated directly to private developers rather than money provided to states; states have no ongoing budgetary commitments which rely on them. 

Finally, it might be argued that our proposal would violate the Supreme Court’s nondelegation doctrine. Under Art. I § 1, all federal legislative power is vested in the Congress, and the Supreme Court has long held that Congress may not delegate such legislative authority to any other actor. Traditionally, this has been an easy bar to clear. The Supreme Court has not struck down a federal statute for violating the nondelegation norm since 1935. However, the Court’s current conservative majority evidently wants to put stricter limits on Congress’s ability to authorize other actors to make legal rules with the force of law.125

Opponents might argue that our proposal unconstitutionally delegates by letting HUD revise or expand the pro-housing standards on the basis of flexible criteria, such as “reducing excessive regulatory barriers to housing development” or “reducing unnecessary complexity in federal pro-housing rules.” This argument is unlikely to succeed. According to its proponents, the nondelegation doctrine exists to ensure that Congress does not delegate to other actors the power to “adopt generally applicable rules of conduct governing future actions by private persons,” particularly when such rules are backed by, at minimum, enforcement mechanisms such as injunctive relief or monetary penalties.126 Our proposal would only empower HUD to propose pro-housing standards. Cities that accept the proposal would do so via the uncoerced choice of their democratically elected representatives. Further, the voluntarily accepted standards would not directly threaten individual liberty interests and would, in fact, enhance the property rights of affected landowners by removing local land-use regulations. This design in no way resembles the delegations of coercive, direct, liberty-threatening authority which the Court’s conservatives fear.

Emphasizing that the nondelegation doctrine has long required administrative actions to be guided by an “intelligible principle” set by Congress,127 opponents might also argue that our proposal leaves cities with standardless discretion to opt into the pro-housing rules. This critique confuses the standards required for a federal agency to exercise quasi-legislative rulemaking authority with those required for an independent sovereign’s participation in a cooperative program. Delegations of the latter sort require no congressional “standards” at all. The whole point is to let another democratically accountable actor decide on the basis of their own criteria whether to join in. Numerous federal statutory schemes, ranging from Medicaid to the Clean Air and Clean Water Acts, give states this choice. 

It bears mentioning too that Congress has sometimes conditioned the legal effect of statutes or administrative regulations on the approval of private entities. The Supreme Court upheld such “on-off switches” in a pair of 1939 decisions,128 and constitutional scholars have argued they are not delegations of legislative authority at all, but merely a way to incorporate local views into a wholly congressionally-designed scheme.129 If the local views are conveyed by a local or state government, as in our proposal, so much the better. The Court has been particularly deferential to delegations of authority to non-federal governments, including states and Native American tribes, due to their “attributes of sovereignty” and “independent authority.”130 In sum, we think the courts would view the discretion exercised by democratically elected local and state governments under our proposal not as congressional overreach, but rather as an effective limit on the scope of federal power.

3.5.6. State law may prevent some cities from opting in 

Some city councils may be constrained by city-charter provisions or other voter-adopted rules, enacted in the exercise of authority conferred by the state’s constitution, that would make it unlawful for the council to opt into the federal pro-housing rules. For example, San Francisco’s city attorney has long taken the position that the city’s charter prohibits ministerial approval of housing projects.131 In other cities, voter-adopted height or density limits, or voter-mandated inclusionary zoning, may prevent the city council from opting into the federal pro-housing rules. 

One way to handle such problems is to leave them to the states. If a city can’t opt in but the state wants it to opt in, the state legislature can opt it in or preempt the conflicting local requirements (to the extent that doing so is within the legislature’s constitutional authority). 

Another possibility for Congress is to conditionally preempt the contrary state or local law, by passing a statute whose preemptive effect is expressly suspended until a city council or other congressionally designated state or local actor says they approve of it. In effect, Congress would authorize the designated actor to opt a city into the federal pro-housing rules even if the actor lacks such authority under state law. Whether the Constitution allows Congress to do this is uncertain. Two state supreme courts have held that Congress may not “enlarge state gubernatorial power” by, for example, authorizing the governor to enter into compacts with native tribes to regulate casino gaming on tribal reservations.132 If this is right, presumably Congress may not enlarge the power of any other state or municipal actor either. The U.S. Supreme Court has also held that Congress may not regulate states or cities as such, as opposed to creating private rights which states and cities must honor.133 “Enlarging the authority” of a state or local government actor arguably violates this principle. 

On the other hand, several Supreme Court cases have upheld local government actions against contrary state laws when the local actions were performed pursuant to express federal authority.134 These decisions support the proposition that the federal government may provide powers to local governments, even when such powers conflict with relevant state or local constitutional, statutory, or charter law, provided that Congress speaks clearly.135

A court could square the circle by holding that if Congress conditions the preemptive effect of the federal law on the approval of a delegatee who happens to be a state or local official, that official acts as an agent of Congress rather than in their state-constitutional capacity when giving (or withholding) their assent. Whatever capacities they have under the constitution and laws of the state would by definition remain unchanged, not enlarged.136

3.5.7. Nobody knows if it will work

Perhaps the strongest objection to our proposal is that it’s too much of a shot in the dark. It borrows from the states, yet there is no model state code that’s been shown to demonstrably increase housing supply in big, high-price cities. It supposes that many big cities would accede to the pro-housing policies rather than lose their LIHTC eligibility, but interest groups that benefit from the big-city status quo could well prevail over affordable housing developers and other allies who want their city to opt in.

We think this objection is less a reason to give up on LIHTC conditionality than to design it in a manner that accounts for epistemic uncertainty. It is precisely because nobody knows the optimal set of federal policies for controlling local permitting, exactions, and zoning that we think a federal agency should be charged with measuring cities’ performance relative to peer jurisdictions and updating the pro-housing rules over time. 

The fact of serious epistemic uncertainty also makes it important to design new federal housing-supply policies in a manner that will enable researchers to plausibly estimate their causal effects. If Congress adopts our proposal, economists will be able to evaluate it by comparing housing production in similar census tracts in cities that are just above and just below the statutory cutoff for “big, expensive city.” The former tracts would be treated by the congressional policy; the latter are controls. However, if there are just a few cities near the cutoff, as seems likely, the resulting estimates of the average difference in production between treated and control units would be pretty uncertain.137 Another option is to match census tracts in all treated cities (not just those near the cutoff) to similar tracts in all control cities, and construct difference-in-difference estimates of housing production. If tracts in the big, expensive cities do not start yielding housing at higher rates than similar tracts in other cities, controlling for pre-treatment trends, that would be a flashing red signal that the LIHTC conditionality program is not working and should be revisited.138

It goes without saying that Congress should also revisit the program if most cities above the cutoff decline LIHTC rather than opting in. A low rate of participation wouldn’t impugn the pro-housing rules, but it would suggest that additional inducements are necessary. Perhaps other federal funding streams should be reserved for cities that opt in, or, if nonparticipating cities object to specific features of the pro-housing rules, perhaps those features could be sanded down.

Ultimately, the fact that there is not a proven state model for inducing big, expensive cities to allow a lot more housing speaks to the politics of the problem. The big, expensive cities are mostly located in Democratic-leaning states, and interest groups aligned with the Democratic party make it very hard to solve the “everything bagel” problem in supermajority-Democrat legislatures.139 Moreover, many of the people who would benefit from liberalizing housing supply in high-demand metros live in other states today.140 State legislators who recognize this point have even less reason to battle the parochial defenders of the status quo. Moving the fight to Congress would empower lawmakers who represent outsiders and are less beholden to urban interest groups.

  1. Affordable Housing Credit Improvement Act of 2023, S. 1557, 118th Cong., 1st sess. Introduced May 11, 2023.

  2. Evan Soltas, “Tax Incentives and the Supply of Low-Income Housing,” May 11, 2024, https://evansoltas.com/ (finding that LIHTC pulls investment forward in time but generates little net new housing).

  3. Soltas, supra, finds that about 50% of LIHTC tax expenditures are captured by developers, and that the effective fiscal cost of the LIHTC program is about a million dollars per net new housing unit. Another study estimates that tenants’ consumption of housing and other goods and services increases by less than $0.30 for every $1 in LIHTC expenditures, versus $0.77 for every $1 in spending on the federal housing-voucher program. Edgar O. Olsen and Dirk W. Early, “The Effects of U.S. Low-Income Housing Programs on Recipient Consumption and Wellbeing,” SSRN Scholarly Paper (Rochester, NY: Social Science Research Network, June 22, 2023).

  4. If LIHTC were replaced with a refundable tax credit, eligibility for the tax credit could also be restricted to renters who reside in places with elastic housing supply or in big cities that have opted into federal pro-housing rules. However, we think that unorganized low-income renters would be a less effective force (compared to firms that specialize in LIHTC projects) for lobbying cities to opt in.

  5. Cf. Jesse Clyde Nichols, Real Estate Subdivisions: The Best Manner of Handling Them., 11 5 (American Civic Association, 1912).(complaining that “good residence centers were being ruined by the building of tall flats, cutting off the air and sunshine from adjoining homes”).

  6. Kenneth T Jackson, Crabgrass Frontier: The Suburbanization of the United States (Oxford University Press, 1987).

  7. Jacob Anbinder, “Building Ecotopia: Environmentalism, Liberalism, and the Making of Antigrowth Political Culture in California, 1950–1990,” in Berkeley Housing Politics and Policy Conference, 2023.

  8. Nathaniel Baum-Snow and Lu Han, “The Microgeography of Housing Supply,” Journal of Political Economy 132, no. 6 (June 2024): 1897–1946; Anthony W. Orlando and Christian L. Redfearn, “Housing Supply Elasticities: A Structural Vector Autoregression Approach,” May 2024; Knut Are Aastveit, Bruno Albuquerque, and André K. Anundsen, “Changing Supply Elasticities and Regional Housing Booms,” Journal of Money, Credit and Banking 55, no. 7 (2023): 1749–83; Albert Saiz, “The Geographic Determinants of Housing Supply*,” The Quarterly Journal of Economics 125, no. 3 (August 1, 2010): 1253–96; “The Impact of Greenbelts on Housing Markets: Evidence from Toronto” (University of Toronto, January 26, 2024).

  9. Nathaniel Baum-Snow, “Constraints on City and Neighborhood Growth: The Central Role of Housing Supply,” Journal of Economic Perspectives 37, no. 2 (May 2023): 53–74. Similarly, from 2000 - 2020, 42% of all housing growth occurred in unincorporated areas, “despite the fact that unincorporated areas are home to only 23% of the overall population, are typically low-demand regions with low prices and rents, face high vacancy rates, and are far from municipal job centers and amenities” Alexander Bartik, Arpit Gupta, and Daniel Milo, “The Costs of Housing Regulation: Evidence From Generative Regulatory Measurement,” September 14, 2024.

  10. Ibid.

  11. Anthony W. Orlando and Christian L. Redfearn, “Houston, You Have a Problem: How Large Cities Accommodate More Housing,” Real Estate Economics 52, no. 4 (2024): 1045–74.

  12. David Broockman, Christopher S. Elmendorf, and Joshua Kalla, “The Symbolic Politics of Housing” (2024); William Marble and Clayton Nall, “Where Self-Interest Trumps Ideology: Liberal Homeowners and Local Opposition to Housing Development,” The Journal of Politics 83, no. 4 (October 2021): 1747–63; Stephanie Ternullo, “The Politics of Concentrated Advantage,” March 20, 2024; Martin Vinæs Larsen and Niels Nyholt, “Understanding Opposition to Apartment Buildings,” May 3, 2024.

  13. Broockman, Elmendorf, and Kalla, “The Symbolic Politics of Housing.”

  14. Luis Ferré-Sadurní and Mihir Zaveri, “A Plan to Force More Housing Development in New York Has Failed,” The New York Times, April 21, 2023.

  15. Grace Ashford, Dana Rubinstein, and Claire Fahy, “How Governor Hochul Decided to Kill Congestion Pricing in New York,” The New York Times, June 9, 2024, Ashford, Rubinstein, and Fahy.

  16. Christopher S. Elmendorf and Clayton Nall, “Plain-Bagel Streamlining? Notes from the California Housing Wars,” Case Western Reserve Law Review, forthcoming.

  17. Ryan Greenaway-McGrevy and Peter C.B. Phillips, “The Impact of Upzoning on Housing Construction in Auckland,” Journal of Urban Economics 136 (2023): 103555; Ryan Greenaway-McGrevy, “Can Zoning Reform Reduce Housing Costs? Evidence from Rents in Auckland,” n.d.; Eleanor West and Marko Garlick, “Upzoning New Zealand,” Works in Progress 13 (November 2023).

  18. Emily Hamilton, “Learning from Houston’s Townhouse Reforms,” Policy Brief (Mercatus Center, George Mason University, April 11, 2023); Anya Martin, “Houston, We Have a Solution,” September 7, 2023.

  19. Muhammad Alameldin and Quinn Underriner, “San Diego’s Success in Spurring Missing Middle Housing: The Accessory Dwelling Unit Bonus Program,” Terner Center (blog), February 15, 2023.

  20. Leah Brooks and Byron Lutz, “From Today’s City to Tomorrow’s City: An Empirical Investigation of Urban Land Assembly,” American Economic Journal: Economic Policy 8, no. 3 (August 2016): 69–105; Rachel Gallagher, Yan Liu, and Thomas Sigler, “Parcel Amalgamation as a Mechanism for Achieving Urban Consolidation Through Densification: The Fixity of Property Boundaries Over Time,” Land Use Policy 89 (December 1, 2019): 104239.

  21. Robert C. Ellickson, “Suburban Growth Controls: An Economic and Legal Analysis,” Yale Law Journal 86, no. 3 (1977): 385–511.

  22. Harvey Molotch, “The City as a Growth Machine: Toward a Political Economy of Place,” American Journal of Sociology 82, no. 2 (1976): 309–32.

  23. Advisory Commission on Regulatory Barriers to Affordable Housing, “‘Not In My Back Yard’: Removing Barriers to Affordable Housing” (Washington, D.C.: U.S. Department of Housing and Urban Development, 1991).

  24. Nathaniel Baum-Snow and Daniel Hartley, “Accounting for Central Neighborhood Change, 1980–2010,” Journal of Urban Economics 117 (May 1, 2020): 103228.

  25. Vicki Been, “Community Benefits: A New Local Government Tool or Another Variation on the Exactions Theme Symposium: Reassessing the State and Local Government Toolkit,” University of Chicago Law Review 77, no. 1 (2010): 5–36; Michael Hankinson and Justin de Benedictis-Kessner, “How Self-Interest and Symbolic Politics Shape the Effectiveness of Compensation for Nearby Housing Development,” Journal of Public Policy, October 21, 2024, 1–24; Michael Hankinson, Asya Magazinnik, and Anna Weissman, “When Do Local Interest Groups Participate in the Housing Entitlement Process?,” Journal of Political Institutions and Political Economy 5, no. 1 (2024): 47–69.

  26. This assumes that the local interests engaged with a project just want money or in-kind benefits, rather than to stop or downsize the project. If there are organized neighbors or others who want to block projects, then a process that invites case-by-case negotiation would probably deter development even if the city council had perfect information about project surpluses and the authority to make interest groups accept the terms of any CBA imposed by the council. Though it’s a useful approximation to think of big-city housing politics as dominated by groups angling to capture value from new projects, and of suburban housing politics as dominated by residents angling to control the size and density of new projects, many of the politically engaged people who live in big cities’ more residential areas also have concerns about size and density, even though they’re more open to density than their suburban counterparts. Vicki Been, “City NIMBYs,” Journal of Land Use & Environmental Law 33, no. 2 (2018): 217–50; Alexander Sahn, “Public Comment and Public Policy,” American Journal of Political Science, August 31, 2024.

  27. Minjee Kim, “Upzoning and Value Capture: How U.S. Local Governments Use Land Use Regulation Power to Create and Capture Value from Real Estate Developments,” Land Use Policy 95 (June 1, 2020): 104624, (finding, in survey of large development projects in five major U.S. cities, that 90% were required to provide public benefits as a condition of entitlement, and that 73 of 90 were required to provide ad hoc benefits rather than benefits in accordance with a predetermined schedule). Kim concludes, “none of the cities had clear standards or evaluation frameworks for determining: how much value was created, what can be asked for in return, and who should benefit from the value captured” (p. 1).

  28. Christopher S. Elmendorf and Timothy G. Duncheon, “When Super-Statutes Collide: CEQA, the Housing Accountability Act, and Tectonic Change in Land Use Law,” Ecology Law Quarterly 49, no. 3 (2023 2022): 655–714; Moira O’Neill et al., “Examining Entitlement in California to Inform Policy and Process: Advancing Social Equity in Housing Development Patterns,” Report to the California Air Resources Board, 2022.

  29. Money has a time value above and beyond explicit project carrying costs. Consider an imaginary construction project that costs $100,000, takes one year to build, owes no out-of-pocket interest costs during construction, and then earns $10,000 per year in perpetuity after completion. The IRR on that 1-year project is 9.9%, falling to 8.3% after 3 years of delay, or 7.3% after 5 years of construction time even in a hypothetical scenario with zero out-of-pocket interest costs. A recent paper estimates that a 25% reduction in approval timelines in Los Angeles would increase housing production by 33%. Stuart Gabriel and Edward Kung, “Development Approval Timelines, Approval Uncertainty, and New Housing Supply: Evidence from Los Angeles,” June 18, 2024.

  30. Chris Elmendorf and Darien Shanske, “Auctioning the Upzone: A New Strategy to Induce Local Government Compliance with State Housing Policies” (California Environmental Law & Policy Center, UC Davis School of Law, December 2018), p. 10.

  31. Elmendorf and Nall, “Plain-Bagel Streamlining?”

  32. Many cities have analogous local rules that raise the cost of development, such as by limiting what can be built, via stringent zoning, unless the developer agrees to provide public benefits in accordance with a predetermined schedule. A recent study finds that 41% of U.S. cities practice such “incentive zoning.” George C Homsy and Ki Eun Kang, “Zoning Incentives: Exploring a Market-Based Land Use Planning Tool,” Journal of the American Planning Association 89, no. 1 (2023): 61–71.

  33. Ezra Klein, “The Economic Mistake the Left Is Finally Confronting,” New York Times, September 19, 2021.

  34. Elmendorf and Nall, “Plain-Bagel Streamlining?”

  35. Nicholas J. Marantz, Christopher S. Elmendorf, and Youjin B. Kim, “Where Will Accessory Dwelling Units Sprout Up When a State Lets Them Grow? Evidence From California,” Cityscape 25, no. 2 (2023): 107–18.

  36. Christopher S. Elmendorf and Darien Shanske, “Tax Development, or What? Fiscal Foundations for the Next Era of Land Use Regulation in California,” SSRN Scholarly Paper (Rochester, NY: Social Science Research Network, May 28, 2023); Christopher S. Elmendorf and Darien Shanske, “Auctioning the Upzone,” Case Western Reserve Law Review 70 (2020): 513–72; Alex Armlovich and Andrew Justus, “An Agenda for Abundant Housing” (Niskanen Center, February 2023).

  37. A land-value or parcel tax would have to be paid regardless of whether the owner develops, so it cannot make (re)development less profitable relative to maintaining the current use of the site. The auction method of capturing value ensures that the price of development allowances does not exceed the site-value surplus on marginal sites, since landowners would not bid more than the surplus for the allowances.

  38. Mark P. Keightley, “An Introduction to the Low-Income Housing Tax Credit” (Congressional Research Service, April 26, 2023).

  39. Economists have critiqued the LIHTC program for subsidizing construction in markets where there’s no shortage of housing. See Edward Glaeser and Joseph Gyourko, Rethinking Federal Housing Policy (American Enterprise Institute, 2008). One might worry that our proposal would exacerbate that problem by disqualifying certain high-demand central cities. However, projects in the suburbs of disqualified high price cities would remain eligible. Ultimately, the problem identified by Glaeser and Gyourko would be better addressed by restricting LIHTC subsidies to high-demand regions, or by limiting new-build & requiring “fix-it-first” rehabilitation use of LIHTC in below-replacement cost regions, independently of whether our proposal for central cities is adopted.

  40. Top-performing cities might be deemed eligible regardless of whether they opt in. See section 3.5.3.

  41. Clancy Mullen, “2019 National Impact Fee Survey,” August 2019, (reporting average total impact fee on a multifamily unit of $8054, which in 2024 dollars is about $10,000).

  42. We would also invite states to act on behalf of their cities and opt them in.

  43. Advisory Commission on Regulatory Barriers to Affordable Housing, “‘Not In My Back Yard’: Removing Barriers to Affordable Housing,” p. 10.

  44. Ibid.

  45. 42 U.S.C. § 12711. There is an exception if the local policy or law violates a specific federal law.

  46. A recent study finds that more than 40% of all cities (with population of 5000 or greater) are already conditioning zoning liberalization on developers’ agreement to provide cash or in-kind benefits. Homsy and Kang, “Zoning Incentives: Exploring a Market-Based Land Use Planning Tool.”

  47. The following discussion of California ADU law was previously published in substantially similar form in Elmendorf “Beyond the Double Veto: Housing Plans as Preemptive Intergovernmental Compacts,” Hastings Law Journal 71 (2019): 79–150. Please refer to the original for citations.

  48. 2002 Cal. Leg. Ch. 1062 (A.B. 1866), § 2.

  49. “A Room of One’s Own? Accessory Dwelling Unit Reforms and Local Parochialism,” Urban Lawyer 45 (2013): 519–69.

  50. Marantz, Elmendorf, and Kim, “Where Will Accessory Dwelling Units Sprout Up When a State Lets Them Grow?”

  51. “Field preemption” is a legal term of art for action by a higher level of government that entirely supplants regulation by a lower level of government over a specified field of activity.

  52. How the Next President Can Solve America’s Housing Crisis,” Substack newsletter, Agglomerations (blog), October 16, 2024, Ozimek and Lettieri would allow cities to opt in for only part of their territory, such as by establishing a special redevelopment district where the federal rules apply. Any local government that opts in would receive a “‘Density Dividend’ [payments from the federal government] proportional to the number of new housing units completed.

  53. However, because local governments would retain control of permitting, there is some risk that the federal zoning standards could be thwarted by bad-faith administration.

  54. Sam Jacobson, “Expand the Fair Housing Initiatives Program to Enforce Federal and State Housing Supply and Affordability Laws,” Federation of American Scientists (blog), February 19, 2024.

  55. Leonardo D’Amico et al., “Why Has Construction Productivity Stagnated? The Role of Land-Use Regulation,” SSRN Scholarly Paper (Rochester, NY: Social Science Research Network, December 30, 2023).

  56. Congress could, in principle, decide to preclude judicial review of the model codes. We would support that approach, particularly since the model codes would only apply in jurisdictions that opt in, but eliminating judicial review of the model codes would probably make the political lift even harder.

  57. To increase the incentive to opt in, Congress could also tie the allocation of Housing Trust Fund (HTF) subsidies to large cities’ acceptance of the pro-housing conditions. Established in 2008, HTF is funded by a percentage set aside from each residential mortgage purchased by Fannie Mae and Freddie Mac. HTF provides an annual formula allocation to states to implement subsidies for construction of housing affordable to households earning 30-50% AMI. Due to its funding mechanism, the annual allocations fluctuate from year to year, with HTF distributing $748 million to states in FY 2022 and $214 million in FY 2024 (United States Department of Housing and Urban Development 2024). In 2024, the minimum state allocation was $3.1 million and the highest state allocation, to California, was $21.6 million (Housing Trust Fund: Fiscal Year 2024 Allocation Notice, 58 Fed. Reg. 58391 (Jul. 18, 2024)).

  58. Department of Administrative Services (DAS), “Oregon Housing Needs Analysis: Draft Methodology” (Oregon, USA: Oregon Department of Administrative Services, September 2024).

  59. E.g., Baum-Snow and Han, “The Microgeography of Housing Supply.”

  60. Cal. Gov’t Code §§ 65589.5(o), 65941.1.

  61. Cal. Gov’t Code § 65589.5(j). There is a health/safety exception.

  62. Cal. Gov’t Code § 65589.5(f)(4).

  63. Cal. Gov’t Code § 65913.4(a).

  64. Cal. Gov’t Code § 65589.5(j); California Renters Legal Advoc. & Educ. Fund v. City of San Mateo, 68 Cal. App. 5th 820, 846, 283 Cal. Rptr. 3d 877, 895 (2021).

  65. Elmendorf and Duncheon, “When Super-Statutes Collide”; Moira O’Neill and Ivy Wang, “How Can Procedural Reform Support Fair Share Housing Production? Assessing the Effects of California’s Senate Bill 35,” Cityscape 25, no. 2 (2023): 143–70.

  66. Cal. Gov’t Code § 65956(b); Ciani v. San Diego Trust & Savings Comm’n, 233 Cal. App. 3d 1604, 285 Cal. Rptr. 699(1991); California YIMBY Webinar, “Implementing AB 1663.”

  67. Cal. Gov’t Code § 65589.5(k)(1)(A)(ii) & (l).

  68. Cal. Gov’t Code § 65589.5(k)(1)(A)(ii).

  69. Cal. Gov’t Code § 65589.5(m)(1).

  70. Cal. Gov’t Code § 65589.5((k)(1)(A)(ii)(II).

  71. Salim Furth, Emily Hamilton, and Charles Gardner, “Housing Reform in the States: A Menu of Options for 2025” (Mercatus Center, George Mason University, August 2024).

  72. Inspection Agencies: List of IPIAs and DAPIA’s,” HUD.gov / U.S. Department of Housing And Urban Development (HUD), accessed November 1, 2024.

  73. Andre Sorensen, Junichiro Okata, and Sayaka Fujii, “Urban Renaissance as Intensification: Building Regulation and the Rescaling of Place Governance in Tokyo’s High-Rise Manshon Boom,” Urban Studies 47, no. 3 (2010): 556–83.

  74. Cities and their consultants are often quite transparent that the purpose of a nexus study is to justify the highest possible fee, and that the actual fee is chosen on some other basis. Elmendorf and Shanske, “Auctioning the Upzone,” pp. 525-27; Hayley Raetz et al., “Residential Impact Fees in California: Current Practices and Policy Considerations to Improve Implementation of Fees Governed by the Mitigation Fee Act” (Terner Center for Housing Innovation, University of California, Berkeley, August 2019), https://ternercenter.berkeley.edu.

  75. Trevor Stockinger et al., “The Impact of Fees: Rethinking Local Revenues for More Multifamily Housing” (California YIMBY Education Fund, April 2024).

  76. Christopher B. Goodman and Megan E. Hatch, “State Preemption and Affordable Housing Policy,” Urban Studies, 2022, 1–18.

  77. California, Oregon, and New Jersey exemplify this approach. Massachusetts, Rhode Island, Illinois, and Connecticut also strongly encourage local inclusionary zoning by exposing cities to a “builder’s remedy” if the city doesn’t ensure that at least 10% of its housing stock consists of deed-restricted affordable housing units. See generally Elmendorf, “Beyond the Double Veto: Housing Plans as Preemptive Intergovernmental Compacts.”

  78. The same can be said about limiting fees and exactions without ensuring an adequate supply of liberally zoned land. A recent study finds that so-called “incentive zoning” (limiting what can be built unless the developer agrees to provide public benefits) is more common in cities whose regulatory authority has been hemmed in by their parent state. Homsy and Kang, “Zoning Incentives: Exploring a Market-Based Land Use Planning Tool.” One of the local officials interviewed by Homsy and Kang frankly acknowledged their incentive-zoning scheme was designed to evade state-law restrictions on inclusionary zoning (p. 68).

  79. Alan A. Altshuler and José A. Gómez-Ibáñez, Regulation for Revenue: The Political Economy of Land Use Exactions (Brookings Institution Press and Lincoln Institute of Land Policy, 1993).

  80. Elmendorf and Shanske, “Auctioning the Upzone”; Elmendorf and Shanske, “Tax Development, or What?” An interesting question (though beyond the scope of this paper) is whether Congress could or should authorize local governments to charge a land value tax, notwithstanding property-tax caps or uniformity provisions in a state constitution that may preclude it.

  81. Tony Strickland, “I Will Always Champion Local Control and Firmly Oppose Rent Control,” Orange County Register, April 12, 2024.

  82. Congress might also stipulate the fees be pegged to unit size and capped at, say, $8.33 per square foot (equivalent to $10,000 for a 1200 square foot dwelling). This would keep cities from making it disproportionately expensive to build smaller homes.

  83. Mullen, “2019 National Impact Fee Survey.”

  84. Assemb. Bill 1893, 2024 Leg., Reg. Sess. (Cal. 2024); Gov’t Code § 65589.5(f)(6)(G).

  85. Cal. Gov’t Code § 65589.5(f)(6)(G)(III).

  86. Konstantin A. Kholodilin, “Rent Control Effects Through the Lens of Empirical Research: An Almost Complete Review of the Literature,” Journal of Housing Economics 63 (March 1, 2024): 101983.

  87. Christopher S. Elmendorf, Clayton Nall, and Stan Oklobdzija, “What State Housing Policies Do Voters Want? Evidence from a Platform-Choice Experiment,” Journal of Political Institutions and Political Economy, 2024.

  88. Cal. Gov’t Code § 65915. The size of the increment varies with the share of deed-restricted units in the project and the affordability of those units.

  89. Fla. Stat. § 166.04151(7)(b).

  90. Fla. Stat. § 166.04151(7)(c).

  91. Fla. Stat. § 166.04151(7)(d).

  92. Fla. Stat. § 166.04151(7)(d).

  93. Cal. Gov’t Code § 65912.123.

  94. Cal. Gov. Code § 65912.113(f).

  95. Assemb. Bill 1893, 2024 Leg., Reg. Sess. (Cal. 2024); Cal Gov’t Code § 65589.5(f)(6)(A)) For an explanation of the process for determining the applicable standards, see https://x.com/CSElmendorf/status/1827084471881101603.

  96. Cal. Gov’t Code § 65915(e).

  97. Bankers Hill 150 v. City of San Diego, 74 Cal.App.5th 755 (2022); Wollmer v. City of Berkeley, 193 Cal.App.4th 1329 (2011).

  98. Cal. Gov’t Code § 66300(b) & (i).

  99. Performance-based evaluations should help to reduce gaming at least on the margins, but for reasons we will discuss in section 3.5.3, it is not a full substitute for the federal pro-housing rules.

  100. Department of Administrative Services (DAS), “Oregon Housing Needs Analysis: Draft Methodology.”

  101. Oregon Department of Land Conservation and Development, “Draft Oregon Administrative Rules Chapter 660, Division 008".

  102. Ibid. (proposed OAR 660-008-0335(4)(b).)

  103. Baum-Snow, “Constraints on City and Neighborhood Growth”; Hempel, “The Impact of Greenbelts on Housing Markets: Evidence from Toronto”; Salim Furth, “Housing Supply in the 2010s,” Mercatus Research Paper, February 14, 2019.

  104. The relevant characteristics also include demand-side attributes like housing prices in the previous time period and exposure to demand shocks during the current time period.

  105. For another peer-benchmarking model, see Salim Furth et al., “HUD Can Use Housing Market Data to Inform Fair Housing Accountability” (Mercatus Center, George Mason University, March 2020).

  106. If big cities’ LIHTC eligibility depended on these reports, they might shade the numbers or simply decline to answer the survey.

  107. For related proposals, see Brian J. Connolly, Heidi Aggeler, and Avila Bueno, “Develop a Housing Production Dashboard to Aid Policymaking and Research,” Federation of American Scientists (blog), February 23, 2024; Nicholas J. Marantz, “Harnessing Federal Programs to Improve Local Housing Permit Data,” Federation of American Scientists (blog), February 21, 2024.

  108. Robert Collinson, Ingrid Gould Ellen, and Jens Ludwig, “Reforming Housing Assistance,” The ANNALS of the American Academy of Political and Social Science 686, no. 1 (November 1, 2019): 250–85.

  109. Christopher Bonastia, Knocking on the Door: The Federal Government’s Attempt to Desegregate the Suburbs (Princeton University Press, 2006).

  110. Benefits to tenants from the LIHTC program as presently constituted appear to be modest. See Soltas, “Tax Incentives and the Supply of Low-Income Housing.” (estimating that about half of the subsidy is captured by developers as profit and a quarter is dissipated in compliance costs).

  111. Evan Mast, “The Effect of New Market-Rate Housing Construction on the Low-Income Housing Market,” Journal of Urban Economics, 2021, 103383; Cristina Bratu, Oskari Harjunen, and Tuukka Saarimaa, “JUE Insight: City-Wide Effects of New Housing Supply: Evidence from Moving Chains,” Journal of Urban Economics 133 (2023): 103528; Andrii Parkhomenko, “Local Causes and Aggregate Implications of Land Use Regulation,” Journal of Urban Economics 138 (November 1, 2023): 103605.

  112. Kevin Corinth and Amelia Irvine, “The Effect of Relaxing Local Housing Market Regulations on Federal Rental Assistance Programs,” Journal of Urban Economics 136 (July 1, 2023): 103572, (estimating that if Los Angeles produced new housing units at the same rate as the 90th percentile metropolitan area for a decade, federal cost savings in housing-choice vouchers would equal $353 million, enough to increase the number of assisted families by nearly 25%; and that doubling the number of units placed in service through the Low Income Housing Tax Credit for a decade would yield only about 1/20th of these cost savings).

  113. We suspect that few top performers would accept the invitation, because the federal rules should help to forestall the emergence and growth of local “value capture” groups — an outcome that representatives of the top performers would presumably like to avoid.

  114. Consider California’s rubric for scoring pro-housing applications, codified as 25 Cal. Code Regs. § 6606(b). The policies that earn the highest scores are: zoning for 150% of the city’s housing target (without regard for fees, approval processes, or any other restrictions that apply to these sites); allowing development of 3-4 unit dwellings by right in single-family areas (without regard to the size of the area in which 3-4 unit buildings are allowed, and the fees and development standards applies to their development); establishment of ministerial approval processes for multiple housing types (without regard to size of area zoned for such types, or the fees and development standards applied to their development); waiver or significant reduction of impact fees on projects with deed-restricted affordable housing (without regard to whether the affordability restrictions are so severe as to make the projects infeasible regardless of the fees; without regard to the size of the district in which the fee-reductions apply; without regard to the liberality or restrictiveness of the zoning in these districts; and without regard to whether projects in these districts are subject to ministerial approval).

  115. United States v. Ptasynski, 462 U.S. 74, 82 (1983). Under this interpretation, Congress can exempt specific categories of Alaskan oil from taxation (Ptasynski) or implement a SALT cap that disproportionately affected high-tax jurisdictions (e.g., New York v. Mnuchin, 408 F.Supp.3d, 399, 420 (S.D.N.Y. 2019)).

  116. E.g., 26 USC § 45(b)(11)(B) (establishing place-based definition of “energy community” for certain tax credits); 26 USC § 48E(a)(3)(A) (providing larger credits to “energy communities,” as defined); 26 USC § 142(n) (providing special broadband subsidies to projects in places defined by characteristics of Census block groups); Daniel J. Hemel, “A Place for Place in Federal Tax Law,” Ohio Northern University Law Review 45 (2019): 525--541 (surveying and assessing place-based federal tax policies).

  117. Nat’l Federation of Indep. Bus. v. Sebelius, 567 U.S. 519, 542 (2012).

  118. And even if the incentive were thought to be a “gun to the head,” our proposed statute would be easy to distinguish from the Medicaid expansion that the Supreme Court invalidated in Sebelius. The difference is that Congress has authority under the Commerce Clause to preempt local regulation of housing development — and thus to force local governments’ housing permitters to comply with federal pro-housing rules, rather than incentivizing cities to opt in — whereas Congress does not have any source of authority to make state officials implement federal healthcare programs. Cf. Hodel v. Virginia Surface Mining, 452 U.S. 264 (1981) (rejecting a commandeering challenge to a federal statute that gave states the choice between (1) designing and implementing a federally-compliant regime to regulate surface mining, or (2) yielding to federal regulation of surface mining). As the Hodel Court remarked, "Congress could constitutionally have enacted a statute prohibiting any state regulation of surface coal mining. We fail to see why the Surface Mining Act should become constitutionally suspect simply because Congress chose to allow the States a regulatory role." Id. at 290.

  119. Gundy v. United States, 588 U.S. 128, 157-59 (2018) (Gorsuch, J., dissenting).

  120. Gundy at 153 (Gorsuch, J., dissenting).

  121. J. W. Hampton, Jr. & Co. v. United States, 276 U.S. 394 (1928); United States v. Rock Royal Cooperative, Inc., 307 U.S. 533 (1939).

  122. Currin v. Wallace, 306 U.S. 1 (1939).

  123. Neil Kinkopf, “Of Devolution, Privatization, and Globalization: Separation of Powers Limits on Congressional Authority to Assign Federal Power to Non-Federal Actors,” Rutgers Law Review 50 (1998): 331–64; Alexander Volokh, “The Myth of the Federal Private Nondelegation Doctrine,” Notre Dame Law Review 99 (2023): 203–32.

  124. United States v. Mazurie, 419 U.S. 544, 556-57 (1975).

  125. Christopher S. Elmendorf, “Lawyering Cities into Housing Shortages: The Curious Case of Discretionary Review Under the San Francisco City Charter,” N.Y.U. Environmental L.J. 32 (2024): 291–337.

  126. State ex rel. Clark v. Johnson, 904 P.2d 11, 21 (N.M. 1995); Fla. House of Representatives v. Crist, 999 So.2d 601, 613 (Fla. 2008)

  127. Murphy v. Nat’l Collegiate Athletic Ass’n, 584 U.S. 453 (2018).

  128. City of Tacoma v. Taxpayers of Tacoma, 357 U.S. 320, 341 (1958) (affirming 9th Circuit’s decision on procedural grounds, but agreeing with 9th Circuit’s substantive ruling that an FPC license to the City of Tacoma to operate a dam preempted contrary state laws and empowered local action which was not otherwise authorized under state law); Lawrence County v. Lead-Deadwood School Dist. No. 40-1, 469 U.S. 256, 269-70 (1985) (upholding local use of funds directly allocated by federal grant against contrary state law, which sought to limit the locality’s use of funds); Missouri v. Jenkins, 495 U.S. 33, 57-58 (1990) (concluding that a District Court could enjoin application of a state law which set a ceiling on local property tax rates, such that a local government could raise its property taxes in order to comply with a judicial remedy to the locality’s 14th Amendment Equal Protection Clause violation).

  129. Roderick M. Hills Jr., “Dissecting the State: The Use of Federal Law to Free State and Local Officials from State Legislatures’ Control,” Michigan Law Review 97, no. 5 (1999): 1201–88.

  130. Another way to square the circle would be for state courts to adopt an interpretive presumption that state laws limiting a state or municipal actor’s authority do not apply to actions expressly authorized by Congress, unless the limiting state law was unmistakably intended to address actions authorized by Congress.

  131. Also, if Congress designs the cutoff with the goal of pulling in specific cities that are believed to be bad actors, that would invalidate the assumption that a near-cutoff city’s position on one side of the cutoff is as-if random, rendering it a good counterfactual for near-cutoff cities on the other side.

  132. It would not be very sensible to evaluate the program by comparing production in cities that opted in with production in cities that did not, as opting in is likely to be correlated with local political support for housing production. Using this design, the apparent “treatment effect” of the congressional policy might just be an artifact of political changes that were occurring anyway in cities that opted in, rather than the actual effect of the congressional policy relative to counterfactual production in the absence of the policy.

  133. Klein, “The Economic Mistake the Left Is Finally Confronting”; Elmendorf and Nall, “Plain-Bagel Streamlining?”

  134. Gilles Duranton and Diego Puga, “Urban Growth and Its Aggregate Implications,” Econometrica 91, no. 6 (2023): 2219–59, https://doi.org/10.3982/ECTA17936; Parkhomenko, “Local Causes and Aggregate Implications of Land Use Regulation.”