The interest of Carroll Gardens’ rezoning is easy to miss, because the moral of the story lies in what was left unsaid by the city’s land-use authorities. During the entire lengthy process, no one calculated — or even mentioned — the potential housing units lost as a result of the downzoning. This omission was especially odd because the affected sites were prime locations for housing, being only a few blocks from subway lines leading to Manhattan and Queens, close to retail, and already occupied by sound residential structures that needed only to be enlarged rather than demolished to accommodate additional occupants. Yet the participants — from the community board, to the planning commission, to the city council — all completely ignored the need for housing, supporting the proposed rezoning solely because it would preserve existing neighborhood character.

That the city would downzone prime residential land without any comment on the consequent loss of housing opportunities is extremely strange. Mayor Bloomberg proposed a long-term plan for New York City — “PlaNYC 2030” — prior to his 2009 reelection that called for the creation of 265,000 new housing units by 2030 to accommodate an estimated 1 million new residents. It is safe to say that the Bloomberg administration and the city that elected him have made a political commitment to increasing the supply of housing, a policy choice that responds to the high cost of housing in New York, where the average apartment in Manhattan cost $1.3 million in 2010 even after the housing crisis, and the average citywide was $854,000.

However strange, the downzoning of prime residential land without considering its effect on the housing supply is not anomalous. Some 59 percent of the Bloomberg administration’s downzonings have eliminated housing in areas served by mass transit, and a disproportionate number of downzonings have eliminated housing densities in neighborhoods that seemed in heavy demand. None of the downzonings seem to be accompanied by any conscious or systematic effort to provide counterbalancing upzonings.

Why do city decisionmakers ignore downzonings’ impact on housing supply? Our central claim is that this sort of neglect of housing need is caused by the piecemeal and seriatim process by which zoning decisions are made — a process creating systemic incentives to impose excessive restrictions on the supply of housing. On any given zoning vote, the supporters of restrictive zoning have an advantage over the supporters of additional housing supply even when less restrictive zoning across a given local government might be preferred by city residents. In effect, local governments exceed their “zoning budgets,” imposing restrictions in excess of what their own planners and politicians declare to be the optimal amount of regulation, because land-use regulation procedure causes them to ignore the long-term effects of individual zoning decisions.

We argue that the policy response to these procedurally induced incentives to downzone land should be to impose actual zoning budgets. The mayor should have to propose a planned annual growth (or decline) of the housing supply at the beginning of the year, which the city council would then accept or reject. All downzonings would have to be matched with some greater percentage of upzonings until the target is hit, and then matched with upzonings one-for-one.

Would such a purely procedural change really make a difference in zoning outcomes? To get a sense of the power of process, we examine the American experience with the closing of obsolete military bases and ratifying reciprocal tariff agreements. In both of those examples, Congress delegated to an extra-legislative agency (a base-closing commission and the president, respectively) the task of packing together decisions on multiple issues, with Congress reserving for itself only the power to ratify the entire package without amendments.

This extra-legislative packaging of base-closing reduced incentives for each member of Congress to save his own military base through log-rolling by assuring him that he would not be left out of a majority coalition of lawmakers seeking to increase their share of base money at non-coalition members’ expense. In the case of reciprocal tariff agreements, the “fast-track” procedure bundled together reductions of both foreign and domestic tariffs, thereby recruiting exporters to fight against and neutralize protectionist constituencies who might otherwise defeat the treaty.

Using the examples of base-closing and reciprocal trade agreements as models, we suggest that extra-legislative procedure can produce similar benefits for zoning, and for similar reasons. Existing “fair share” systems for allocating affordable housing or other locally unwanted land uses (“LULUs,” in land-use speak) among local governments or neighborhoods constitute a version of issue-bundling closely analogous to the base closing commission: by relieving each legislator of the fear that his district will be the dumping ground for LULUs that other jurisdictions avoid, the bundle reduces lawmakers’ incentives to roll logs to fight off all LULUs whatsoever. To deal with the general anti-development politic caused by homeowners across the city disliking increases in housing supply, a procedure similar to reciprocal trade treaties is needed. By requiring downzonings to be matched with upzonings, our zoning budget proposal would recruit developers to fight for general increases in the housing supply, just as reciprocal trade treaties link the interests of exporters to the fight for reduced import tariffs.

We offer this “coupling mechanism” as a heuristic rather than a strict recipe. The important point is that it provides a method for pitting geographically concentrated interest groups against each other in ways that it would be difficult for a legislature to undo. In urban legislatures, where political party competition is weak, legislative procedure can balance specific interest groups and force legislatures to consider more general interests. Reform efforts inside cities should focus on changing the politics of urban land use when they have the opportunity, rather than making one-time policy changes that will be whittled away over time.

The Public Choice of Land-Use Decisions

The sides in virtually all land-use disputes are the same. On one side are incumbent property owners seeking to limit or stop new development. On the other are renters, future residents, and — crucially — developers. When zoning decisions are made seriatim, and particularly when individual developers have no existing interest in downzoned land, this is hardly a fair fight. The benefits of new development are dispersed both geographically and across many individuals. In contrast, the costs are concentrated on neighbors who have a great deal invested in the outcome of land-use decisions.

That land-use disputes involve geographically concentrated harms and widely geographically dispersed benefits should be clear. Virtually all of the slang acronyms used in debates about land use — Not in My Back Yard (NIMBY), LULUs, Build Absolutely Nothing Anywhere Near Anyone (BANANA), Citizens Against Virtually Everything (CAVE people), etc. — are premised on the idea that neighbors resist locally concentrated costs of regionally beneficial land uses. Homeowners in the neighborhood of a development may see fewer scenic views, increased shadow, more traffic and less parking, more children in their school’s catchment area, or simply more people when they would rather see fewer. More importantly, they will see more competition for buyers and renters of housing, as a home in their neighborhood — which, for most homeowners, is their largest asset by a huge margin — becomes less scarce.

As Mancur Olson famously argued, smaller groups facing lower costs of organization and higher payoffs from political success will have an easier time overcoming collective action problems of becoming active in politics. And the situation of homeowners’ abutting a proposed residential development presents an Olsonian perfect storm. Homeowners are easy to organize because they are physically close together and thus can easily monitor each other’s contributions to the anti-NIMBY cause. Moreover, to the extent that the neighbors are also owner-occupiers of their structure, their large and undiversified investment in their home gives them big incentives to be politically active to protect their investment from neighborhood change. The beneficiaries of development are theoretical, distant, or easy to caricature: rambunctious young new residents and fat-cat developers. Anyone who has been to a community board meeting understands the physical embodiment of this phenomenon: local residents screaming at a developer’s representative for hours with little dissent.

In theory, big developers stand as an interest that can represent those absent consumers of housing. When the dispute concerns a proposed upzoning or other site-specific approval for a specific new project in which a specific developer has invested, then the developer with money in the parcel will fight neighborhood groups, sometimes winning and sometimes losing. By contrast, the developer’s incentive to fight downzonings is missing when the developer has not yet made any site-specific investment in a new project. Because zoning decisions are made seriatim, developers have little incentive to get involved in opposing downzonings where they have no skin in the game yet, splitting the pro-development coalition between developers and renters/​buyers. Opposition to downzonings differs from support for upzonings, therefore, in that the former has only the support of a theory — the idea that excessively restricting supply in the face of strong demand is costly. Naked theories, unadorned by powerful groups with individually valuable interests, fare poorly in the rough and tumble of urban politics.

Does this imbalance lead to an inefficiently low level of housing? In theory, developers could simply bribe the neighbors into accept greater housing density in their neighborhood whenever the time is ripe to build in the downzoned neighborhood. “Community benefits agreements” under which developers promise various public amenities — jobs in the proposed development, playgrounds, affordable housing, etc. — are a practical way in which developers can insure that local opposition does not thwart cost-justified residential development. But a single developer has to overcome the high costs of dealing with the city bureaucracy and cannot internalize all of the benefits of residential density — for instance, reduced traffic and accompanying air pollution from shorter commuting times, greater concentration of consumer demand for richer retail opportunities, or greater “agglomeration economies” from more opportunity for a more diverse pool of residents to interact with each other.

These costs to lobbying for a map amendment, or even a variance, are often worth it for big players, but make life tough on small developers seeking to build a pop-up or rent a “granny flat,” distorting development toward mammoth projects. And large developments are not a perfect substitute for natural housing growth.

If developers cannot lobby or bribe their way into development rights, then it is unlikely that their potential tenants or homebuyers will do so. Dispersed interests such as potential entrants into a neighborhood have only a small interest in preserving housing options where no specific housing is actually being proposed. This leaves the field to neighborhood activists who want to preserve the status quo.

The regional effects of zoning restrictions on new housing can be quite large. Ed Glaeser, Joe Gyorko, and Raven Saks have estimated that, for instance, zoning restrictions are responsible for almost half of the cost of homes in the San Francisco region. For a city like New York that itself covers a huge housing market, the cost effects can be dramatic: zoning restrictions increase the cost of housing in Manhattan by nearly 50 percent. Thus, although it is difficult to get an exact measure of the net benefits of zoning restriction for cost-benefit analysis purposes, there is a great deal of evidence that neighborhood dominance of land use produces substantial costs.

Local legislative blues | The local legislature could, in theory, take into account the city’s overall interest in more housing and zone accordingly. In practice, however, each individual legislator is likely to ignore citywide interests and instead focus on the district-specific interests of his constituents. The reason is rooted in the structure of local politics: cities typically lack the mass political parties that can champion abstract, jurisdiction-wide benefits.

Why are parties needed to advance general policy goals? The problem is that legislatures need some mechanism to overcome what social choice theorists call the problem of “cycling” or strategic coordination problems. As famously diagnosed by Kenneth Arrow, legislatures cannot produce stable policy outcomes if each member has an unlimited ability to propose and amend new pieces of legislation. An unorganized legislature can feature “cycling,” a situation where Proposal A beats Proposal B, Proposal B beats Proposal C, and yet — violating the rational ideal of preference transitivity — Proposal C beats Proposal A. In this situation, without some legislative leader to cabin the agenda-setting process, the legislature will lack the capacity to get anything done at all.

Strong party leaders can use selective incentives, like plum committee positions or pork, to “whip” wavering members of their caucus into voting for legislation that serves the collective ends of the partisan majority. Members of a majority caucus are willing to cede power to leaders because they are confident that leaders like being leaders — it is far more fun to be house speaker than a lowly backbencher — and will therefore order votes in a way that maximizes the political benefits to the party caucus as a whole through the development of a (hopefully) popular party brand.

The power exercised by party bosses might seem undemocratic — a relic of cigar-chomping insiders making decisions behind closed doors in smoky rooms — but delegation of power by rank-and-file legislators to party leaders is what makes popular democratic control over policy possible. Individual voters have no incentives to monitor the behavior of dozens of local legislators closely and little ability to determine whether the thousands of votes taken by each member between elections represented constituents’ long-term or general interests. The members’ delegation of agenda control to a party leadership gives voters a chance to express their preferences about the overall performance of one highly visible coalition marked by a distinctive “brand label” — that is, the party. Monitoring a party retrospectively based on the overall state of a city is much easier than monitoring dozens of obscure legislators voting on hundreds of obscure bills. Thus does party control of legislatures (and party names appearing on the ballot) make it possible for a poorly-informed electorate to use abstract policy preferences — for instance, a preference for “less regulation” as a general matter as a basis for deciding how to vote for low-visibility candidates in low-visibility local elections.

In urban legislatures, however, strong party leaders are often missing, destroying the capacity of the legislature to set its own agenda according to some overall vision of the jurisdiction’s general interest. There are many reasons for this absence of partisan control. A majority of urban elections, for instance, are nonpartisan. In nonpartisan elections, voters are denied the one piece of useful information about candidates — that is, their party “brand”— that could give them a clue about the candidate’s general ideological bent. Absent information about policy stances through information about parties, voters use whatever information they can get their hands on: usually ethnic, racial, and status variables. In such an environment, the candidate’s incumbency status, the personal networks of candidates, and political interest groups take on outsized influence. The absence of party competition enhances the degree to which urban political conflict is formless and thus provides little check on the power of concentrated groups in politics.

In a Hobbesian legislature where each member stands and falls by herself, it is hardly surprising that each member focuses on her own backyard, even if this focus is bad for the city as a whole.

Even cities with partisan elections provide only weak signals of candidates’ general ideologies compared to state legislative and congressional elections. The problem, as one of us has elsewhere argued, is partisan “mismatch.” While our major national political parties now consist of relatively ideologically coherent coalitions, there is little correlation between national issues (e.g., abortion, national defense, the deficit) and beliefs about local issues like zoning. However, election laws ensure that national parties appear on local ballots without much change in the shape of their coalition and, because they lack much other information about local candidates, voters rely on their national party preference when making local voting decisions. City voters rely heavily on national partisan labels — Democratic and Republican — and, as a result, they end up selecting candidates who are representative of local voters on national issues but not on local ones.

The result is that partisan big-city local elections, outside of some mayoral elections and a few cities in which national party preferences are relatively equal, often involve only a single effective party. If everyone in the big city is a Democrat, then the only election that matters is the Democratic primary. However, a primary is just like a nonpartisan election: there are no party labels on the ballot to aid voters in determining the candidates’ general ideologies. Voters turn out in small numbers and are forced to use non-ideological variables in making their decisions. The result is that partisan city elections are often not much more ideologically competitive than nonpartisan ones.

Why should the essentially nonpartisan character of local politics matter for land-use regulation? The reason is that local legislatures, deprived of the capacity to control their agendas and their electoral fortunes by delegating agenda control and electoral “branding” to party leadership, rely on alternative mechanisms for controlling legislation. One popular mechanism for managing the legislative agenda is the universal log roll. In the absence of some strong party leader, legislators face a prisoner’s dilemma: they may prefer that some new development be sited in their neighborhood over their own constituents’ objections that every neighborhood be permitted to exclude all new construction. However, no legislator will vote to allow a new development in her district unless she can be sure everybody will reciprocate by taking their fair share of the housing needed to meet demand. Without political parties to fix a contract to divide up the costs of these decisions fairly and promulgate the benefits of the overall bargain, this prisoner’s dilemma leads each legislator to block every project, even though allocating a “fair share” of the undesired uses to each legislator might be a superior result. Moreover, each legislator votes to support his fellow members’ efforts to exclude LULUs out of fear that, were he to press for unwanted housing in someone else’s backyard, his district would be targeted for unwanted uses.

In sum, the absence of party leaders means that individual legislators cannot take credit for the overall benefits of housing nor fairly apportion the electoral blame of individual votes to allow more housing into specific districts. In a Hobbesian legislature where each member stands and falls by herself, it is hardly surprising that each member focuses on her own backyard, even if this focus is bad for the city as a whole.

How Unconventional Legislative Procedure Can Aid Dispersed Interests

We suggest that a change in voting procedure could mitigate the power of neighborhood groups and induce better land-use decisionmaking. By bundling together decisions about the use of land in different neighborhoods for a single legislative vote, voting procedures could improve land-use decisionmaking in two different ways. First, such issue-bundling could reassure members of the local legislature that they will not be left out of a coalition that seeks to avoid any share of locally undesirable land uses. Second, by requiring that every downzoning be paired with an equivalent upzoning, voting rules could pit neighborhoods against each other and thereby mitigate the power of geographically concentrated groups. To illustrate the power of such issue-bundling procedures, it is useful to examine them in a context outside land-use law where they have built up a longer track record. Accordingly, we examine Congress’s use of extra-legislative issue-bundling in two contexts where issues cut across party lines, so that parties could not successfully organize votes.

Base closings | It is not hard to see why it is politically difficult to close military bases. Domestic military bases bring enormous amounts of money to specific towns and states, and are therefore jealously guarded by the people elected to represent those places. In contrast, national defense is the classic public good; it benefits all citizens in a non-rival, non-excludable manner. When a current base structure ceases to serve the military’s interest, this harm will be felt relatively equally by all Americans. Similarly, the harms created by wasteful government spending — marginally higher taxes, higher borrowing costs — are also felt generally. The diffuse character of inefficient bases’ harm makes these costs less salient to constituents. This does not mean that voters and their representatives do not care at all about getting rid of obsolete bases; it means only that they will protect their own bases unless they have the clearest possible assurance that everyone else is adhering impartially to the criterion of military efficiency for base-closing.

Thus, one can understand base closing as a strategic problem inside Congress, approximating a prisoner’s dilemma. If military bases are retained for non-military (i.e., “pork”-based) reasons, then each member will prefer to have a base in his district. Each member, however, also prefers closing all inefficient bases (including his district’s) to allowing all inefficient bases to remain open. Absent cooperation, each member will protect his district’s base — or “defect” in game-theoretic lingo — and no bases will be closed. Only if they can somehow agree not to protect their own bases and make that agreement stick will lawmakers get to their preferred result of all bases being in the proper place for military purposes.

Their mechanism for such a commitment is the Base Closure and Realignment Act (BCRA).The story of the BCRA begins with the collapse in the 1970s of the purely presidential system of base closing after members of Congress became suspicious that President Richard Nixon was playing political favorites in selecting bases to close. In response to distrust for the untrustworthy presidential agent, Congress refused to approve presidential suggestions about base closures. But lawmakers couldn’t reach an agreement on their own because of the prisoner’s dilemma described above and because the issue cut across party coalitions. And so no bases closed. This inaction generated ridiculous results, like the continued operation in the 1980s of a base created to protect stagecoaches against Native American attacks. Congress needed a procedure to save itself from its own parochial tendencies.

The BCRA provided that answer by essentially delegating to an independent commission the task of determining which bases to close based on an initial list formulated by the armed services and reviewed by the secretary of defense. The commission’s final recommendation was forwarded to the president for approval, subject only to a joint resolution of disapproval by Congress to be enacted within 45 days of the president’s approval of the final list. The joint resolution, however, required an up-or-down vote on all base closures, barring all amendments of the list. In effect, Congress was faced collectively with the choice of either ignoring entirely the general need to close bases or accepting the package of base closures bundled together by the executive actors defined in the BCRA for militarily sound reasons.

By bundling together decisions about the use of land in different neighborhoods for a single legislative vote, voting procedure could improve land-use decisionmaking.

This procedure was strikingly successful in eliminating bases. The success is directly tied to two elements of the process: First, the process forces Congress to make a single vote on an entire list of base closings. This stops individual members from dissolving the deal by proposing an alternative list appealing to a different coalition within Congress — a process that eventually leads to a universal log-roll and no base closings. Second, it gave an outside group, the commission, the power to define the set of base closings. As long as the commission’s decisions are credible as an expression of military need, it can solve the coordination problem faced by Congress of spreading the costs of base-closings over the entire body in a manner acceptable to a critical mass of Congress. To insure that the commission is a trustworthy agent, Congress built into the BCRA a number of safeguards such as requiring specific substantive criteria for base-closing and Senate confirmation of the commission as well as presidential consultation with all four House and Senate party leaders before making nominations in order to ensure partisan balance.

The structure of the decisionmaking process thus allows Congress to pass laws that serve a general interest in national defense against opposition from geographically concentrated groups. It is attractive enough that they have successfully completed five rounds of base closings.

Trade balancing | International trade is considered the classic concentrated-cost/dispersed-benefit policy area. It is likely the most widely agreed-upon belief in modern economics that tariffs and other trade barriers are economically harmful. However, the benefits of removing tariffs fall relatively evenly on all consumers of imports, benefiting everyone diffusely but no one immensely. In contrast, there are industries that are severely harmed when tariffs are removed, specifically the firms and workers in industries that compete with imported goods. While the costs are smaller in welfare terms than the benefits, this distribution of costs and benefits creates a political problem. Those who are harmed have an incentive to spend resources to fight reductions in tariffs or other trade barriers, while those who benefit do not. The political economy of trade, the story goes, is as slanted toward protectionism as the economic benefits are in favor of free trade.

The story, though, has a problem. The United States has approved dozens of trade deals since the last major outburst of protectionism in the 1930s and has historically low tariffs and non-tariff trade barriers. With a few notable exceptions, U.S. policy can be described as pro–free trade. If the political economy of trade is so weighted in favor of protectionism, why are protectionist U.S. policies the exception rather than the rule? One answer to the question lies in the legislative procedure that governs trade deals. Two procedural rules have helped determine the shape and speed of American trade liberalization: reciprocity and fast track.

In the 40 years leading up the Great Depression, Congress regularly increased tariffs. Despite its infamy, the Smoot-Hawley Tariff of 1930 was merely the last in a generation of protectionist measures enacted since the McKinley Tariff of 1890. What changed in the 1930s to liberalize trade policy permanently? The answer lies in changes in legislative procedure. When power swung to the more pro–free trade Democrats in 1932, they passed the Reciprocal Trade Agreement Act (RTAA), which gave the president the power to enter into agreements with foreign countries to reduce tariffs unilaterally. The Trade Expansion Act of 1962 further enhanced presidential power to strike deals, leading to the “Kennedy Round” of tariff reductions, which slashed tariffs to the point where they were no longer an important restriction on trade. Later trade acts added “fast track,” a requirement that Congress vote up-or-down without amendments on trade deals.

One reason why these statutes were so successful in reducing tariffs is that they empowered the president — who, with his national constituency, is generally considered more pro–free trade than Congress. But why would a protectionist Congress agree to empower the president? As Michael Gilligan argues in his excellent book Empowering Exporters, the answer lies in the requirement of reciprocity. Exporters desperately wanted to see reduced tariffs abroad. The RTAA gave the president a tool to achieve such reductions through reciprocal treaties. Such treaties, however, tied the fates of exporters to the fates of import consumers, ensuring there were concentrated interests in favor of trade deals to combat the concentrated interest groups opposed to them. By controlling the order and shape of trade votes, the RTAA moved Congress from regularly anti–free trade majorities to regularly pro–free trade majorities.

Fast track further cemented this process. By requiring that there be only one vote, on a package designed by the president in negotiation with other countries that could not be amended by Congress, fast track ensured that exporters would lobby in favor of trade deals as a whole. Had amendments been allowed, opponents of free trade could pick apart a coalition of exporters and importers favoring free trade by proposing domestic tariff increases on goods that did not threaten exporters with foreign retaliation. The president, by contrast, will negotiate for deals that will bring on enough exporter support to offset increased import-competing industry opposition.

Can Land Use Benefit From Extra Legislative Procedure?

The stories of base closings and reciprocal trade agreements suggest that extra-legislative procedures can play a critical role in controlling concentrated interests. Are there any lessons here for land-use regulation and its apparent need to control the power of neighborhoods to defeat a dispersed interest in housing? We think that extra-legislative procedures not only could play, but have already played, an important role in solving coordination problems in local legislatures as well as insuring that geographically concentrated interests do not go unchallenged.

“Fair share” requirements | “Fair share” requirements are most familiar from the Mount Laurel litigation in New Jersey. Starting in 1975, the New Jersey Supreme Court construed the state constitution to require that each local government in New Jersey accommodate its fair share of the regional need for affordable housing in its zoning ordinance. To calculate these shares, the Mount Laurel II court held in 1983 that three lower state courts, using experts in urban planning, would devise formulae for apportioning the regional need for affordable housing by calculating each community’s share of its region’s land, employment, substandard housing, and population. With the New Jersey Fair Housing Act of 1985, the New Jersey legislature delegated this apportioning function to a statewide agency, the Council on Affordable Housing (COAH). A 12-person agency appointed by the governor with the advice and consent of the state senate, COAH has detailed statutory guidelines and procedures by which to calculate each community’s share of the regional need. COAH’s certification of a community’s land-use plan and zoning resolution creates a rebuttable presumption that the community satisfies the constitutional requirements of Mount Laurel.

It is not hard to see how COAH’s bundling of different communities’ “fair share” obligations into a single statewide plan solves a coordination problem in the state legislature. The concentration of poverty and substandard housing in a handful of dilapidated and near-bankrupt cities like Camden and Newark has costs for the state in terms of peer effects depressing employment and school performance. In order to avoid those peer effects, New Jersey citizens might prefer that each New Jersey local government, including their own local government, forgo its power to exclude all such housing from its territory over the world in which every local government excludes all such housing. The worst possible world, however, would be that in which one’s own community allowed affordable housing while every other community did not, for in such a case the peer effects of concentrated poverty would remain, but the non-exclusive community would bear any fiscal or social costs of hosting low-income households. Because monitoring of other communities’ zoning behavior is costly, local governments acting individually might adopt a position of total exclusion. The state legislature could attempt to apportion low-income households among communities, but each legislator would be tempted to adopt standards that would exempt his own electoral district from such an obligation as much as possible. If the leadership of the major political parties did not take any clear position on how affordable housing should be apportioned, then instability in the legislature might lead to the legislature’s simply refusing to address the issue of zoning, which seems to have been the approach of the state legislature before the New Jersey Supreme Court forced the legislature to confront the issue.

After Mount Laurel II seemed to threaten each local government with a judicially calculated “fair share” of housing, the state legislature delegated the delegation of “fair share” to COAH, consisting of a mix of developers, housing advocates, and suburban mayors. Why? The obvious answer is that the legislature has greater collective control over COAH than of the courts. Justices serve seven-year terms, subject to state senate confirmation and gubernatorial reappointment, but strong norms favoring re-appointment of justices give the court substantial — albeit not complete — independence from the political branches. Moreover, the court need not contain representatives of suburbs and central cities, nor need it be balanced among the two major political parties. There is a danger, therefore, that the courts might not be a faithful agent of the entire legislature. By contrast, COAH’s members must include both big city and suburban elected officials, as well as equal numbers of Republicans and Democrats. COAH, therefore, poses less of a risk that it will disregard the interests of any region or party within the legislature. Moreover, because COAH owes its very existence to the state legislature, it is more amenable than the state supreme court to anticipating and adopting likely legislative reactions to “fair share” calculations.

Perhaps as a result, COAH’s estimates of regional needs for affordable housing have been persistently criticized by advocates of affordable housing as too low. This criticism, however, hardly indicates that COAH has fallen prey to the parochial forces that it is supposed to supersede. Instead, it seems likely that the median legislator in the New Jersey state legislature believes that the amount of affordable housing necessary to avoid the costs of excessive zoning restrictions is lower than the preferences of housing advocates. In effect, COAH operates as an impartial and transparent agent for the legislature, spreading the costs of affordable housing in ways that avoid a scramble among legislators to form unstable burden-shifting coalitions.

COAH is not unique. New York City has used a roughly analogous method for apportioning LULUs among boroughs to insure that low-income neighborhoods do not have to bear more than their “fair share” of noxious facilities like bus depots. As in New Jersey, the city charter delegates the task of working out criteria for siting to an agency — the city’s planning commission — while the implementation of the criteria is given to various exceutive agencies in charge of different categories of LULUs.

There is one salient difference between COAH and the base-closing commission: the state legislature is not bound to vote up-or-down on COAH’s package of “fair share” decisions. The legislature, however, has not revised COAH’s rules in an ad hoc manner, preferring instead to stay out of the housing fray even when loudly criticizing COAH’s performance. COAH’s calculations seem to be self-enforcing because they provide a salient focal point on which legislators can avoid eternal cycling. No one wants to disturb the norm of having some number of Mount Laurel units thrust upon one’s suburb by the agency for fear that the alternative would be the instability of ever-shifting regional coalitions of legislators trying to shift the burden of affordable housing on to their neighbors.

Maintaining a “zoning budget” | COAH’s packaging of multiple upzoning decisions into a single bundle solves a collective action problem faced by a state legislature that lacks strong partisan leadership on the question of siting affordable housing. However, a different approach is needed for increasing housing generally, as opponents of new housing are not determined exclusively be geography but also by status (home ownership). We offer a second sort of extra-legislative procedure as a way of not merely solving coordination problems in the legislature, but also inducing conflict among well-organized interest groups. We call this proposal “the zoning budget balance bill.”

As with other extra-legislative procedures, our proposal begins with a delegation of agenda-setting power to some administrative agency — most plausibly, the city’s planning commission. The planning commission would be charged with setting an overall annual zoning “budget” for the city consisting of the optimal increase (or theoretically, a decrease) in the number of housing units within the city or the number of potential units permitted by the zoning envelope. The planning commission would also be charged with devising a ratio of upzonings to downzonings in light of its zoning “budget.” So long as the city’s housing stock (or zoning envelope) falls below this housing target, the planning commission would prohibit map amendments reducing the number of housing units unless those amendments are matched at some set ratio by an upzoning elsewhere. If the budget is positive, this ratio would have to be above 1, so that each approved downzoning would result in more housing. The ratio would apply until sufficient new projects have been approved to meet the number called for in the budget. After that point, all downzonings or denials would be offset at a ratio of one to one. This package would be voted on, up-or-down, by the local legislature, subject to a closed rule, just like “fast track” for trade deals.

Once the budget is passed, the law would establish a procedure for considering zoning changes. For each proposed map amendment rezoning city land, the agency would present a package of down- and upzonings to the local legislature and the legislature would be required to accept or reject the entire package without uncoupling the two decisions. (The package would include as many upzonings relative to downzonings as are called for in the ratio adopted in the budget.) Supporters of the downzoning — typically, neighborhood activists seeking to preserve community character — would be forced to identify and defend a potential and suitable upzoning in some other neighborhood, thereby becoming advocates for new building. Likewise, the residents of that alternative neighborhood would have an incentive to lobby for maintaining housing in the area of a proposed downzoning by pointing to the relative unsuitability of their own area for more housing. In this way, the downzoning/​upzoning ratio would have the effect of enlisting interests that normally fight against housing proposals to lobby for housing in competing neighborhoods. The idea would be roughly analogous to “pay-as-you-go” budget rules designed to force advocates of new spending programs to identify cuts in other spending programs that would eliminate the budgetary effects of the new proposed spending. By pitting interests against each other, such rules would insure that the relative merits of the interests’ rival proposals are brought to the legislature’s attention.

By pitting different neighborhoods’ activists against each other, such rules would insure that the relative merits of the interests’ rival proposals are brought to the legislature’s attention.

Such a proposal might have some of the same conflict-inducing features as a “fast track” procedure allowing the president to bundle together domestic and foreign tariff reductions. Just as exporters become advocates for reducing domestic tariffs not because they care about consumers, but merely because they want to obtain access to foreign markets, so too the residents of neighborhoods proposed for upzoning would lobby to preserve housing opportunities elsewhere, while the residents seeking a downzoning would lobby for expanded housing in the former residents’ neighborhoods.

This procedure could work to create a permanent coalition in favor of an increased housing stock. Developers would be given new allies, specifically the communities most interested in downzonings, who would lobby together and hence create a powerful coalition in favor of new projects. It would also ensure competition among downzoning projects, with council members’ willingness to fight for their projects serving as a proxy (admittedly imperfect) for the true value of a downzoning. Further, as the matching procedure would provide a political outlet, the system as a whole would be at least modestly secure against truly unpopular denials of downzonings.

The analogy to “pay-as-you-go” budgeting, however, suggests an immediate difference between reciprocal trade agreements and proposals to link down- and upzonings. Like such budgeting, the proposed zoning procedure linking upzoning and downzoning might be easily waived by the very legislature that it is supposed to constrain. There is a causal and temporal relationship between foreign and domestic tariff reductions: exporters cannot get the benefit of the former without the latter, regardless of how Congress votes, because foreign governments will not give the United States something for nothing. It would, therefore, be difficult for exporters to sever the foreign from the domestic tariff reductions and lobby just to obtain the former, ignoring the latter. By contrast, developers, but for the procedural rule, could practically obtain an upzoning without lobbying for a downzoning elsewhere and, but for the proposed bundling rule, neighbors seeking a downzoning would not need to obtain an upzoning. The temptation, therefore, would be great for both developers and neighbors to lobby the local legislature to waive the procedural rule whenever it threatens to derail their zoning agendas. If the rule were routinely waived, then it would not induce neighbors to lobby for housing; they would, instead, lobby for waiver of the rule.

The challenge of the offset mechanism in our “zoning budget,” therefore, parallels the challenge of “pay-as-you-go” budgeting procedures: there needs to be a rule, norm, or institutional mechanism for entrenching the bundling of up- and downzonings. In the following section, we fill in the details of the “zoning budget” system, explaining why we believe that this challenge is surmountable.

In crafting a detailed proposal for a zoning budget, we are guided by the example of both the base-closing commissions and presidential treaty negotiation. In both cases, the institution’s transparency and credibility as an honest broker for legislative interests can induce the legislature to refrain from unbundling the institution’s packages of issues.

The ordinary zoning process in place in most American cities offers the ingredients for such an extra-legislative bundling process. The planning staff and planning commission certainly have the tools to cultivate a reputation for bureaucratic impartiality. The members of planning commissions typically are appointed by mayors sensitive to the various real estate constituencies in the city. Further, their decisionmaking process is transparent. Planning commissions typically must hold hearings before approving any proposals for zoning map amendments, and these hearings are generally well-attended when they affect the neighborhoods of persons having an equity interest in their homes or businesses. There are thus fire alarms built into the zoning amendment process to apprise legislators of substantial community opposition to any proposal. When the city’s planning staff initiates a zoning map amendment (usually at the behest of developers or neighborhood groups), the initial proposal is crafted with an eye to surviving this gauntlet. Like the president negotiating a package of tariff reductions, the planning staff and commission craft their package of zoning proposals with an eye to how it will be received by the legislature that must ultimately approve it.

Packages of upzonings and downzonings should have the same character: the planning staff should be expected to seek out land for compensatory residential upzoning that mobilizes the strongest interests in defense of the entire package. This would mean land that contains the weakest neighborhood groups but the strongest development interests. One would expect such land to be land with low residential densities but high residential value — say, old warehouses in an area that is gradually becoming “hip,” held by speculators capable of, and interested in, developing the land to capture this difference in current and future value.

The “zoning budget” process can magnify and direct these inherent incentives to organize effective coalitions in defense of upzoning. We suggest, as an example of such a mechanism, that the planning staff should be required to submit a “housing impact statement” with any downzoning/​upzoning proposal, identifying not only the loss of housing resulting from any proposed downzoning, but also the quantity of housing likely to be produced by the proposed upzoning. Such a statement would identify the developer(s) likely to propose this new housing, with the expectation that such a developer would submit plats, site plans, or other documentation indicating a readiness to develop the land in question. The planning staff would also identify the reasons why the proposed upzoning moved the land to a superior use and why failure to upzone the land, when coupled with a downzoning of another neighborhood, would endanger the local government’s overall housing goals.

It is likely that the danger that the local legislature’s uncoupling of the planning staff’s bundle of upzoning and downzoning proposals would emerge only if residents from the two affected neighborhoods made common cause with each other to resist new development in either neighborhood. The housing impact statement, however, would highlight the systemic effect of such resistance and force the local legislature explicitly to acknowledge abandonment of its own housing goals. Moreover, the housing impact statement’s assessment that the proposed upzoning site was unsuitably zoned would be ammunition in the hands of the developer seeking to challenge that zoning as a violation of substantive due process or as inconsistent with the city’s own comprehensive plan. Such lawsuits seldom succeed before state courts, in part because the courts are reluctant to second-guess local officials’ own estimates of the proper timing for implementing the comprehensive plan. The housing impact statement’s specific recommendation that the land be developed simultaneously with the downzoning of other land would neutralize such grounds for deference.

The threat of judicial review would not, by itself, suffice to lock in the planning staff’s and commission’s coupling of the down- and upzoning. Instead, the detailed factual findings in the housing impact statement, the ratification of the zoning-budget procedure by the local legislature itself, the lobbying by affected developers, and even the developer’s threat of a lawsuit to challenge downzoning inconsistent with the zoning budget would all enable the local legislature to characterize the bundling procedure as preexisting legal standards the waiver of which would constitute a breach of “rule of law” values. The stigma of acting lawlessly would provide political cover for local legislators to refrain from picking apart the bundle of up- and downzoning proposals. To the extent that this coupling procedure becomes viewed as a given exogenous to the political decision, the residents of the affected neighborhoods will have no choice but to oppose each other’s proposals rather than join forces to urge the uncoupling of the planning commission’s bundle of proposals. Each will argue that the other neighborhood is a more appropriate site for housing given the infrastructure, market demand, presence of industry or other nuisances, etc., and thereby providing information to the local legislature not only about their relative preferences but also their relative levels of political organization — a critical factor for electorally minded politicians seeking information about the risks of alternative policies.

In sum, the planning commission and staff could perform the same role as the president in tariff negotiations, bundling proposals to assemble maximum political support. The powers here urged for the planning commission are not remote from the sorts of powers that such commissions typically exercise. Environmental impact statements can and do contain assessments of effects on housing. Extending this requirement to include the effects of downzonings as well as upzonings is hardly revolutionary. Likewise, such statements typically require some reference to mitigation of harmful effects, including off-site mitigation. Including upzoning proposals as part of such mitigation is not radically different from including recommendations of street widenings or traffic lights to mitigate, say, traffic effects. Further, the idea that zoning ought to be consistent with a comprehensive plan reflecting the city’s overall development goals is hardly novel, being embodied not only in the text of the earliest zoning legislation, but also the exhortations of generations of land-use scholars. The zoning budget’s overall housing goal is merely a specific instance of using citywide goals to discipline piecemeal neighborhood-by-neighborhood decisions.

Unlike traditional supporters of citywide planning (and the procedures they backed), we hold no illusions that planners are omniscient in their prescriptions for city development. The point of our proposal is to realign interest group incentives rather than bring planning expertise to bear. We urge a greater role for executive agencies in agenda-setting simply because such bundles show promise of inducing interest group conflict that might break the impasse over enlarging a city’s zoning envelope. Similar mechanisms have liberalized trade and shut down obsolete military bases. It might be time to experiment with extra-legislative procedures in zoning politics.