The Yale Law Journal

June 2024

The Subdivided City

abstract. Subunits within a city provide different levels of public services to discrete areas. In theory, these subunits, which typically combine public and private characteristics, enhance preference satisfaction and opportunities for democratic participation in city governance. But they also raise issues concerning their ability to serve the interests of their constituents and of the city that hosts them. This Feature addresses the extent to which those who manage subunits are likely to exhibit fidelity to their constituents and the city. Investigation of managers’ incentives leads to a distinction among subunits, largely based on their function. Private-based subunits that serve purely economic interests are likely to be faithful to their constituents but demonstrate less fidelity to the city. Subunits that serve broad community needs are likely to demonstrate less fidelity to their constituents’ service needs but more fidelity to the city; however, conflict with the city is more likely where these community-based subunits seek to exercise political authority. Since community-based subunits seek to raise services for their constituents to the level enjoyed elsewhere in the city, any conflict with the city implicates its obligation to provide all residents equal services on a nondiscriminatory basis. The ability of subunits to help the city perform that obligation entails a need for the city’s deference to community demands for those services that the city fails to provide.

author. Max E. Greenberg Professor of Contract Law, NYU School of Law. Thanks to participants in a faculty workshop at NYU School of Law and especially to Sheila Foster, Ed Rock, and David Shleicher for helpful suggestions. Callen DiGiovanni provided excellent research assistance.


Conversations about “the city” imply that governance of dense urban areas comprises a single monolithic administration that provides local public goods and regulatory functions within defined boundaries. From this view, cities offer their residents a basket of services that define a distinct local character and enhance common local government objectives, including sorting, self-government, and community development.1 Sizeable contemporary cities, however, rarely involve a single entity that performs local government functions uniformly across an entire jurisdiction. Instead, cities authorize multiple subunits within different parts of a municipality to provide services or regulations that deviate from the default level the city presumptively offers to all residents.2

The governance and operation of these subunits frequently involves the participation of private individuals and firms or community organizations, even though city officials may dictate the terms of their creation, define the scope of their authority, and share in their funding and management. These subunits, therefore, differ from purely governmental subdivisions that serve as administrative arms of the city, such as bureaus, departments, or local authorities. Similarly, they differ from fully private entities that provide services for a select group of residents or commercial tenants, such as homeowners’ associations or shopping malls. These public-private subunits include business improvement districts, neighborhood improvement districts, tax increment financing districts, community benefit agreement panels, participatory budgeting delegates, and community land trusts.

Combining public and private characteristics to provide public services within a subunit potentially generates significant benefits in the provision of public goods, participation, and regulation. In this Feature, however, I focus on some risks that accompany the resulting subunits. For one, including both public and private participants in the provision of local public goods may cause those who determine and implement subunit policies (whom I refer to as “managers”) to deviate from the interests of their constituents. Conversely, efforts to serve the interests of those constituents may cause subunit managers to deviate from the interests of the city. I refer to the resulting conflicts as raising issues of fidelity—fidelity of subunit managers to their immediate constituents and fidelity of those managers to the broader city. The central claim of this Feature is that subunits that primarily conjoin the city with entities that pursue private economic interests—which I refer to as “private-based subunits”—are more likely to exhibit fidelity to their constituents than subunits that conjoin the city with organizations that pursue the broader objectives of their community—which I refer to as “community-based subunits.” On the other hand, private-based subunits are less likely to exhibit fidelity to the objectives of the city than community-based subunits. That conclusion, however, comes with an important proviso when the latter seek additional political authority rather than simply to provide additional services.

There are good reasons for cities to create subunits. By differentiating the quantity and quality of goods and services available to different areas of the city, subunits implicitly recognize that the optimal boundaries for achieving traditional local government objectives rarely correspond to municipal boundaries drawn at distant times and for purposes unrelated to contemporary urban demands. Urban areas imperfectly fit the highly idealized conditions of perfect information, perfect mobility, full choice of residence, and absence of externalities under which local governments would efficiently provide the public goods that their residents prefer.3 There is also little reason to believe that the heterogeneous residents of these cities all prefer the same level of a particular service or have similar access to the level of service they desire. Consequently, subdividing municipal authority may promote the objectives of local government by allocating decision-making to relatively homogeneous groups within the city. Residents or members of a subunit, for example, may desire more frequent trash collection than the city offers and may be willing to pay for it. The city may advance that objective by facilitating the formation of a business improvement district or neighborhood improvement district that provides the additional service for which users within the district then pay a fee.

Similarly, cities do not easily satisfy the conditions for robust participation that local government theoretically should offer.4 Even a moderately sized city will struggle to involve a significant percentage of residents in deliberating and deciding issues of common interest. Subunits facilitate the democratic functions of local government by enhancing opportunities for participation and community self-governance, even if their deliberation involves only a few of the numerous traditional local government functions.5 A community land trust, for example, may permit residents to debate the appropriate use of a vacant lot held by the trust as opposed to proceeding through citywide planning processes. A committee of residents may negotiate an agreement with a developer to minimize the adverse effects of a project and secure offsetting benefits for the community. These relatively small decision-making bodies permit discrete members of an affected community to have more input and decisional authority than would be possible if city officials made the relevant choices. To the extent that community control through subunits involves relatively disadvantaged residents, subunits may even permit a more equitable distribution of public goods that cities fail to achieve due to market forces or political dynamics. For example, a neighborhood improvement district may develop broadband infrastructure and thus increase wireless access for low-income communities that commercial providers consider insufficiently profitable.6

Before embracing the proliferation of subunits, however, it is appropriate to assess how well these entities serve the constituents they represent. Similarly, the fact that subunits exist to pursue their constituents’ desired level of public goods or participation or regulation—a preferred level that differs from what the city is already providing—raises the prospect that subunit policies will adversely affect nonconstituent residents and the city writ large. If managers exhibit infidelity either to their constituents or to the city, then the presence of subunits may frustrate rather than facilitate the city’s optimal achievement of its objectives. For these purposes (and for this Feature generally), the term “constituents” varies with the subunit. Some subunits are membership-based, and those members are the relevant constituents, even if they are nonrepresentative of all persons the subunit affects. Other subunits purport to serve all residents of a distinct community, and those residents qualify as the relevant constituents. Still other subunits are formed to pursue the welfare of a particular subgroup of residents or firms within the subunit (such as garment firms within a district the city has established to support garment manufacturing), in which case those persons or firms are the relevant constituents.

It is plausible that subunit managers could be highly faithful in one relationship and less faithful in another. For example, subunit managers could serve as faithful agents of their constituents but, by serving those interests, be unfaithful to the city. The possibility that subunit objectives could conflict with those of the city raises a third issue that is crucial for determining the appropriate design and authority of subunits. That is, in the event of conflict between a city and its subunits, which entity should prevail?

In this Feature, I contend that analysis of these issues depends critically on each subunit’s function, the incentives that its managers face, and its governance structure. Those characteristics vary among types of subunits. For example, some subunits serve primarily economic interests of their members, and that role may induce members to participate in or monitor subunit governance to ensure satisfaction of those interests. Other subunits serve broader and more diverse objectives that are difficult to verify, which complicates members’ capacity to participate in or monitor subunit governance. The result is that different types of subunits are likely to exhibit different levels of fidelity.

These issues of fidelity and conflict between a centralized entity and its decentralized subdivisions arise in other contexts. Students of federalism will recognize the analogous tension between allowing national residents to sort themselves among states in accordance with preferences and the subsequent dilution of national identity.7 State and local government scholars consider how local autonomy enables interlocal competition and democratic participation but also fosters inconsistency among state policies.8 Institutionalists who analyze corporate structures recognize that firms often organize with different divisions or subsidiaries to achieve efficiencies in production or marketing of their goods.9

Nations, states, and firms organize with subunits to minimize transactions costs and agency costs10 that would exist if all decisions were made at a centralized level.11 Firms, for example, reduce the costs of negotiating, renegotiating, and implementing inherently incomplete contracts with independent entities by vertically integrating functions but retaining a decentralized organizational structure.12 Decisions that are insensitive to variations among customers, suppliers, or employees—such as payroll administration—may be amenable to treatment at a centralized level. But where the entity faces customers with different tastes, suppliers with goods or services of different quality, or employees who develop different products for the firm, decentralized decision-making is superior to a one-size-fits-all model.

Cities face analogous problems in their efforts to provide services and opportunities for self-governance. Similar to firms, cities may optimally perform various functions at centralized levels, such as a school system that purports to deliver the same level of service citywide or zoning decisions that implement a citywide comprehensive plan. But if residents within a discrete area of the city prefer a higher level of services than the city’s default and have difficulty contracting for the increment—either because of the service’s nature as a public good or inequality in resources—then the city’s creation of a subunit may reduce transactions costs, just as firms achieve by decentralizing production. Even where it may be reasonable to discriminate among service levels for dissimilar residents—such as where different neighborhoods face different needs for publicly funded services—cities face significant transactions costs in negotiating and implementing diverse arrangements for the relevant residents. Creation of a subunit may reduce political or financial obstacles by enhancing opportunities for democratic participation or localizing redistribution, just as different divisions of a firm may pursue separate product lines for different groups of customers.13

City subunits may also reduce agency costs that arise under centralized decision-making. For example, centralized political actors may ignore highly localized interests in favor of providing uniform services across the jurisdiction.14 As in the case of intrafirm divisions or affiliates that allow the firm to provide different qualities of goods to customers with different preferences or abilities to pay, city subunits may reduce friction between constituents with atypical service demands and officials attentive to citywide service levels.15

Functionally and organizationally, therefore, public-private subunits have characteristics that provide a healthy addition to the range of options for delivering local public goods and promoting democratic values. At the same time, decentralization may promote inefficiencies. Within firms, subdivisions may pursue objectives that not only differ from but also conflict with those of central
management.16 Similarly, subunits of a city may attempt to implement policies that conflict with those of the city. First, creating a sublocal layer for supplying public goods and opportunities for participation may increase agency costs between those constituents and city officials. It may also increase transactions costs of developing appropriate citywide policies. For example, subunits may promote rather than constrain the NIMBYism17 that often characterizes local opposition to a development that would benefit a city, notwithstanding adverse effects on the developed area.18 A business improvement district may increase safety and reduce crime within its boundaries, but only by driving the same conduct to less protected parts of the city.19 Second, the involvement of private individuals in subunit governance raises issues about those parties’ incentives to pursue public versus private objectives. Compounding this issue, subunit participation costs raise questions about the representativeness of those who purport to act on behalf of affected residents.20 Finally, the very existence of subunits implies fragmented municipal decision-making that can dilute accountability by increasing the costs of monitoring decision makers and subjecting municipal policies to multiple veto points.21 In short, the wrong form of decentralization can increase rather than decrease transactions costs and agency costs.

Most of the literature that addresses subunits deals with only a particular form of the phenomenon—for example, by focusing exclusively on business improvement
districts.22 That work tends to explore issues that could affect multiple subunits, such as the representativeness of managers and the ability to generate adverse external effects. But by focusing only on specific forms of subunits, the existing literature does not propose general theories about the fidelity of subunit governance either to their constituents or to the city.

The relatively small literature that deals with multiple subunits tends to focus on different, albeit related issues. Richard Briffault’s influential work on sublocal units was primarily concerned with privatization’s effect on local government.23 Nadav Shoked’s work on “micro-localism” primarily critiques the claim that sublocal units promote efficiency and democratic participation,24 but it does not distinguish between the incentives of private-based and community-based subunits that motivate my analysis. Sheila R. Foster and Christian Iaione’s distinction between “top-down” and “bottom-up” subunits largely corresponds to the private-based and community-based organizations on which I focus.25 Indeed, the community-based subunits I discuss largely comprise the “public-community partnerships” they endorse.26 Foster and Iaione’s elegant model of how public-community partnerships can develop the common resources of the city, however, does not concentrate on the implications of those categories for the fidelity issues that I address. K. Sabeel Rahman and Jocelyn Simonson’s work on community control addresses issues of representation and service delivery in different forms of community-based subunits, but it does not compare them to private-based ones.27

By focusing on the general phenomenon of subunits and distinguishing among their different types, I build on the foregoing work to provide a more complete picture of the conditions under which decentralized, sublocal governance can facilitate city provision of local public goods and democratic participation or, more problematically, interfere with those same objectives. Further, I argue that subunits that attempt to equalize services across the city should receive more deference than those subunits that attempt to increase services above the city default level.

Part I of this Feature describes a representative selection of subunits. Part II explains crucial similarities and differences among subunits and accordingly classifies these subunits as private based and community based. Part III argues that the alignment of interests, capacity for monitoring, and likelihood of constituent complaint induces greater fidelity to constituent interests in private-based subunits than in community-based subunits. Part IV explains why, compared to community-based subunits, private-based subunits are more likely to be unfaithful to the interests of the city that authorizes them. Finally, Part V proposes institutional and organizational reforms that might increase fidelity for both classes of subunits but argues that a different analysis is appropriate to address conflict between community-based subunits and the city. That Part contends that the equal services doctrine, which requires cities to provide services in a nondiscriminatory manner, justifies granting substantial latitude to community-based subunit policies because those entities may compensate for cities’ failure to comply with their legal obligations. The Feature concludes by summarizing the argument that subunits can enhance service provision and democratic participation, but they require both institutional changes and occasional city deference to play that role effectively.