A Polycentric Vision for Governing the Urban Food Commons

Abstract

This chapter uses polycentric governance as a lens for examining alternative governance visions for urban agriculture, or what I refer to as the urban food commons (UFCs). I draw ideas from the political and institutional economics of Vincent and Elinor Ostrom to discuss why UFCs cannot be governed through a one-size-fits-all institutional arrangement. I interlace my conceptual arguments with empirical examples, including examples from Kaufman and Bailkey’s classic work. Governing UFCs effectively, I argue, requires a mix of institutional arrangements involving centralized, decentralized, competitive market, cooperative, and command-and-control governance models. A polycentric governance vision moves us closer to building such diverse institutional arrangements to manage undesired outcomes, including the capacity to effectively internalize spillover effects from UFCs. I delineate four institutional design parameters (IDPs) to guide the design and evaluation of a polycentric governance vision for UFCs: (1) multiple decision centers (or actors) that can make decisions about UFCs, (2) opportunities for cross-scale and cross-sector interactions among UFC actors, (3) an overarching system of rules to guide decisions and interactions, and (4) adaptability of the rules to align with actors’ incentives and decisions emanating from the rules. I use Chicago’s nonprofit urban land trust NeighborSpace to illustrate how these IDPs manifest empirically.

1 Introduction

In Farming Inside Cities, Kaufman and Bailkey (2000) have discussed the vision and reality of urban agriculture (UA) in the United States. The vision of UA is “where inner-city residents grow food in the soil, in raised planting beds or in greenhouses, then market their produce at farmer’s markets, to local restaurants, or to city and suburban residents eager for fresh, locally-grown food.” In reality, however, this vision stood on a “wobbly three-legged stool” that could be made sturdier through “a more supportive institutional climate” (Kaufman and Bailkey 2000). Since this publication, several studies have discussed government and non-government actors’ governance of community gardens, backyard gardens, or community-supported agriculture (CSA) around the globe (Battersby and Muwowo 2018, Bryld 2003, Mok et al. 2014, Raja and Whittaker 2018, Smit 2018). These studies remind us that achieving an envisioned, supportive institutional climate for UA is a moving target that still requires the attention of food scholars, practitioners, and policymakers. This chapter presents ideas to help us conceptualize and design alternative governance visions for UA in the United States.¹ The UN Organization for Economic Cooperation and Development (cited in Kaufman and Bailkey 2000) defines urban agriculture as “production of food and nonfood plant and tree crops, and animal husbandry, both within and fringing urban areas.” Kaufman and Bailkey (2000) employ this definition to demonstrate the compositional elements of urban agriculture: location (urban, suburban, or peri-urban), activities (crop farming, animal husbandry, horticulture, and/or aquaculture), production stages (from growth to distribution, consumption, and waste production), and purposes (consumption or as input for other production). Direct and indirect benefits result from UA: increased food supply, improved nutritional health of children, employment opportunities within the food-value chain, improved aesthetics in cities, and reduced impervious surfaces in urban areas and their associated benefits, such as reduced stormwater runoff and decreased urban heat-island effect (Hynes and Howe 2002; Maxwell et al. 1998; Mok et al. 2014; Smit and Nasr 1992; Twiss et al. 2003). These derived direct and indirect benefits for everyone explain why some scholars and practitioners consider UA as a “public good” (see Vitiello’s discussion in this book, Vitiello 2024). Considering UA as a public good has a governance implication: how to ensure that those impacted (directly and indirectly) by UA get to pay for and decide how UAs are created and managed. This implication underlies Olson Jr. (1965) “logic of collective action,” which animated and was also animated by long-standing governance debates on how to effectively govern public and other types of goods (see summary of debate in Ostrom 2010).

This chapter is organized as follows: the next section reconciles how UA fits into the long-standing governance debate about different institutional arrangements and models needed to manage various types of goods (public, private, club, and CPRs). I then turn to polycentric governance, which I advance as an alternative institutional arrangement for managing urban farms or community/neighborhood gardens, collectively referred to as urban food commons (UFCs).² Polycentric governance refers to multiple semi-autonomous decision-making centers that have competitive or cooperative interactions within an overarching system of rules (i.e., institutions). After mapping the theoretical contours of polycentric governance, I provide in the next section a framework to analyze and design the polycentric governance of UFCs. I use Chicago’s nonprofit urban land trust NeighborSpace as a proof of concept. I highlight empirical features evidencing Chicago’s polycentric governance of UFCs and how such features can be harnessed to develop a coherent governance vision for UFCs in other U.S. cities. A brief summary concludes the chapter.

2 Governing the Urban Food Commons: Beyond Market, State, Centralized, and Decentralized Models

The term “commons” signifies a natural area that is owned, managed, and valued through a social collective process, or “commoning” (Huron 2015; Nonini 2007). Even though scholars have widely discussed the urban commons (eg. in Armiero 2011, Gioielli 2011, Harvey 2011, Lee and Webster 2006, Newman 2013), Huron (2015) maintains that “…much of the extant work on the urban commons takes the city as a site for the commons without theorizing what may be distinct about a specifically urban commons.”

According to Huron (2015), two distinct features characterize the urban commons: concentration of people, competing uses, and financial investment; and representatives of the collective work of strangers. Drawing on Eizenberg’s (2012) work on community gardens in New York City, Huron argues that people working together collectively built these gardens, but some scholars and policymakers also criticized the gardens for the opportunity costs they presented; that is, community gardens occupied lands that could be used for other purposes. Huron believes that theorizing the distinct features of specific commons should include the misconception that the commons is freely accessible to all. This view criticizes Hardin’s classic piece “Tragedy of the Commons,” which begins with “Picture a pasture open to all” (Hardin 1968) and ends by claiming that “Individuals locked into the logic of the commons are free only to bring on universal ruin.” Some scholars believe that Hardin’s conclusion is based on a flawed premise, “picture a pasture open to all,” because the commons is not open to all but only to a group of users who collectively manage it (Bromley 1991; Ciaracy-Wantrup and Bishop 1975). That is, enclosure of the commons was the norm in medieval Europe and the European colonies (see Cronon 2013; Hoffmann 2014). Lee and Webster (2006) also observed that the current urban enclosures in China, evidenced by gated communities, are no different from the enclosures of rural commons in sixteenth-century Britain or Ebenezer Howard’s (1965) utopian garden city in the twentieth century.

In what follows, I discuss why enclosure is not necessarily an inherent attribute of urban commons. Rather, underlying institutional arrangements often exist to ensure that certain people are included or excluded from benefiting, contributing to, and participating in decisions about the urban commons. I begin this conversation by (re)visiting two overlapping questions: (1) Should UFCs be governed as public or private commons? (2) Should UFCs be governed as a local or nonlocal (regional, state, federal, or global) commons? I refer to the first question as the public-private, state-market debate

Paul Samuelson’s (1954) argument about public-private goods and state-market institutional forms is illustrative. Goods (or in this case commons) are defined by their degree of subtractability and excludability. Commons are considered a public good when they have low excludability (very difficult if not impossible to prevent people from enjoying the good) and low subtractability (a person using the good does not diminish what is available for other people to use). Private goods, such as a bottle of fruit juice sold in a supermarket, have high excludability (only those who pay can use it) and high subtractability (Gulick 1957; Hardin 1978; Latin 1984; Ophuls 1973; Steinzor 1998; Ackerman and Stewart 1987; Demsetz 1983; Stewart 1985; Welch 1983). Ostrom (1990, 2010), James Buchanan (1965) identified an example of commons with high excludability but low subtractability: club goods. Ostrom and Ostrom (1977) also introduced the concept of common-pool resources (CPRs) to exemplify commons that have low excludability but high subtractability.

More important, the first debate, particularly informed by Elinor Ostrom’s observation, demonstrates how UFCs under certain conditions can exhibit characteristics of all four types of goods: public, private, club, and CPR (see Fig. 7.1). For instance, urban farm areas with free access (i.e., low excludability) and sufficient land for everyone (low subtractability) could be considered purely public UFCs. On the other hand, urban farm areas belonging to farmer cooperatives with members-only access (i.e., high excludability) and limited available land (high subtractability) can be considered purely private UFCs. If UFCs are purely private or public commons, then we can employ market pricing or state regulatory mechanisms (e.g., zoning and taxation), respectively, to ensure that anyone who enjoys purely private or public UFCs pays their fair share. Urban farm areas owned by farmer cooperatives (i.e., high excludability) but with sufficient spaces for every cooperative farmer (low excludability) exhibit characteristics of club goods, but urban farm areas with free access (i.e., low excludability) and insufficient space for everyone (high subtractability) exhibit CPR characteristics. Because UFCs can reflect all four types of goods, they present complicated management or governance challenges. For instance, introducing price and/or regulatory (e.g., zoning) mechanisms to manage UFCs can exclude certain individuals (especially vulnerable populations) from benefiting from UFCs. Conversely, adopting a laissez-fair (free-for-all) approach to managing UFCs, thereby relying on people’s altruism to pay for them, can constrain available resources. This second debate further complicates these challenges, especially when we consider the indirect and often non-local effects (benefits and costs) of UFCs .

Fig. 7.1

Conditions under which UFCs reflect all types of goods. The conditions listed are not exhaustive. In reality, various conditions in specific local contexts can make UFCs reflect any or all four types of goods

The second debate, regarding institutional scale, centers primarily on the choice of small governments (decentralized local governing entities) versus large governments (centralized governing entities). In the U.S., some reformers claim that having several municipal governments and special-purpose jurisdictions (e.g., school districts) leads to inefficiencies. (Lind 1997) referred to these decentralized governments as a “horde of Lilliputian governments,” portraying metropolitan governance in the U.S. as “organized chaos” or a “crazy-quilt pattern” (see also Ostrom et al. 1961). Here, I draw ideas from environmental federalism and the decentralization theorem (Esty 1996; Oates 1972, 2001), especially the three benchmark models (local decentralized, large centralized, and cooperation) of Oates (2001) to discuss why the choice between small decentralized scale and large centralized scale governance models matter for UFCs. Figure 7.2 helps us understand how these benchmark models align with how we think about UFCs. As noted, large centralized governance or market models can better provide UFCs as purely public or purely private goods, respectively. Thinking of UFCs as local public goods (with no spillover effects) allows one to argue, based on the principle of subsidiarity, that the least centralized decision-making authority can, in theory, better provide UFCs as purely local public goods. However, the spillover effects of public goods require a mix of state-market and local-nonlocal institutional arrangements or governance models to better manage these effects.

Fig. 7.2

Thinking through institutional arrangements to govern UFCs based on their characteristics and scale of institutions. A mixed institutional arrangement often accurately portrays the varied and changing urban conditions that make UFCs reflect any or all four types of goods

Attending to the mix of institutional arrangements needed to manage UFCs is important for both theoretical and empirical reasons. In theory, and as discussed above in the public-private goods and state-market institutional debate, there is more fluidity in the classification of goods: the prevailing conditions in urban areas (see Fig. 7.1) make it generally impossible to classify UFCs as purely public, private, club, or CPRs. Thus, Oates (2001) concludes that due to changing local conditions and contexts, none of the institutional arrangements is generally preferable and that “Prescribing one alternative to the exclusion of the others is, in my view, unwise.” In empirical terms, the governance of American societies resembles more of a mix of institutional arrangements (Ostrom 1972a, 2007; Ostrom et al. 1961). In reality, well-functioning and sustainable community gardens or urban farms require government, non-government, local, and nonlocal support. Kaufman and Bailkey (2000) highlighted the weakness of Chicago’s community gardens because of weak non-governmental support, compared to those of Boston and Philadelphia. Martinez (2016) documents current and specific federal, state, and local government policies that address structural barriers in local farming and food issues in the U.S. Pothukuchi and Kaufman (1999) specifically examine the role of city institutions (planning and food departments and food policy councils) for supporting local food issues (see also Raja and Whittaker 2018; Vitiello et al. 2015) . Thus, in reality, a mix of institutional arrangements inadvertently affect UFCs in U.S. cities. The challenge, as recognized by Ostrom (2005) and others, is to design the right mix of institutional arrangements that can affect UFCs positively: the intentional design of diverse institutional arrangements that are flexible or adaptive in different conditions to ensure that the costs and benefits of UFCs are properly and equitably internalized. I turn to the theory of polycentricity for answers on how such an institutional mix could be designed for UFCs.

3 Polycentric UFCs: Governing UFCs Through an Institutional Mix

Polycentricity is variously defined, discussed, and debated among scholars and sometimes across disciplines. This section will not recapitulate the various disciplinary histories and contestations of this concept (see detailed discussions in Aligica and Tarko 2012, Batty 2001, Frimpong Boamah 2018b, Green 2007, Hall and Pain 2006). In this essay, polycentricity, or what Vincent Ostrom refers to as the “compound republic,” represents a constellation of actors functioning together as a coherent governance system through interactions structured by an overarching mix of rules:

“Polycentric” connotes many centers of decision-making which are formally independent of each other…To the extent that they take each other into account in competitive relationships, enter into various contractual and cooperative undertakings or have re-course to central mechanisms to resolve conflicts, the various political jurisdictions in a metropolitan area may function in a coherent manner with consistent and predictable patterns of interacting behavior. To the extent that this is so, they may be said to function as a “system.” (Ostrom et al. 1961)

For our context of UFCs, we can consider polycentricity as the interactions among multiple actors (e.g., farmers, local government agencies and nonprofits) within diverse institutions (rules) working to ensure that under different conditions, anyone benefiting from a given commons gets to pay their fair share and decide on the commons. In other words, consider the market as made of a single institutional arrangement—the competitive market rule—whereby consumers and producers use price signals to decide when to buy, what to buy, and how much to buy. The buying and selling interactions between consumers and producers within this competitive market rule ensures equilibrium over time, so that anyone who benefits from the good (as consumers or producers) gets to pay for and decide on it. As discussed, operations within this single competitive market rule can reach equilibrium if we assume that consumers and sellers deal with purely private goods. Some scholars (e.g. Ostrom et al. 1993; Stiglitz 1989) have argued that even with purely private goods, a market institutional arrangement cannot function effectively in reality unless it is supported by other institutional arrangements such as the state’s command-and control regulatory powers. This regulation is often needed to discipline the market through activities such as limiting market monopolies, providing conflict-resolution mechanisms (e.g., courts) to resolve private ownership conflicts, and cooperation and negotiation mechanisms to ensure the enforcement of contract agreements, for instance.

Similarly, we can define a state (either centralized or decentralized) institutional arrangement as comprising command-and-control rules that allow the state to regulate the actions and interactions of actors. These rules achieve equilibrium by ensuring uniformity in standards and enforcement—for example, uniform taxation and state-sanctioned permit standards. Command-and-control rules will work if goods are purely public, but even then, the state enacts and enforces its command-and-control rules often through the use of other formal or informal institutional arrangements, including lobbying and backroom negotiations. In other words, when we think of polycentricity, we should consider how (1) multiple actors (e.g., urban farmers, community garden owners, real estate developers, local government agencies, state and federal government agencies, etc.) (2) who are autonomous can nonetheless (3) interact formally and informally with one another through cooperative, competitive, command-and-control, and conflict relationships to ensure (4) polycentric order, “spontaneous order,” or equilibrium within the system (see Aligica and Tarko 2012; Ostrom 1972b, 1991).

I now delineate the two key conceptual pillars of what could constitute a polycentric UFC system. These pillars include (1) multiple decision centers or actors with opportunities for cross-scale interactions for making decisions about UFCs and (2) flexible and adaptive rules based on actors’ incentives and rule outcomes. These pillars draw from global lessons from the governance of commons by Ostrom and colleagues in their design and deployment of the institutional analysis and development (IAD) framework (Ostrom 1990, 1998, 2005, 2010; McGinnis 1999, 2011; McGinnis and Ostrom 2012). Polycentric governance of UFCs should include at least these two features to ensure that the governance system can address tragic outcomes. The pillars discussed here are not exhaustive but are presented to stimulate conversations on the need for alternative institutional arrangements to govern UCFs, beyond conventional approaches such as state, market, local, and nonlocal governance models.

• 1. First conceptual pillar of a polycentric UFC system: Multiple decision centers that allow cross-scale and cross-sector interactions for making decisions about UFCs. At the heart of polycentricity is the need to distribute power to ensure that a single decision-making center (e.g., individual or organization) does not hold absolute power: “…the governance of metropolitan areas can occur in a polycentric political system so long as no single set of decision makers is able to gain dominance over all decision-making structures” (Ostrom 1972b, p. 21).

For effective polycentric governance of UFCs, a single decision-making center cannot dominate decisions about where, what, and how to farm or garden in cities. Decisions about farming in cities, such as location and pricing of urban farm areas, consumers’ access to farm produce, and disposal of farm waste, must address the following questions: (1) Who is making these decisions? (2) Are there opportunities for others to challenge and change these decisions? From an empirical standpoint, for instance, scholars have variously discussed the difficulty for actors such as farmers, gardeners, and food policy advocates to challenge intentional and unintentional (re)zoning practices that affect farming and food issues in U.S. cities (Cohen 2018; Heckler 2012; Raja et al. 2014; Voigt 2011).

Similarly, others should be able to challenge and change decisions made by farmers and/or nonprofits engaged in urban farming or gardening. This is where opportunities for cross-scale and cross-sector interactions and engagements become important so that actors can discuss, challenge, or agree on decisions. Figure 7.3 illustrates a hypothetical group of urban farmers who have local interactions with other farmers, local nonprofits, local real estate developers, and city authorities. These farmers also have cross-scale (local-regional/state/federal) interactions with the State Department of Health and the USDA. We also see cross-sector interactions between farmers (food sector) and real estate developers (housing sector) and between real estate developers (private for-profit sector) and nonprofit 1 (private non-profit sector). I discuss the nature of such interactions in terms of the second conceptual pillar.

Fig. 7.3

Network of actors and interactions structured by different rule relationships. The links between nodes (actors) represent the dominant rules by which they interact. In reality, however, more than one rule can structure the interactions between two actors, even though a specific rule can dominate at a given point in time. Interactions between actors can also change over time by shifting from one rule to another

A polycentric governance system can eventually break down into a hierarchy, or worse, chaos, if multiple decision centers cannot make, challenge, or change decisions (c.f. Aligica and Tarko 2012, Pahl-Wostl and Knieper 2014, Rudel and Meyfroidt 2014) More important, a unique advantage that cities offer to UFCs is density of people and organizations, which facilitates collective action and innovation (see Glaiser 2011; Huron 2015). Individuals’ ability to exercise diverse opinions about UFCs is necessary to (1) ensure innovation in how UFCs are organized in cities and (2) create opportunities to monitor and adjust decisions and outcomes. However, because rules structure the decisions of and interactions among multiple decision centers, the system of rules available within UFCs is also important. I discuss this next.

• 2. Second conceptual pillar of a polycentric UFC system: Flexible and adaptive rules based on actors’ incentives and rule outcomes. Rules are defined as when groups of individuals develop a common understanding of “who must, must not, or may take” certain actions in particular situations; those who fail to conform are subject to sanctions (Crawford and Ostrom 1995; Ostrom et al. 1994). Figure 7.3 shows how a group of urban farmers can have interactions characterized by multiple formal and informal rules. For example, farmers can have a cooperative interaction evidenced by the mutual sharing of farm inputs and information—some form of decentralized cooperation à la Axelrod (1984) and Taylor (1987) or local self-organization à la Ostrom (1990) and Ostrom and Gardner (1993). This cooperative rule interaction can be formal (e.g., bylaws of a farmers’ association) or informal (e.g., cooperation based on friendship or familial ties). Farmers can also have competitive market rule (buyer-seller) interactions with their urban consumer residents or can have conflicting rule interactions with consumer residents through, for example, residents’ not-in-my-back-yard (NIMBY) protests. State and federal agencies (e.g., Department of Agriculture and Environmental Protection) can have cooperative (e.g., sharing information) or conflicting rule interactions that can affect farming within cities. For instance, Institute of Medicine (2003) highlights how both national and state governments’ concurrent exercise of public health powers (see Gade v. National Solid Waste Management Association 1992) leads to contestations over public health infrastructure and decisions across multiple issues, such as nutrition, sanitation, and infectious diseases, and multiple governing scales (see also Gostin 2000; Mozaffarian et al. 2018). Within a polycentric system, a mix of rules structures the set of interactions between actors within and across governing scales: “Within a set of rules, autonomous decision makers are free to pursue their own interests subject to the constraint inherent in those particular rules being enforced. The many autonomous elements or units seek to order their relationships with one another rather than by reference to some external authority” (Ostrom 1994)

In other words, what separates polycentric from monocentric systems is that in the former, actors interact and adjust their actions based on multiple rules, while a monocentric system imposes a single, top-down, command-and-control rule on actors’ interactions and decisions (Pahl-Wostl and Knieper 2014). The most important consideration here is that actors be able to design, redesign, choose, and enforce among a system of rules that align with their incentives and desired outcomes: “If individuals can know the relationship between particular rules and the social consequences that those rules tend to evoke under specifiable conditions, then specific polycentric orders can be created as a matter of conscious design” (Ostrom 1972b).

Ostrom’s point aligns with the logic underlying Buchanan’s constitutional economics (Brennan and Buchanan 1985; Buchanan 2002; Buchanan and Tullock 1962). Constitutional economics, in simple terms, posits that actors do not only make decisions and interact with each other within predetermined rules but can also choose the rules that shape their decisions and interactions (Frimpong Boamah 2018a; Kurrild-Klitgaard 2012). Rules of cooperation, competition, negotiation, and command-and-control have outcomes that differ based on multiple factors, such as the incentives facing actors. Decisions within command-and-control rules achieve different outcomes, positive (e.g., coerce free riders to pay for benefits enjoyed) or negative (e.g., exclude poor and vulnerable actors from participating in decisions), depending on the incentives of those making and enforcing these rules. For example, local governments incentivized to prioritize certain land uses, such as residential and commercial, over others, such as farmlands, in cities are likely to enforce zoning rules favoring the former rather than the latter land use, if we assume all other things are equal. Similarly, just as competitive rules, such as market rules, can lead to positive and negative outcomes, so do cooperative rules: “Competitive rivalry among public enterprises can generate adverse social consequences as well as beneficial effects. Cooperative arrangements among public entrepreneurs can also degenerate into collusive efforts [e.g. monopoly power or patron-client relations] to raid the public treasury” (Ostrom 1972b, p. 20).

Actors’ decisions and interactions in cities are driven by their incentives, which can be (mis)aligned with the formal-informal rules within which they are embedded. Arguing for the use of one rule over others in decisions about UFCs misses the potential benefits that could result from bundling different rules. Ostrom and colleagues’ empirical studies of cases across U.S. cities found that a mix of institutional arrangements are necessary for an effective governance system, whether of UFCs or other urban commons (Ostrom 2010; Ostrom et al. 1978). Effective polycentric systems allow actors to choose and readjust the rules needed to achieve desirable outcomes: “At least conceptually, we know that markets work and have desirable welfare features because of price signals. The polycentric order is supposed to describe the conditions under which desirable emergent outcomes occur absent price signals” (Frimpong Boamah 2018b).

Cooperative rules are sometimes needed to resolve the negative effects of overly competitive rules, and command-and-control rules (e.g., anti-trust law) are also sometimes needed to address the negative effects of overly cooperative rules (e.g., monopoly). This issue invokes large-scale farming corporations such as Hantz farms in Detroit, Michigan, which was criticized for exploiting and deepening racial and income differences and environmental degradation within neighborhoods (Delind 2011; Vitiello and Wolf-Powers 2014). Eventually, the city exercised its command-and-control powers by voting to sell 140 acres of land to Hantz farms and later prohibited the farms from growing food (Vitiello and Wolf-Powers 2014).

Furthermore, arguing for the governance of UFCs through local community organizations (nonprofits or community boards) potentially downplays how the decisions and rules of nonlocal actors impact local decisions over time. For instance, Congress and the USDA defunded the Penn State Urban Gardening Program in 1996, which reminds us of how certain state or federal decisions, such as about funding, can be vital to the success or failure of urban farm projects (Lawson 2005; Reynolds 2011). In fact, Kaufman and Bailkey (2000) commented on this issue when discussing how, across their case studies, state and federal actors were impeding UA:

Any state’s perception of agriculture solely as a rural activity, if not prohibiting the direct marketing of urban-grown food, can affect the regulations governing such sales. Thus, urban growers in Boston find the Massachusetts laws relevant to the selling of their products confusing and open to different interpretations.

Similarly, one cannot also underestimate how local decisions and rules impact centrally governed UFCs. Martinez (2016) reminds us that because farming in the U.S. is primarily a local activity, even state and federally sanctioned food initiatives must deal with local decisions on issues such as zoning, licensing, and market ordinances (see also Broad Leib 2013; Raja and Diao 2016). Kaufman and Bailkey (2000) averred that conflicts among various city agencies, such as the case of the Vacant Property Review Committee in Philadelphia and the idea that urban agriculture may not be the best use of city land, were among the key forces that fettered UA in their case-study cities. Several case studies on the commons in the U.S. and globally suggest that without these two conceptual pillars as the foundation, governance of urban commons such as UFCs could, over time, be dominated by competitive, conflicting, or command-and-control rule interactions, which can lead to tragic outcomes (e.g. Andersson and Ostrom 2008; Ostrom 1990). A polycentric governance vision represents an attempt to prevent or remedy such outcomes, by calling for a governance system that (1) recognizes the role of multiple actors in decisions about UFCs and allows cross-scale and cross-sector interactions for making such decisions and (2) allows actors to design, redesign, and choose different rules that align with prevailing conditions, for example actors’ incentives, desired outcomes, and other contextual factors.

In the next section, I synthesize the two conceptual pillars of a polycentric UFC system by breaking them down into four institutional design parameters (IDPs) and their associated key issues or questions. I discuss how these parameters and associated issues help to achieve desired outcomes such as the effective internalizing or managing of spillovers associated with UFCs. I then use Chicago’s NeighborSpace to provide empirical instances of these guiding parameters. Note that these institutional design parameters are not exhaustive (see Decaro et al. 2017; Ostrom 2005; Sarker and Itoh 2001). Again, the parameters can vary depending on other factors (e.g., type of commons under consideration, specific local conditions affecting the commons) and must be fine-tuned over time by subjecting them to different UFCs in the U.S. and elsewhere.

4 Achieving Polycentric Order in Governing UFCs: A Framework for Analysis and Design

An effective polycentric UFC should aim to achieve polycentric order, which means ensuring that the system can self-organize to address undesired outcomes. This requires ensuring that decisions about UFCs include the two above-noted conceptual pillars. Here, I classify these pillars into four institutional design parameters with key issues or questions, which Fig. 7.4 outlines and Table 7.1 summarizes. The first pillar is classified into two institutional design parameters: multiple decision centers or actors making decisions about UFCs (IDP 1) and opportunities for cross-scale and cross-sector interactions among UFC actors (IDP 2). The second pillar is also classified into two institutional design parameters: the overarching system of rules that guide decisions and interactions (IDP 3) and adaptability of the rules to align with actors’ incentives and decision outcomes (IDP 4).

Fig. 7.4

Decisions and their impacts (externalities) as actors interact within and across scales to participate in decisions about UFCs

Table 7.1 Institutional design parameters for evaluating and designing effective polycentric governance of UFCs

Figure 7.4 shows that we can consider two main decisions about UFCs: decision X (such as where to locate farm areas in cities) and decision Y (such as the value of land parcels used for farming compared to other land uses within cities). To evaluate and design effective polycentric governance of UFCs, decisions should involve multiple decision centers (IDP 1). Table 7.1 lists key issues to consider under each IDP, such as who is involved in making certain key decisions that directly or indirectly affect UFCs and which actor(s) wield(s) greater influence in specific decisions. Figure 7.4 shows that actors 1, 2, 3, 4, and 5 are involved in making decision X, and actors 4, 5, 6, 7, and 8 are involved in decision Y. Outcomes from both decisions (X and Y) can have direct and indirect impacts on each other. For instance, deciding where to locate farm areas in cities (decision X) could affect and also be affected by outcomes from appreciation or depreciation of surrounding land values (decision Y). These impacts create (1) payoff externality in decisions—when decisions impact each other—and (2) strategy externality in decisions—actors’ ability to transfer behavior, beliefs, and incentives from one decision to another (Bednar and Page 2007; Lubell 2013). We account for these externalities by attending to the rules and interactions that frame decisions about UFCs, which leads to the three remaining IDPs.

Decision centers or actors should be able to interact within and across governance scales and sectors, to make and challenge decisions and build consensus around decisions (IDP 2). These cross-scale and cross-sector interactions for making, challenging, and building consensus should be supported by diverse rules (IDP 3), which can also be adapted to align with actors’ interests and observed outcomes (IDP 4). In a polycentric governance system, interactions within and across scales and sectors for making decisions and/or changing rules present opportunities for actors to freely enter and exit decision-making, rule-making, and rule-enforcing arenas (Aligica and Tarko 2012; Ostrom 1972b). Such interactions ensure that outcomes from decision X can be used to drive outcomes from decision Y, or vice versa (positive payoff externality). For instance, because actors 4 and 5 are involved in making decisions X and Y, they can learn from one decision outcome to influence the outcome of the other decision (strategic externality). Other actors can also freely enter and exit decision arenas (X and Y) or interact to learn from one another, change and/or enforce rules, and resolve conflicts. Through such interactions and free entry and exit, actors can constantly adjust their relationships (self-organize) to address undesired outcomes as conditions in cities change. We use Chicago’s NeighborSpace to provide more-concrete instances of these IDPs at work.

5 Case Study: Chicago’s NeighborSpace Illustrates Emerging Polycentric Governance of UFCs

Chicago’s NeighborSpace illustrates how some of the four IDPs work in practice. My aim is not to demonstrate that NeighborSpace epitomizes the full extent of polycentric governance of UFCs; after all, there is more to be learned as NeighborSpace evolves over time. Kaufman and Bailkey (2000) observed signs of emerging polycentric governance of Chicago’s UFCs, evidenced by a constellation of actors involved in UFC decisions in Chicago: “Chicago’s motto, urbs in horto, the ‘city in a garden,’ is being realized by organizations in and out of government now creating the institutional context for entrepreneurial urban agriculture.”

Multiple overlapping decisions occur within Chicago’s UFCs, such as land acquisition and use decisions. Multiple actors (IDP 1) make these decisions through NeighborSpace, a nonprofit urban land trust formed through a 20-year intergovernmental agreement signed by three public agencies: the City of Chicago, the Chicago Park District, and the Cook County Forest Preserve District. NeighborSpace purchases land titles to vacant lands in Chicago, tests and deeds the land to community organizations, secures permits, and provides funding to insure the land. The executive director of NeighborSpace, Ben Helphand, summarizes their mission: “NeighborSpace shoulders the responsibility of property […], so that community groups can focus on gardening and community building” (Helphand 2015).

The intergovernmental agreement ensures interagency interactions (IDP 2) that allow representatives from each of the three public agencies and other NeighborSpace board members to influence decisions about land acquisition and use. Community organizations also influence which lots they want NeighborSpace to secure and what the land is used for, such as growing food crops and/or ornamental plants. NeighborSpace requires each community organization seeking partnership to have three leaders and at least 10 community stakeholders. There is a nested system of stakeholders and interactions (IDPs 1 and 2) involved in making decisions within and across organizations (from community organizations to NeighborSpace) and governance scales (from the neighborhood to state level), which safeguard against the dominance of a single agency or individual in the decision-making process.

Supporting this nested, polycentric decision-making system are the diverse and adaptive rules (IDPs 3 and 4) that structure actors’ decision making and interactions. Discussing all the formal and informal rules exceeds the scope of this chapter, so I address a few here. Kaufman and Bailkey (2000) highlighted an instance of informal cooperative rules used to govern how Neil Dunaetz (an organic vegetable farmer) interacted with young people in his neighborhood: these young people sold Neil’s farm produce in exchange for 20 percent of Neil’s profits. In terms of formal rules, the home clause in the Illinois State Constitution allowed the City of Chicago to establish NeighborSpace as a subsidiary jurisdiction. Furthermore, federal (e.g., US Code Title 26 §501(c)3) and Illinois state laws (POI-37, Illinois Department of Revenue and 1986 Illinois General Not for Profit Corporation Act, IL Admin. Code tit. 86 §130.120) support the creation of nonprofits such as NeighborSpace with self-governing and tax-exempt powers. Again, Mayor Richard Daley’s greening policy (Kaufman and Bailkey 2000) and the 20-year intergovernmental agreement were key sources of institutional support, which ensured the functioning of NeighborSpace. At the neighborhood level, NeighborSpaces’s fiscal agency and partnership agreements also specify fiscal procedures, site guidelines, roles and responsibilities, and conflict-resolution procedures for community organizations. More important, members of community organizations not only operate within existing rules but are also allowed to design their own rules for cooperation: “NeighborSpace meddles with the management of the gardens as little as possible. We do require loose standards for safety and insurance purposes, but, beyond that, step out of the way so community can act” (Helphand 2015).

At least some of these diverse rules are adaptive (IDP 4). For instance, as narrated by the executive director of NeighborSpace, an alderman was hesitant about the permanent transfer of a vacant lot in his ward to NeighborSpace. Thus, rather than working within the existing rules of land acquisition and use, the alderman instituted ad hoc rules that required community gardens in his ward to demonstrate success for three seasons before he would sign off on the permanent transfer of land. Here, we see the free entry of an alderman in rule-making decisions regarding land acquisition, which helped to build trust among him, NeighborSpace, and community groups. Similarly, Kaufman and Bailkey (2000) demonstrated how NeighborSpace had to adapt its internal procedures and decisions as a result of conflicting interactions with community residents who opposed the organization’s proposed agriculture demonstration site along North Sheffield Avenue. NeighborSpace’s ability to adapt their rules and decisions helped them to learn and reconsider how they could better align this proposal with residents’ interests.

This case analysis of NeighborSpace does not claim that Chicago has figured everything out. Rather, I highlight polycentric elements, such as IDPs, that allow actors to readjust their decisions and interactions in order to address tragic outcomes in the governance of Chicago’s UFCs. For instance, Vitiello (see Chap. 5 in this book) discusses how NeighborSpace and community gardeners readjusted when the support of Heifer International and the Greencorps program ceased in 2011. An effective polycentric governance system is expected to reorganize and re-establish order and stability by allowing actors to (re)adjust their decisions, interactions, and the rules by which they interact and make decisions, to address undesired outcomes. This is precisely what happened. NeighborSpace helped to form the Chicago Community Gardens Association to offer new support systems to community gardens, by sponsoring plant distributions, developing and maintaining shared resources such as tools and equipment, coordinating garden volunteers, and organizing trainings and workshops for community gardeners. Whether it was deliberate or not, the emergence of Chicago’s polycentric UFCs, evidenced through opportunities for local, state, and national actors to interact and decide within diverse and adaptive rules, is aptly expressed in the words of NeighborSpace’s Executive Director: “It [NeighborSpace] is like an estuary where the saltwater of government agencies mix with the freshwater of communities” (Helphand 2015).

6 Conclusion

In this chapter, I presented alternative conceptualizations for how to effectively govern urban food commons in U.S. cities. I first discussed the challenges of defining UFCs as a purely public or private good. Drawing from prior debates by Vincent and Elinor Ostrom and other scholars, I discussed how UFCs can, under different conditions, exhibit characteristics of all four types of goods: public, private, club, and CPRs. This makes it difficult to prescribe a one-size-fits-all institutional arrangement for governing UFCs. I then drew ideas from environmental federalism to discuss why UFCs should be governed through a mix of diverse institutional arrangements at multiple governance scales—including competitive market, cooperative, command-and-control, decentralized, and centralized models—to ensure that undesired outcomes are properly addressed, such as the capacity to effectively internalize spillover effects from UFCs.

To design such a mix of institutional arrangements, I used the concept of polycentric governance. Here, I identified the two conceptual pillars of polycentric governance and distilled them into four institutional design parameters to guide the design and evaluation of a polycentric UFC system. These IDPs included (1) multiple decision centers (or actors) making decisions about UFCs, (2) opportunities for cross-scale and cross-sector interactions among UFC actors, (3) an overarching system of rules to guide decisions and interactions, and (4) adaptability of the rules to align with actors’ incentives and decision outcomes. To ground these concepts in empirical examples, I used the case of NeighborSpace to illustrate how some of these IDPs manifest in Chicago’s UFCs. Through this case analysis, I underscored that polycentric governance is about safeguarding against the dominance of actors and the use of single, inflexible command-and-control rules in actors’ decisions and interactions. Multiple actors participating in decisions and using diverse, adaptive rules within and across scales and sectors can better adapt to changing conditions within a system. Whether Chicago’s NeighborSpace can endure as a model for designing polycentric UFCs is an empirical question that requires further analysis. Such analysis could occur, for example, through a comparative case study of Chicago and other UFCs, including a polycentric focus on food policy councils in the U.S., to refine and develop this chapter’s analysis of polycentric IDPs.