Conflict on the Red River: Applying the institutional collective action framework to regional flood policy

Abstract

In reaction to major flooding, local governments in Minnesota and North Dakota formed a cooperative network to construct a $1.2 billion flood diversion along the Red River of the North. Threatened by this diversion, a second set of governments formed their own network in opposition to flood policy. This study uses propositions derived from the Institutional Collective Action framework to examine formal contracts at the core of these cooperative networks, as well as the circumstances under which the contracts were negotiated. It considers the ability of the framework to understand interlocal cooperation where regional consensus is nonexistent. The study finds that the two sets of governments faced very different transaction costs, resulting in contrasting approaches to governance.

Keywords

Institutional Collective Action interlocal cooperation new regionalism local governance flood policy regional organizations

Introduction

On 28 March 2009, the Red River of the North crested at more than 20 feet above flood level, nearly overwhelming the city of Fargo, North Dakota (Nowatzki, 2009). The flood caused an estimated $55 million in damages and displaced hundreds of people from their homes and businesses (Kolpack and Suhr, 2010; National Oceanic and Atmospheric Administration, 2010). Unfortunately, only the scope of this flood was unusual: the Red River had surpassed flood levels nearly every spring since 1993, forcing local governments to spend some $65 million each year on preparation and clean-up (United States Army Corps of Engineers, 2009). Despite the ongoing threat, much of the Fargo-Moorhead metropolitan area remained under the protection of an assortment of temporary and permanent dikes barely capable of handling the annual stress. In response, local officials, working with the United States Army Corps of Engineers, called for the construction of an extensive set of flood-control infrastructure using funds from federal, state, and local sources. The centerpiece would be a 36-mile diversion around Fargo for an estimated cost of $1.2 billion (Kolpack, 2010). In 2011, the city of Fargo signed a Joint Powers Agreement (JPA) with other local governments, thereby creating a public entity to oversee the planning and implementation of flood policy.

While local actors were initially united on flood-fighting, the consensus soon wavered. Many smaller communities rebelled as it became clear that this policy required storing excess water on thousands of acres south of Fargo during a major flood (Reuer, 2010). Most notably, a collection of business owners, farmers, and residents formed the MnDak Upstream Coalition in opposition to the diversion. The Coalition proved remarkably effective at putting together its own cooperative network, cumulating with a JPA between Richland County, North Dakota and Wilkin County, Minnesota in 2012.

This study considers the evolving circumstances of Red River flood policy, which initially seemed like an example of ‘New Regionalism’ in action (Bauroth, 2013). It uses the Institutional Collective Action (ICA) framework (Feiock, 2013) to better understand the behavior of supporters and opponents of a Red River diversion. Specifically, the study examines the JPAs to discern the ICA dilemmas at work during the post-flood period and how solutions to these dilemmas varied by policy goal. In addition, the study recognizes the zero-sum nature of Red River flood policy and considers how it impacted efforts at regional cooperation.

Literature review

Traditionally, urban scholars could be divided into those who believed regional matters were best dealt with by a single entity and those who preferred a more fragmented approach. Progressive reformers claimed that the patchwork collections of municipalities, school districts, and special districts found within metropolitan areas were inherently corrupt and needed to be replaced by metropolitan-wide government (Hawley and Zimmer, 1961; Jones, 1942). Conversely, public choice scholars argued that a plethora of local governments provide citizens with more service options, thereby creating competition and efficiency (Hirsch, 1964; Tiebout, 1956). The consolidation-public choice debate dates to an era in which local entities generally focused upon local matters while state governments handled more regional concerns. In the twenty-first century, though, fragmentation has not only become a permanent feature of the metropolitan area, but spread to the suburbs and countryside (Savitch and Vogel, 2000). In addition, states have grown leery of taking the lead on regional matters (Collins, 2006). Consequently, local governments seeking regional policy must engage with other governments across the metropolitan area (Kraus, 2012).

A scholarly movement known as ‘New Regionalism’ attempts to not only understand the various manifestations of local cooperation, but prescribe ways of making such cooperation more effective and sustainable (Parks and Oakerson, 2000). The movement is based upon a distinction between ‘government’ and ‘governance.’ “The classic idea of government entailed formal institutions and elections and established decision-making processes and administrative structures” (Savitch and Vogel, 2000: 161). This approach typically produces a hierarchy of political authority with some entities, such as consolidated governments, granted broad powers over a wide range of services, and other entities, such as small special districts, confined to narrowly defined missions (Koch, 2013). The state implements regional policy by placing additional administrative layers across a metropolitan area in the form of a public authority or empowered county government.

In contrast, the idea of ‘governance’ holds that cooperation between local governments, nonprofits, and private corporations can be an effective way of underwriting regional services (Savitch and Vogel, 2000) and spreading the associated risks across multiple entities (Steinacker, 2010). Such cooperation manifests itself through contracts, memoranda of understanding, and other such arrangements. The governance perspective asserts that routine interactions between local entities can lead to the formation of cooperative networks (Bae and Feiock, 2012). Regional policy does not have imposed from above to be successful.

Despite the seemingly obvious benefits of New Regional ‘governance,’ many local governments have limited experience with interlocal cooperation. Indeed, officials report feelings of unease and distrust towards neighboring jurisdictions (Collins, 2006). Thus, even seemingly well-managed entities had difficulties in developing intra-jurisdictional networks. New Regionalism has also been criticized for its emphasis on efficiency over democracy (Gjertsen, 2014). The local electorate is often treated as just another obstacle for cooperating governments to overcome.

The ICA framework provides a means for understanding the growth in interlocal cooperation as well as the factors hindering further cooperation (Feiock and Scholz, 2010). Drawing upon the collective action literature, the framework starts with the premise that cooperation is inherently difficult in a fragmented environment due to the innate tendency by local actors to pursue short-term interests with little regard for neighboring jurisdictions. Most entities are reluctant to change their behavior unless some mechanism is put into place that incentivizes coordination. This is not easily done since the process of negotiations itself creates obstacles to cooperation in the form of transaction costs (Feiock, 2013).

Transaction costs are those costs associated with planning, adapting, and monitoring policy (Williamson, 1981). They arise from incomplete information, uncertainty about the motives of other actors, worries about autonomy, the likelihood of legal conflict, and other such matters (Brown and Potoski, 2005). These costs can make the risks associated with an agreement appear even greater than the benefits, thereby precluding cooperation. In addition, transaction costs vary by policy. For example, large capital projects are dependent upon long-term financial support by multiple partners. Defection by just a single partner can leave a project politically and financially vulnerable (Feiock, 2009). Consequently, the possibility of defection is a central component of negotiations. Such fears are less prevalent for policies involving smaller commitments, such as coordination of traffic lights between jurisdictions or information sharing between police departments (Feiock and Scholz, 2010). However, even these arrangements have transaction costs.

The ICA framework is used to recognize and surmount transaction costs associated with interlocal cooperation (Feiock and Scholz, 2010). Towards these ends, cooperation is considered along two dimensions: first, the extent to which individual autonomy is preserved while still imposing obligations upon participants. The resulting forms of governance range from informal networks that participants can easily exit at one end to consolidated metropolitan government with its rigid hierarchy at the other. Second, cooperation is placed in relation to the comprehensiveness of the policy under review. Local actors might focus upon a narrowly defined issue between two municipalities or a broad set of policies across multiple levels of government. The formal and informal rules directing these networks are key to understanding cooperation.

The institutional arrangements associated with interlocal cooperation can be categorized along three approaches. First, there is the ‘managed’ approach whereby a higher-level government, such as the state, mandates cooperation between local governments (Feiock, 2009). An existing regional institution, such as a council of governments, is then used to broker the specifics of an agreement (Carr et al., 2009; Hawkins and Carr, 2015). Under the managed approach, defection and noncompliance becomes costly for individual governments since the state can: (1) compel participants to follow through with their obligations; and (2) withhold funding.

Second, there is the ‘policy network’ approach where cooperation is based upon established social relationships (Hawkins and Carr, 2015). A history of interactions between actors enhances their sense of trust for one another, thereby reducing costs. Participants come to a consensus that is sealed through memorandums of understanding or other semi-formal arrangements. Enforcement procedures tend to be minimal: typically, participants can withdraw at little or no direct cost to themselves. However, withdrawal may damage long-established relationships, which serves as a means of enforcement on its own.

Finally, there is the ‘contract network’ approach where local governments make connections through formal agreements, such as service contracts and joint venture agreements (Hawkins and Carr, 2015). A contract network serves as a means of exchanging information, reducing predatory behavior, and coordinating resources (Andrew, 2010). Typically, participants have already had numerous interactions with one another over a range of policies before engaging in regional policy. The accumulated trust means there is less need for the state to play the role of facilitator. Local officials can compose regional policy largely on their own.

However, a reliance upon formal contracts can be problematic. For cooperation to work, the initial negotiations must establish some means of evaluating whether participants have fulfilled their obligations as well as a way of penalizing noncompliant participants (Hawkins and Carr, 2015). These necessities complicate negotiations since an effective enforcement mechanism reduces autonomy.

There are several dynamics to consider when applying the ICA framework: first, the self-organizational solutions require a certain degree of cooperation between local actors (Carr and Never, 2013). Cooperation is unlikely, though, without some preexisting consensus on policy. This implies the framework is best suited for resolving low-conflict issues where substantial agreement already exists. Second, local actors are generally willing to engage in cooperative relationships so long as costs remain low and benefits are roughly equivalent between participants (Anderson and Pierre, 2010). They are less likely to cooperate if they perceive themselves as bearing a greater proportion of the costs without an associated increase in benefits.

Finally, local actors may not be able to reach consensus on the more contentious matters. Certain regional issues are zero-sum in nature such that any effort at resolving matters produces clear winners and losers (Jones, 2010). Examples of such ‘wicked’ problems include: land use policies that encourage urban sprawl; economic policies that force local governments to compete with one another for new businesses; and welfare policies that focus on urban poverty (Carr and Never, 2013). While some actors would benefit greatly from solutions on such issues, others would assume additional costs for little apparent reward, reducing their incentive for regional cooperation.

The ability of the ICA framework to promote interlocal cooperation where regional consensus is nonexistent remains in question (Jones, 2010). The framework suggests that, under certain circumstances, uncooperative actors may be forced into giving up some autonomy to a regional actor, such as consolidated government (Feiock and Scholz, 2010). In the interests of inclusion, these uncooperative actors may be given some role regarding oversight of that regional actor (Chen et al., 2015) or greater access to federal and state resources (Spicer, 2015). However, they would have little say in the establishment of regional policy. Not surprisingly, the aggrieved actors would see this approach as fundamentally coercive.

There are two questions to keep in mind as this study applies the ICA framework to the issue of flood control in the Red River Valley. First, can meaningful cooperation in a fragmented environment occur without broad consensus across the major actors? Second, in situations where regional cooperation is impossible, does the forging of cooperative networks serve as another means through which governments compete over policy?

Propositions and methodology

This study examines conflicting approaches to regional flood policy in the Red River Valley. Towards that end, it compares the creation of two JPAs, one derived through the ‘contract network’ approach to interlocal cooperation and the other via the ‘policy network’ approach. This allows for a consideration of: (1) what each set of actors believed to be their most pressing transaction costs; (2) the mechanisms used to overcome those costs; and (3) the value of the ICA framework in a conflicted setting. The analysis focuses upon three propositions derived from the ICA literature.

Proposition one: Local entities must overcome transaction costs to achieve interlocal cooperation

The ICA framework applies theories of individual collective action to institutional actors to better understand cooperative behavior (Hawkins and Andrew, 2011). However, transaction costs, such as incomplete information, worries about autonomy, and uncertainty about the intentions of others, hinder cooperation by highlighting the risks associated with policy. To achieve their goals, local actors must agree upon some form of governance capable of alleviating transaction costs (Feiock, 2009; Feiock and Scholz, 2010).

Proposition two: Transaction costs vary by the policy and institutional arrangements under negotiation

While all efforts at coordinating policy face transaction costs, some policy areas are more susceptible to their debilitating effect than others (Feiock, 2009). Indeed, the impact of transaction costs varies by the costs and complexity associated with a specific policy area (Feiock, 2013; Feiock and Scholz, 2009). Thus, a policy of information-sharing between adjoining municipalities will be a relatively straightforward affair while large capital projects must overcome deep-seated fears regarding loss of autonomy and the burden of long-term obligations. This, in turn, has a decisive influence upon the institutional arrangements associated with local governance. Large capital projects generally require formal agreements that define responsibilities, address accountability, and is legally enforceable (Hawkins and Andrews, 2011). In contrast, modest cooperative efforts require little more than a memorandum of understanding.

Proposition three: The formal rules used to establish interlocal cooperation reflect those initial transaction costs

Effective cooperation requires some means of enforcing individual obligations (Andrew and Carr, 2013). The strength and complexity of these enforcement mechanisms are determined during initial negotiations over governance (Berado and Scholz, 2010). A careful analysis of the enforcement mechanisms as well as the circumstances under which they arose can reveal the transaction costs driving negotiations. This allows for a clearer sense of the dynamics at work in regional policymaking as well as possible causes of future conflict.

Scholars applying the ICA framework to new circumstances often rely upon the case study approach for their analysis (Callahan, 2007; Idt et al., 2012; Jang et al., 2012). This research follows that approach with an overview of Red River flood policy and its attendant actors. The research describes the zero-sum nature of this dispute and the resulting forms of interlocal cooperation using three sources: (1) newspaper articles; (2) documents from federal, state, and local governments; and (3) documents from the MnDak Upstream Coalition. The analysis pays close attention to contracts between governments as a gauge of intent.

The development of regional flood policy in the Red River Valley

The Red River basin is an intricate set of streams and tributaries across eastern North Dakota, northwestern Minnesota, and southern Manitoba (Ramsey and Skroch, 1996). The topography of this region is generally flat, which can lead to flooding after a rapid spring thaw or sudden thunderstorms. This flatness makes it difficult to implement standard flood control techniques, such as dams and reservoirs. At the center of the watershed is the Red River itself, which flows north along the North Dakota-Minnesota border. Situated on the Red River, the City of Fargo is the largest municipality in North Dakota with a population of 115,863 in 2014, or 15.7 percent of the state’s population (United States Bureau of the Census, 2015a). Fargo is in Cass County, which had a total population of 167,005. The city of Moorhead, directly across the river from Fargo, had 38,504 people in 2014. Moorhead is in Clay County, Minnesota which had a total population of 61,281.

As the 2009 flood receded, Fargo and its neighbors began their pursuit of a regional solution to perennial flooding, enlisting the United States Army Corps of Engineers to oversee the process (2009). Fargo, Moorhead, Cass County, and Clay County organized the Metropolitan Flood Management Committee during the summer of 2009 (Schmidt, 2010a). This Committee then formed a subgroup called the Metro Flood Management Work Group to work directly with the Corps. The Work Group consisted of 11 officials: two Fargo city commissioners, three Moorhead commissioners, two Clay County commissioners, three Cass County commissioners, an administrator from the Buffalo Red River Watershed District, and a manager from the Southeast Cass Water Resource District (Olson, 2009).

The Corps demanded that local officials submit their preferred flood-fighting plan by 15 April 2010 as well as an enumeration of the finances behind such a plan by 15 July 2010 (Schmidt, 2010b). Consequently, the process was designed for speed. As soon as the Corps released its analysis on certain aspects of flood policy, the Work Group would hold public hearings and make a recommendation (see figure 1). The member governments of the Committee would then vote on that recommendation.

Figure 1.

Red River of the North (Wikimedia, 2016).

As local governments considered their options, Fargo pressed ahead with securing a revenue stream to cover its share of future expenses. On 30 June 2009, nearly 91 percent of Fargo voters approved a half-cent sales tax to help pay for the still-to-be-determined plan (Fargo Forum staff, 2009). The tax was expected to raise $200 million over the next 20 years (Schmidt, 2009b).

On 20 October 2009, the Corps of Engineers presented a cost–benefit analyzes of 11 flood-fighting options (Schmidt, 2009a). They explained that the federal government would pay up to 65 percent of project costs so long as the benefits-to-cost ratio was greater than 1.0 per year. Corps officials recommended the Minnesota option whereby $871 million would be spent on a 25-mile channel east around Moorhead (Schmidt, 2010a). This would protect Fargo against a once-every-100-years flood. However, the Work Group expressed greater interest in the North Dakota option whereby $1.3 billion would be spent on a 36-mile diversion west around Fargo (Kolpack, 2010). Such a diversion would protect against a once-every-500-years flood. Initially, the Corps pegged the benefits-to-cost ratio for the North Dakota diversion at less than 1.0 per year. However, they raised those estimates to 1.29 after a second review, making the North Dakota diversion eligible for federal funding (Fargo Forum staff, 2010) (see Figure 2).

Figure 2.

Proposed Red River Diversion (Floodlist, 2014).

Preliminary projections suggested the North Dakota diversion would only add two inches to flood crests downstream from Fargo. However, later estimates predicted an impact of up to 10 inches, greatly increasing the likelihood of flooding in the northern cities (Nowatzki, 2010a).

On 18 March 2010, the Metro Flood Study Group recommended the North Dakota diversion to the Metropolitan Flood Management Committee (Fargo Forum staff, 2010). This recommendation was summarily approved. To secure its share of the costs, Cass County placed a sales tax levy on the November ballot. The measure won 64 percent of the vote, with much of that support coming from Fargo and West Fargo (Nowatzki, 2010b).

The City of Moorhead and Clay County did not submit a sales tax to the voters during this initial period (Abbott, 2016).

On 18 November 2010, 16 days after the Cass County election, the Corps released new plans for the North Dakota diversion (Shaffer, 2010). Under these revisions, the Corps would alleviate the Diversion’s downstream impacts by storing excess water on an undetermined number of acres south of Fargo. The Corps acknowledged water storage would have a negative effect upon upstream communities, though how substantial remained unclear (Nowatzki, 2010c).

On 30 March 2011, the Corps announced that water retention south of Fargo would be more impactful than originally supposed, adding at least three feet to upstream water levels during a major flood (Reisenauer, 2011). Thousands of acres could be submerged for weeks. Shortly after this announcement, residents, farmers, and property owners from south of Fargo formed the MnDak Upstream Coalition to oppose the plan. In addition, the Richland County Commission passed a resolution calling for policy that took the interests of the entire basin into consideration (Fargo Forum staff, 2011b).

On 11 July 2011, the cities of Fargo and Moorhead, Cass County, Clay County, the Cass County Joint Water Resources District, and the Buffalo-Red River Watershed District entered into a JPA to form the Fargo-Moorhead Area Diversion Authority (2011). A nine-member board would govern the Authority with three appointed by Fargo, one by Moorhead, three by Cass County, one by the Cass County Joint Water Resource District, and one in accordance to a separate JPA between Clay County and the Buffalo-Red River Watershed District.

The MnDak Upstream Coalition criticized Diversion Authority leadership for excluding smaller communities from the planning process (Fargo Forum staff, 2011a). The group retained a lawyer and urged residents to contact elected officials (Fargo Forum staff, 2011b). These efforts allowed the Coalition to become a key voice of opposition. The Coalition began coordinating efforts with Richland County, North Dakota and Wilken County, Minnesota. Richland County, just south of Cass County, was home to 16,432 residents while Wilkin County had 6,495 (United States Bureau of the Census, 2015b).

In January 2012, Richland County and Wilkin County created a JPA to bring legal challenges against the Diversion Authority (Berg, 2012). On 8 February 2012, Wilkin County Commissioners set aside $125,000 in county funds for use by the joint powers board (McDermott, 2012). As intended, the Richland–Wilkin JPA became the primary means through which other public and private entities could express opposition to the Diversion (Administrator, 2012).

On 19 August 2013, the Richland–Wilkin JPA filed a federal lawsuit against the Corps of Engineers, accusing them of unnecessarily expanding the scope of the flood project (Potter, 2013). The suit alleged this expansion was done not only to protect Fargo from a catastrophic flood, but to ensure that undeveloped land was removed from the floodplain. The lawsuit claimed the project would cause irreparable damage to upstream property values. The Upstream Coalition contributed $90,000 of the initial $156,000 spent on this litigation (Potter, 2014).

Analysis

The zero-sum nature of flood policy in the Red River Valley made regional cooperation difficult, if not impossible. While the proposed diversion would protect the cities of Fargo and Moorhead during a major flood, it would inundate thousands of acres in Richland and Wilkins counties. Consequently, flood policy became something of a conflict resolution problem where the majority attempted to force the minority into accepting their fate while the minority sought relief through legal action (Jones, 2010). In preparation for conflict, the two sets of governments sealed their respective alliances by signing JPAs.

A JPA is a contract between local governments for the shared administration of any power legally designated to those governments (North Dakota Century Code, n.d.). The contract specifies the powers to be shared as well as the mechanism for administering those powers. Whereas the powers of a special district are derived from the state, a JPA is the creation of a contract between local governments. Thus, it is a political subdivision and not a truly independent entity. A JPA is well-suited for regional policy since it allows for a variety of governing arrangements, membership can include entities from other states, and incorporation is not dependent upon state approval. Consequently, JPAs are a standard means of establishing cooperative relationships in North Dakota.

The Metro Flood Diversion Authority JPA

While the states of Minnesota and North Dakota demonstrated a willingness to help underwrite regional flood policy, they left the creation and implementation of that policy to Fargo, Moorhead, Cass County, and Clay County. These neighboring jurisdictions had a long history of cooperation with one another on an assortment of policy areas. However, the proposed Diversion would be, by far, their most expensive and risky project to date. Given the ambitious nature of the project, participants selected a ‘contract network’ approach whereby local governments are linked together “… in formalized and legally binding agreements” that, hopefully, allows for local autonomy (Hawkins and Carr, 2015: 25). While the threat of a devastating flood remained an immediate concern, the governments had several issues to work through before they could achieve regional policy. These issues can be categorized along three areas.

First, while fear of another flood brought governments together, certain wariness still lingered. Indeed, large infrastructure projects such as the diversion typically involve high transaction costs due to the greater risks of defection (Feiock and Scholz, 2010). However, these concerns over defection must be balanced against the desire for autonomy. Thus, each government worried that the other governments might abandon the JPA as the costs of flood control became apparent, leaving them to fund policy on their own. Conversely, each government also worried that their partners might force the JPA to pursue excessively expensive or foolhardy policies, leaving them with no way of escaping the fiscal and political consequences.

To reconcile these conflicting impulses, the participants negotiated a JPA contract that was a deliberate mixture of vagueness and obligation. The initial contract did not actually describe the specifics of the Diversion itself but, instead, noted “…the Project subject to the terms of this Agreement is the project arising out if the Fargo Moorhead Metropolitan Feasibility Study developed by the United States Army Corps of Engineers” (Fargo-Moorhead Area Diversion Authority, 2011: 7). Since that Study remained incomplete, the JPA contract did not require a detailed policy statement or division of labor for it to become operative. Indeed, the section of the contract ostensibly devoted to ‘Project Operations and Maintenance Costs’ was purposely left blank (Fargo-Moorhead Area Diversion Authority, 2011: 13). Clearly, participants did not wish to bind themselves to any plan just yet.

However, at some point, participants would have to commit to a more specified project. This commitment would have to be done in a manner that did not allow for defections as costs came due. Towards these ends, Article XIII of the JPA noted that the 2011 contract would be in effect until terminated upon the unanimous approval of participating governments. In addition, “it is the intent of the members that this agreement will terminate on or before the execution of the PPA [Project Partnership Agreement]” (Fargo-Moorhead Area Diversion Authority, 2011: 13). The PPA is the formal agreement between JPA participants and the Army Corps of Engineers to build flood-fighting infrastructure. It is, therefore, the last step before construction begins. After signing the PPA, participating governments would unanimously vote to terminate the old contract and bring the initial JPA to an end. The participants would then create a new JPA devoted to implementing the PPA. This adjustment would do two things: (1) provide governments with one last chance to exercise autonomy and remove themselves from the contract network; and (2) force governments to firmly commit themselves to implementation no matter how costly it might seem.

The Corps of Engineers produced an acceptable PPA in July 2016 (Associated Press, 2016). The participating governments subsequently dissolved the old JPA and replaced it with a new contract. However, the Buffalo Red River Watershed District opposed certain aspects of the PPA and decided against signing the new JPA (Springer, 2016). The Watershed District subsequently lost its seat upon the Diversion Authority’s governing board.

The second area of concern during negotiations was that the enforcement mechanisms in the final agreement would prove ineffectual at preventing defections. Indeed, with its focus on autonomy, the initial JPA had little within its bylaws that could be used to punish recalcitrant members. Member governments were obligated to remain a part of the JPA until terminated by unanimous vote. However, this principle of unanimity extended to the passage of budgets as well as modifications to the agreement. Consequently, a single participant could slow down and even scuttle plans altogether by refusing to support changes to the JPA.

For example, the Buffalo-Red River Watershed District protested the evolving PPA by refusing to vote in favor of the 2015 Diversion budget (Tran, 2015). The Watershed District’s protest quickly ended, but it placed the Diversion Authority in fiscal limbo.

The initial JPA was amended twice through unanimous agreement. However, enforcement mechanisms were not altered until the creation of the second JPA. Since the new JPA would implement the project, defection and recalcitrance by member governments became a central concern. Consequently, Article XX of the second JPA laid out a mediation process for disputes between members as well as between members and the Division Authority itself (Fargo-Moorhead Area Diversion, 2016). Notably, participants waived any right to a jury trial upon signing the JPA. In addition, “the Member Entities recognize and agree that they may be requested to make politically unpopular decisions and that it would not be fair or reasonable to withhold taking necessary action absent a Rational Basis for said inaction” (Fargo-Moorhead Area Diversion, 2016: 69). Non-performance of a participant’s obligations could result in suspension from the Authority’s decision-making process and, ultimately, removal from the JPA altogether. However, suspension or removal “…shall not be deemed as construed to alter any of the obligations of the Member Entity established, created, or accrued prior to its removal” (70), including debt.

The final area of concern was that the interests of North Dakota governments might diverge from those of Minnesota governments. Consequently, the initial JPA made clear distinctions between their obligations in terms of leadership and commitments. The contract specified that membership of the Diversion Authority would consist of nine members, seven appointed by North Dakota members and two appointed by Minnesota members (Fargo-Moorhead Area Diversion Authority, 2011: 13). North Dakota participants were responsible for 90 percent of local costs accrued in the development of the project, including execution of the PPA, while Minnesota participants were responsible for 10 percent (12). The North Dakota participants would even cover the Minnesota participants’ preliminary obligations until the PPA was signed (13). Thus, Moorhead did not have a direct financial obligation for the project nor did it pass any tax increases in support of the project (Abbott, 2016). The city was only required to request money from the state of Minnesota. Finally, the JPA also stipulated that while flood policy would be determined via majority vote, this majority had to include at least one Minnesota member.

The new JPA retained most of these features. However, the new contract placed greater liability on the North Dakota members since they would receive greater benefits from the Diversion (Tran, 2016). The Minnesota members’ share was capped at $100 million while the North Dakota members’ share had no cap. Thus, the North Dakota members pledged to cover any additional costs that might accrue during implementation of the PPA.

The Richland–Wilkin JPA

The contract at the core of the Richland–Wilkin JPA was much shorter and less complicated than that of the Metro Flood Diversion Authority JPA. Negotiations surrounding the formation of the Richland–Wilkin JPA were also more straightforward. The two counties seemed unconcerned about issues related to possible defection or loss of autonomy. At first glance, the purpose of the Richland–Wilkin JPA appears quite expansive: “to take any and all steps necessary… to oppose the planned construction of a Dam on the Wild Rice and Red River” (Richland-Wilkin Joint Powers Agreement, 2012: 1). In practice, though, this meant publicizing the downstream communities’ issues with the Diversion and pursuing legal action. While legal action could be expensive, the JPA contract made it clear that only the two member counties were liable for such costs. Consequently, public and private entities could join the JPA without putting themselves at financial risk.

Richland and Wilkin counties took a ‘policy network’ approach in their efforts to establish “…direct and indirect ties among public, nonprofit and private organizations within a region” (Hawkins and Carr, 2015: 25). Such an approach “provides the greatest degree of autonomy in decisions to enter and exit agreements” (25). Other upstream entities could participate in the JPA as they pleased with little, if any, direct cost to themselves.

However, there was one dilemma for the two counties to overcome before the cooperative network could proceed: the possibility of losing control over the JPA to other participants. Thus, the Richland–Wikin JPA contract has numerous safeguards to ensure this control. For example, the contract makes clear distinctions between ‘initial members,’ defined as Richland County and Wilkin County, and any subsequent ‘additional member’ (Richland-Wilkin Joint Powers Agreement, 2012: 1). The ‘initial member’ governments got to appoint the four-member governing board, which controlled budgeting, determined strategy, engaged lobbyists, and decided whether other entities could join the JPA. While the ‘additional members’ could petition for changes in the JPA, governing board determined whether such proposals could proceed. In addition, the contract designated the Richland County Auditor as treasurer and depository of all funds (2).

These arrangements made it much easier to recruit additional members. With Richland and Wilkin counties assuming financial responsibility, there was little risk in joining the JPA. Indeed, by announcing their involvement with the JPA, a tiny upstream government achieved a regional prominence they would not otherwise have had. Should they decide to abandon the JPA, these entities would only have to file a written notice of intent with the governing board. Thus, membership in the JPA became the primary way for upstream governments to protest Diversion planning.

The Richland–Wilkin JPA proved effective: the two counties would pursue legal action in the federal courts as the ‘voice’ of the downstream communities. The upstream communities found this policy network approach agreeable. By 2013, JPA participants included 10 municipalities, 15 townships, two school districts, three counties, a special district, and several nonprofit organizations (Administrator, 2012).

Discussion

While the zero-sum nature of Red River flood policy rendered regional cooperation impossible, the threat of another major flood hastened the development of competing cooperative networks. Even under duress, the participating governments paid close attention to the wording of their JPA contracts and its implications for future cooperation. The institutional arrangements behind cooperation shaped those contracts (Feiock, 2009).

Three propositions derived from the ICA literature guide the analysis. The first proposition asserts that transaction costs greatly hinder negotiations between local entities seeking regional cooperation. The process of overcoming those costs shapes both policy and the governing structure put in place to implement that policy. Even though they had just experienced a devastating flood, the governments of Fargo, Moorhead, Cass County, and Clay County still had to overcome significant ICA dilemmas related to possible defections and loss of autonomy before meaningful cooperation on flood control policy was possible. Likewise, the alliance between Richland County, Wilkin County, and other upstream entities would seem to be a straightforward affair requiring a loose-knit “policy network.” However, the two counties feared losing control over policy to the assortment of much smaller governments.

The second proposition argues that transaction costs vary by policy and institutional arrangements under negotiation (Feiock, 2009). The member governments of the Diversion Authority believed extensive infrastructure was needed to prevent massive flooding in the Fargo-Moorhead area. Unsettled by the size and cost of this capital project, though, they took great care in negotiating matters related to autonomy, enforcement mechanisms, and the diverging interests of Minnesota and North Dakota. This led to a complicated process by which an initial JPA oversaw the design phase of the project and a second JPA was set up to implement that design. In contrast, opponents to the Diversion had more straightforward negotiations, with the most serious concern pertaining to control over the JPA. This resulted in a much simpler contract and form of governance, though one dominated by Richland and Wilkin counties.

The final proposition asserts that the formal rules used to establish interlocal cooperation reflect the initial transaction costs. Therefore, an examination of JPA contracts will give a sense of the fears hindering negotiations between local governments in the Red River Valley. These fears were not publically discussed at the time, with much of the debate focusing on the extent the Diversion would damage upstream interests. Thus, the JPA contracts give a better sense of the dilemmas negotiators faced. The initial Diversion JPA was set up to ensure that member governments would see flood policy through to the end. However, the policy itself was deliberately vague and the initial JPA contract had several loopholes to ensure that individual members could slow down and even halt questionable policy. Their agreement also allowed for one final moment when a government could withdraw from the network before implementation could proceed. The subsequent JPA was more specific regarding enforcement mechanisms.

In contrast, the Richland–Wilkin JPA was written so that two counties would assume all financial risk. This meant that the “decision costs” associated with joining this JPA were quite low for other upstream entities, thereby allowing for the expansion of the cooperative network (Feiock and Scholz, 2010: 15).

Conclusion

The conflict over flood policy in the Red River Valley provides a useful illustration of the ICA framework. The contrast between the approaches taken by the two sets of actors allows for four observations. First, regional flood policy quickly degenerated into a zero-sum game. The solutions offered by Fargo and its allies would inundate the upstream communities during a major flood. Conversely, eliminating certain elements of that proposal would leave the largest city in North Dakota vulnerable to massive flooding. Faced by such existential threats, the two sets of governments were unwilling to compromise on most issues. This conflict highlights the limitations of interlocal cooperation (Jones, 2010) and serves as a reminder of the role played by coercion in metropolitan politics, including governance (Post, 2004).

Second, the central concern in negotiations between local officials was, ultimately, loss of control over policy “through the emergence of new governance relationships” (Zeemering, 2016: 2348). This loss of control pertains not only to decision-making but also the allocation of valuable resources as determined by that new form of governance. Officials took considerable fiscal and political risks in giving up such powers. Consequently, they insisted upon intricate negotiations over the nature of that governance even in the face of such immediate dangers as flooding or destruction of property.

Third, the JPAs used by both sets of government are examples of “self-organizing federalism” in action. Self-organizing federalism is “… the endogenous development and maintenance of institutional mechanisms that mitigate a recognized ICA dilemma by those directly affected by the dilemma” (Feiock and Scholz, 2010: 5). The creation of these mechanisms is a “bottom-up” action whereby local governments, rather than the state or national government, determine the means through which ICA dilemmas are overcome. In this way, cooperative networks can achieve regional policy without sacrificing control to higher forms of government.

Finally, by placing the formal contracts in relation to the circumstances under which they were negotiated, this study provided a deeper understanding of regional flood policy in the Red River Valley. What seemed, at first, to be an example of New Regionalism, proved more complicated as cooperative networks arose in conflict with one another.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Administrator (2012) Fargo, Cass County and the diversion authority face rising opposition. FMDam.org. Available at: http://fmdam.org/fargo-cass-county-and-the-diversion-authority-face-rising-opposition/ (accessed 10 December 2015).

Anderson

Pierre

(2010) Exploring the strategic region: Rationality, context, and institutional collective action. Urban Affairs Review 46(2): 218–240.

Andrew

(2010) Adaptive versus restrictive contracts: Can they resolve different risk problems? In: Feiock

Scholz

(eds) Self-organizing Federalism: Collaborative Mechanisms to Mitigate Institutional Collective Action Dilemmas, New York City, New York: Cambridge University Press, pp. 91–113.

Andrew

Carr

(2013) Mitigating uncertainty and risk in planning for regional preparedness: The role of bonding and bridging relationships. Urban Studies 50(4): 709–724.

Associated Press (2016) ‘Signing ceremony scheduled for Red River project.’. Albert Lea Tribune. 11 July. Available at: http://www.albertleatribune.com/2016/07/signing-ceremony-scheduled-for-red-river-project/ (accessed 8 August 2016).

Bae

Feiock

(2012) Managing multiplicity coordinating multiple services at a regional level. State and Local Government Review 44(2): 162–168.

Bauroth

(2013) Government fragmentation and emergency planning: Findings from the 2009 Red River flood and its aftermath. In: Cepiku

Jesuit

Roberge

(eds) Making Multilevel Public Management Work: Stories of Success and Failure from Europe and North America, Boca Raton, FL: CRC Press, pp. 141–158.

Berado

Scholz

(2010) Self-organizing policy networks: Risk, partner selection, and cooperation in estuaries. American Journal of Political Science 54(3): 632–649.

Berg S (2012) Join Us In Our Efforts. Wahpeton Daily News. Available at: http://www.wahpetondailynews.com/opinion/letters_to_editor/join-us-in-our-efforts/article_02db701c-6d22-11e1-a8b6-001871e3ce6c.html (accessed 10 December 2015).

10.

Brown

Potoski

(2003) Transaction costs and institutional explanations for government service production decisions. Journal of Public Administration Research and Theory 13(4): 441–468.

11.

Callahan RF (2007) Governance: The collision of politics and cooperation. Public Administration Review 67(2): 299–301.

12.

Carr

LeRoux

Shrestha

(2009) Institutional ties, transaction costs, and external service production. Urban Affairs Review 44(3): 403–427.

13.

Carr J and Never B (2013) Confronting wicked problems in the Metropolis: Is it possible to make these problems tame? Paper prepared for annual meeting of the American Political Science Association, August 2013.

14.

Chen

S-H

Feiock

Hsieh

(2015) Regional partnerships and metropolitan economic development. Journal of Urban Affairs 38(2): 196–213.

15.

Collins

(2006) Interlocal Service-Sharing Agreements, Washington, DC: International City/County Management Association.

16.

Fargo Forum staff. Fargo voters approve flood sales tax by whopping margin. 2009. Fargo Forum: A1, 30 June.

17.

Fargo Forum staff (2010) Corps to request approval of North Dakota diversion. Fargo Forum, 2 April. Available at: http://beta.inforum.com/event/article/id/274147/ (accessed 10 December 2015).

18.

Fargo Forum staff. Local councils reaffirm support for ND diversion. 2011a. Fargo Forum: A1, 5 April.

19.

Fargo Forum staff. Richland county opposes F-M diversion. 2011b. Fargo Forum: A6, 19 April.

20.

Fargo-Moorhead Area Diversion Authority (2011) Limited Joint Powers Agreement: Metro Flood Diversion Project. Available at: http://www.fmdiversion.com/pdf/JPA_w%201st%20and%202nd%20amendment_Limited%20Joint%20Powers%20Agreement%20-%20Fully%20Executed.pdf (accessed 10 March 2016).

21.

Fargo-Moorhead Area Diversion (2016) Joint Powers Agreement By and Between City of Moorhead, Minnesota and City of Fargo, North Dakota and Clay County, Minnesota and Cass County, North Dakota and Cass County Joint Water Resource District. West Fargo, North Dakota: Fargo-Moorhead Area Diversion.

22.

Feiock

(2009) Metropolitan governance and institutional collective action. Urban Affairs Review 44(3): 356–377.

23.

Feiock

(2013) The institutional collective action framework. Policy Studies Journal 41(3): 397–425.

24.

Feiock

Scholz

(2010) Self-organizing governance of institutional collective action dilemmas. In: Feiock Scholz (ed.) Self-Organizing Federalism: Collaborative Mechanisms to Mitigate Institutional Collective Action Dilemmas, New York, NY: Cambridge University Press, pp. 3–32.

25.

Floodlist (2014) Fargo Moorhead Flood Diversion Project. Available at: http://floodlist.com/america/usa/fargo-moorhead-flood-diversion-project (accessed 10 December 2015).

26.

Gjertsen

(2014) Legitimacy in interlocal partnerships: Balancing efficiency and democracy. Urban Studies 51(9): 1926–1942.

27.

Hawkins

Andrew

(2011) Understanding horizontal and vertical relations in the context of economic development joint venture agreements. Urban Affairs Review 47(3): 385–412.

28.

Hawkins

Carr

(2015) The costs of services cooperation: A review of the literature. In: Henderson

(ed.) Municipal Shared Services and Consolidation: A Public Solutions Handbook, New York City, New York: Routledge, pp. 17–35.

29.

Hawley

Zimmer

(1961) Resistance to unification in a metropolitan community. In: Janowitz

(ed.) Community Political Systems, New York City, New York: Free Press.

30.

Hirsch

(1964) Local versus areawide urban government services. National Tax Journal 17: 331–339.

31.

Idt J, Nadia A and Marie-Pierre L (2012) Metropolitan collective action: when the project is not linked to common interest. The case of the intermodal transit network in the Lyon metropolitan area. Urban Research & Practice 5(2): 223–238.

32.

Jang HS, Jesus V, Jung WK, et al. (2012) Understanding the diverse forms of nonprofit collaborations: A case study of communities in schools of north texas and its partner organizations. Journal of Public and Nonprofit Affairs 1(2): 100–117.

33.

Jones B (2010) Conflict, power, and irreconcilable preferences: Some limits to self-organizing mechanisms. In: Feiock and Scholz (eds) Self-Organizing Federalism: Collaborative Mechanisms to Mitigate Institutional Collective Action Dilemmas. New York City, New York: Cambridge University Press, pp. 73–90.

34.

Jones

(1942) Metropolitan Government, Chicago: University of Chicago Press.

35.

Koch

(2013) Overestimating the shift from government to governance: Evidence from Swiss metropolitan areas. Governance 26(3): 397–423.

36.

Kolpack

(2010) Area officials debate flood projects. Fargo Forum: C1, 1 February.

37.

Kolpack

Suhr

(2010) F-M not alone in US flooding. Fargo Forum: A1, 6 March.

38.

Krause R (2012) An Assessment of the Impact that Participation in Local Climate Networks Has on Cities' Implementation of Climate, Energy, and Transportation Policies. Review of Policy Research 29(5): 585–604.

39.

McDermott

(2012) Set money aside: Board dedicates funds for opposition. Wahpeton Daily News. 8 February. Available at: http://wahpetondailynews.com/news/board-dedicates-funds-for-opposition/article_afac10d2-5281-11e1-8855-0019bb2963f4.html (accessed 10 December 2015).

40.

National Oceanic and Atmospheric Administration (2010) Flood Damages Suffered in the United States During Water Year 2009, Washington, DC: NOAA.

41.

North Dakota Century Code (n.d.) Chapter 54-40.3: Joint powers agreements. Available at: http://www.legis.nd.gov/cencode/t54c40-3.pdf?20160306115557 (accessed 14 February 2016).

42.

Nowatzki

(2009) Fargo defends decision not to evacuate. Fargo Forum: C3, 30 March.

43.

Nowatzki

(2010a) Landowners organize to oppose diversion. Fargo Forum: A1, 11 April.

44.

Nowatzki

(2010b) Cass approves diversion sales tax. Fargo Forum: A1, 2 November.

45.

Nowatzki

(2010c) New plan is ‘scary’ for those upstream. Fargo Forum: A1, 20 November.

46.

Olson

(2009) Flood management work group aims to find consensus. Fargo Forum: A1, 24 October.

47.

Parks

Oakerson

(2002) Regionalism, localism, and metropolitan governance. State & Local Government Review 32(3): 169–179.

48.

Post SS (2004) Metropolitan area governance and institutional collective action. In: Feiock (ed) Metropolitan Governance: Conflict, Competition, and Cooperation. Washington, D.C.: Georgetown University Press, pp. 67–92.

49.

Potter K (2013) F-M diversion opponents sue U.S. Army Corps of Engineers. Fargo Forum, 19 August. Available at: http://www.inforum.com/content/f-m-diversion-opponents-sue-us-army-corps-engineers (accessed 10 December 2015).

50.

Potter K (2014) Nearly $500K spent so far in F-M diversion lawsuit. Fargo Forum, 10 May. Available at: http://www.inforum.com/content/nearly-500k-spent-so-far-f-m-diversion-lawsuit (accessed 10 December 2015).

51.

Ramsey

Skroch

(1996) The Raging Red: The 1950 Red River Valley Flood, Grand Forks, ND: Valley Heritage Press.

52.

Reisenauer

(2011) Corps diversion information shows buyouts warranted for all of Oxbow, Hickson, and Bakke subdivision. Fargo Forum. 30 March. Available at: http://www.inforum.com/content/latest-corps-diversion-information-shows-buyouts-warranted-all-oxbow-hickson-and-bakke (accessed 10 December 2015).

53.

Reuer

(2010) Upstream residents rally over diversion. Fargo Forum. 9 December. Available at: http://www.inforum.com/content/upstream-residents-rally-over-diversion.

54.

Richland–Wilkin Joint Powers Agreement (2012) Joint Powers Agreement establishing a board of directors to organize and direct a challenge to the construction of a dam. Available at: http://fmdam.org/wp-content/uploads/2012/03/Joint-Powers-FINAL.pdf (accessed 13 March 2016).

55.

Savitch

Vogel

(2000) A symposium: New regionalism and its policy agenda. State & Local Government Review 32(3): 158–168.

56.

Schmidt

(2009a) Dilworth, WF unhappy about being left out of diversion talks. Fargo Forum: A1, 25 October.

57.

Schmidt

(2009b) Fargo considers budget issues for 2010. Fargo Forum: A1, 27 December.

58.

Schmidt

(2010a) ND plans fill Corps criteria, but Minn. option preferred. Fargo Forum: A1, 2 February.

59.

Schmidt

(2010b) Officials weigh how to pay for diversion project. Fargo Forum: A4, 2 February.

60.

Shaffer

(2010) New details of Fargo-Moorhead diversion. Fargo Forum, 18 November.

61.

Spicer

(2015) Cooperation, coordination and competition: Why do municipalities participate in economic development alliances? Canadian Public Administration 58(4): 549–573.

62.

Springer

(2016) Buffalo Red River watershed board rejects diversion agreement. Fargo Forum: A1, 24 May.

63.

Steinacker

(2010) The institutional collective action perspective on self-organizing mechanisms. In: Feiock Scholz (ed.) Self-Organizing Federalism: Collaborative Mechanisms to Mitigate Institutional Collective Action Dilemmas, New York, NY: Cambridge University Press, pp. 51–72.

64.

Tiebout

(1956) A pure theory of local expenditures. Journal of Political Economy 64(5): 416–424.

65.

Tran

T-U

(2015) Divisions in ranks of diversion authority. Fargo Forum: A1, 10 April.

66.

Tran T-U (2016) ND entities given more liability in new Diversion Authority ‘constitution.’ Fargo Forum: A1, 25 February.

67.

United States Army Corp of Engineers (2009) Scoping Document: Fargo-Moorhead Metropolitan Area Flood Risk Management Environmental Impact Statement, St. Paul, MN: U.S. Army Corps of Engineers.

68.

United States Bureau of the Census (2015a) Quick Facts: North Dakota. Available at: http://quickfacts.census.gov (accessed 13 March 2016).

69.

United States Bureau of the Census (2015b) Quick Facts: Minnesota. Available at: http://quickfacts.census.gov.

70.

Wikimedia (2016) Map of the Red River of the North. Available at: https://upload.wikimedia.org/wikipedia/commons/6/66/Redrivernorthmap.png (accessed 26 September 2016).

71.

Williamson

(1981) The economics of organization: The transaction cost approach. American Journal of Sociology 87(3): 548–577.

72.

Zeemering

(2016) Assessing local elected officials’ concerns about interlocal agreements. Urban Studies 53(11): 2347–2362.