Reasons for Collaboration

Organizations collaborate for many reasons. The two reasons cited most frequently in the literature are to obtain needed resources and to solve complex problems. Analyses that emphasize resource needs start with the premise that “no organization is an island.” Organizations seek relationships with other organizations to obtain the resources they need (Pfeffer and Salancik, 1978). Cyert and March (1963) frame the resource problem as one of uncertainty. That is, a main cause of uncertainty is resource scarcity. An organization is vulnerable if the resources it needs are not consistently available. From this standpoint, organizations adopt collaboration as a strategy for reducing uncertainty and achieving stability. Although organizations often need tangible resources, such as buildings or equipment, the literature also devotes attention to collaborations that provide access to intangible resources, such as expertise, technology, knowledge necessary for innovation (Powell, Koput, and Smith-Doerr, 1996), or even legitimacy (Van Slyke, 2007).

Another stream of research emphasizes the importance of resource bundles (Barney, 1991). The concept of bundles refers to combinations of an organization's internal and external resources that can yield a synergy (Harrison et al., 2001). According to this way of thinking, a collaboration is motivated not as much by pursuit of resources critical to survival but rather by the quest for resources that can provide a strategic advantage, e.g., by making the organization more competitive. This perspective emphasizes the complementarity of resources and the value of resources when they are combined. The value of combined internal and external resources depends on an organization's objectives (Soda and Fulotti, 2017). Whether resources are complementary depends on several factors, including the degree to which the resources sought differ from those that the organization already has (Teece, 1986), the potential synergies when resources are combined (Dyer and Singh, 1998), and whether combining resources will accomplish strategic aims (Wang and Zajac, 2007).

Empirical research firmly establishes that collaboration is more likely to occur under both conditions, that is, when it reduces uncertainty over resources an organization needs (Hillman et al., 2009) and when benefits can be expected to accrue from resource complementarity (Harrison et al., 2001). For example, Weisbrod (1997) makes the case that competition for resources has prompted nonprofit organizations to consider alternatives to going it alone. Choi and Choi (2005) discuss partnerships between state governments and external organizations when the federal government conditions federal aid (i.e., resources) on collaboration. Bryson, Crosby, and Stone (2006, p. 46) point out that “cross-sector collaborations are more likely to form in turbulent environments,” that is, in fluctuating circumstances that create uncertainty about resources.

Analysts of nonprofit organizations often refer to collaboration as imperative (Butcher and Gilchrist, 2016; Murphy and Bendell, 1999). They also emphasize, however, that those seeking collaborative arrangements must carefully consider the role of stakeholders in affecting the collaboration decision. External stakeholders can affect the decision to collaborate. In a study by Nasi et al. (1997), stakeholders cautioned a forestry organization to consider profitability in deciding on a collaboration aimed at environmental preservation. In a nascent partnership between a Canadian supermarket chain and a national advocacy group called Pollution Probe, external stakeholders objected to a partnership (Lawrence and Hardy, 1999). According to Ebers (2004), parties are not always equally enthusiastic about collaborations. Nonprofits are cautiously optimistic about partnering with business in comparison (Elers, 2004).

Major social problems have wide scope and complexity and often involve multiple policy areas (Vurro, Dacin, and Perrini, 2010; Westley and Vrendenburg, 1997). Such problems have led those confronting them to coin the term “wicked problems.” Wicked problems have “circular causality, persistence, absence of well-structured alternative solutions, or relative lack of room for trial and error learning” (Rittel and Webber, 1973; Dorado and Ventresca, 2013, p. 69).

For example, the problem of homelessness has implications beyond housing policy. Homelessness involves challenges in health, drug, and employment policies. Homelessness often disproportionality affects military veterans. A single organization usually lacks the capacity and range of expert knowledge to address complex social problems adequately (Eisenhardt and Schoonhoven, 1996; Waddock and Post, 1995; Bryson, Crosby, and Stone, 2006). Government increasingly turns to cross-sector collaboration to address complex social issues (Stadtler and Karakulak, 2020) in many policy areas. Research on cross-sectoral arrangements occurs in the domains of environmental policy (Grady and Chen, 2006), emergency management (McGuire, 2009), and disaster relief (Curnin and O'Hara, 2019), among other policy areas.

Problems can also be complex when they involve a range of different stakeholders with different views on how the problem should be addressed. In the delivery of health care public, private, and nonprofit lab centers, dialysis centers, and hospices all play some part in shaping how any health program will operate. Community hospitals, sometimes public and sometimes nonprofit, treat thousands of patients on a daily basis. Private sector insurers, for their part, make decisions on coverage and charge copayments. All of these entities likely have different views on such issues as who should receive health services, how services should be delivered, how much it should cost, and who should pay. In addition, efforts to solve one aspect of the problem such as eligibility for coverage may have cascading effects, e.g., on government budgets and insurer solvency. The range of effects, long-term and short-term, intended and unintended, are more likely to surface and to be considered when multiple and diverse stakeholders are involved in the service. Collaboration that includes several parties can be a way to obtain buy-in from stakeholders as well as to preempt or deal with adversarial relationships (Pasquero, 1991).

Partner Selection

Collaborations can include a range of organizational partners: large and small, public and private, and representing multiple industries. Partner selection is one of the most important decisions that organizations make when they consider collaborating (e.g., Rothaermel and Boeker, 2008; Shah and Swaminathan, 2008; Beamish and Lupton, 2009; Dekker and Van den Abbeele, 2010).

The rationale for a public manager to seek collaboration partners from the same sector – e.g., other public agencies instead of partners from profit-driven organizations – depends on whether sector distinctions have implications for outcomes. Consider a government agency that considers a partnership with a private, for-profit organization to vaccinate a population at a specified (low) cost by a given date. The government agency may have a mandate to provide the same health services to all, and to provide vaccines for all segments of the population, regardless of their ability to pay. The private organization may have different incentives. The private organization may need to consider profitability and proprietary control of their resources. Thus, government agencies differ from their private sector counterparts in their incentives to offer services to those who are constrained in their ability to pay. As pharmaceutical corporations began to make vaccines available in the COVID-19 pandemic, the vaccines disseminated in wealthy nations earlier and more rapidly than in nations with lower levels of economic wealth and development. Representatives of the less wealthy nations, and of international organizations such as the United Nations, began to call for the pharmaceutical corporations to share their vaccines, and related knowledge and information, with the less fortunate nations. The corporations reportedly balked at sharing knowledge that would compromise their ownership of patents for the vaccines.

In a recent theoretical contribution that addresses cross-sector collaborations, Lazzarini (2020) offers a theory of organizational alignment by identifying comparative outcomes for different forms of collaboration, including government in partnership with a for-profit organization and government in partnership with a nonprofit social enterprise. According to the theory, distinct organizational forms change the managerial incentives to provide inclusive high-quality social services based on three main factors: the relative attractiveness of beneficiary segments, their cost structure, and extent of fixed costs to implement and operate services. Some combinations of such factors can create disincentives for nongovernmental, for-profit participants (and therefore compromised outcomes) when a public program aims to create social value, defined broadly as social benefits to a given population (including more vulnerable groups).

The literature relevant to partner selection often extols the attributes of nonprofits. For example, Van Slyke (2007) describes the unique strengths that nonprofits bring to partnerships with government for the provision of social services. Van Slyke (2007; pp. 143, 150) describes the potential advantages of government-nonprofit relationships. Nonprofits often have general experience in serving clients and elected officials, citizens, and the media often perceive nonprofits as legitimate. Similarly, Waddell and Post (1991) point to nonprofits' experience with managing stakeholder relationships (p. 3).¹ Witesman and Fernandez (2013) find a number of differences in the way public managers structure and manage contractual relationships with nonprofits compared to relationships with for-profits. They conclude that nonprofits are perceived to be more trustworthy and reliable partners for government.

Cross-sector partnerships are generally considered valuable because they can provide a range of competencies for addressing complex social problems (Austin, 2000; Waddell, 2000). Yet there are important contingencies to consider. Bryson and Stone et al. (2006, 52) caution that cross-sector relationships need to be built on already strong relationships with key political and professional constituencies. If they are, the success of cross-sector relationships is more likely. Gulati (1995) makes a similar case regarding the advantage of prior working relationships. On the other hand, in assessing R&D collaborations, Diestre and Rajagopalan (2012) observe that governments prefer government-business partnerships. In these conditions, the perception that businesses embody a culture of entrepreneurship is likely to be a factor in partner selection.

In collaborations initiated by government, a public agency or public department is the entity making the choice among partners. A growing literature in public management emphasizes the importance of shared values when selecting partners. Drawing on Moore's (1995) work on public values, Stoker (2006) articulates a public value management model of collaboration, which he describes as an overarching framework for collaborative governance (Stoker, 2006, p. 41). Other public management scholars also emphasize the importance of finding partners that fit with the pursuit of public value, as well as management skills focused on conflict resolution, trust building, information, and goal clarity (Domberger and Fernandez, 1999; Entwistle and Martin, 2005).

At the same time, van Thiel and van der Wal (2010) caution that we should not assume value congruence within the public domain. This is important since multi-partner collaborations often include multiple government agencies, and value incongruence among potential partners makes collaboration and partnering more challenging. Combining data from two Dutch surveys (n=324), they find evidence of value incongruence between ministries (that is, the parent public organizations in the Netherlands) and semi-autonomous executive agencies under the auspices of the ministry, called quangos. Quangos receive government funding but operate at “arm's length” (with high levels of autonomy) from the ministries. Quangos are “quasi-NGOs”; NGOs are nongovernment organizations. A quango usually administers a specific function, as with an Arts Council or a Forestry Commission. Van Thiel and van der Wal also find different levels of value incongruence among quangos. They conclude that the quality of relationships between parent ministries and quangos also varies (see also Lyons et al., 2006).

Collaboration Mechanisms

Collaboration mechanisms are the tools that facilitate continuity, lessen the potential for conflict, and make collaboration less vulnerable to dissolution.² They are the formal and informal rules that provide structure for interaction (Hoffman, 1999). Put another way, mechanisms are the glue intended to sustain the collaboration. Some scholars predict collaboration failure without the right mechanisms in place to transform the relationship from a lose-lose situation to a win-win situation (e.g., Brown, Potoski, and Van Slyke, 2016).³

Rules can serve as collaboration mechanisms. In describing the multi-level institutional elements of governance, Nobel Laureate Elinor Ostrom (1990) identified three types of rules that guide behavior: operation rules, policy rules, and constitutional rules. In a collaborative or network setting, operation rules govern day-to-day network activities. Policy rules determine what activities are allowed in networks, and constitutional rules decide who can participate in the collective decision-making process and how collective choice rules can be changed. These levels of rules can play important roles in governing collaborations and partnerships (see also, Stone et al., 2010, 2013). Uncertainty and the lack of information are key factors that create the need for collaborative mechanisms that foster accountability. Since parties cannot anticipate all possible scenarios ahead of time, there will always be gaps and omissions in the agreement. Collaborating partners frequently address this challenge by inserting clauses in the formal contract documents that specify how contingencies will be handled. Alternatively, partners may appoint members to a board dedicated to settling disputes should they arise (Williamson, 1991, p. 280). The possibility that parties with an information advantage may use that advantage to exploit other parties is another reason for accountability mechanisms. This can occur when service quality or level of effort is not easily measured. But it can also occur when a party makes upfront investments specific to the relationship at hand (Williamson, 1986, 1991). In such a situation, the investing party depends on contract continuation, at least until the investment is recouped; this could lead the non-investing party to behave opportunistically (Coase, 2000), e.g., by forcing a renegotiation for more favorable terms. This is called the hold-up problem, as the non-investing party can hold up the former for the value of the commitment. A contract provision that requires a reimbursement of the investment in the event the contract terminates prematurely is one possible mechanism to address the hold-up problem (Rogerson 1992).

Rosell and Saz-Carranza (2020) and their colleagues analyze different structural mechanisms for accountability. They empirically examine the governance configurations of 37 rule-enforcing networks in the European Union, finding three basic structures used for accountability: a configuration with legal accountability, which is characterized by having a board of appeals; a configuration with administrative accountability that, in addition to a board of appeals, has powerful executive boards and professional experts in the network plenary; and configuration providing for democratic accountability, which incorporates legislative representatives into the network (Saz-Carranza, Albareda, and Ryan Federo, 2020). Devarakonda and Reuer (2018) provide a similar example. They explore how administrative controls in collaborations regulate knowledge transfers across partners. These administrative controls can take the form of board-like joint committees having explicitly delineated authority over certain alliance activities. This research shows that governing committees can serve as important instruments for administrative control and demonstrates how the contract serves as a mechanism to facilitate knowledge flows. It also describes how governing committees safeguard against a partner misappropriating resources, particularly when a partner possesses the incentive and ability to engage in such behavior.

The need for accountability is not the only reason for collaboration mechanisms. Administrative mechanisms are often employed to facilitate communication and coordinate tasks. Collaboration agreements can involve multiple actors simultaneously completing different tasks, all toward a common end. Complex tasks, such as those that require reciprocal coordination, require more elaborate administrative mechanisms (Thompson, 1967). Such cases require more interactive coordination and mutual adjustment (Adler, 1995; Thompson, 1967). One way to do this is to specify communication and decision-making processes in the collaboration agreement (Gulati and Singh, 1998). As for the logical underpinnings of coordination, Reuer and Devarakonda (2016) develop a theory of the contractual delegation of authority and posit that partners are more likely to employ steering committees to help partners guide their interactions and address unanticipated contingencies when there is a substantial need for coordinated adaptation.

In addition to categorizing mechanisms according to the challenges they address, the literature frequently distinguishes between formal and informal mechanisms. Many of the formal mechanisms discussed in the literature are explicit in contract documents. In general, written contracts set standards of conduct and guide the parties' behavior. Within the contract documents themselves, partners commonly encourage cooperative behavior with incentives and discourage noncooperative behavior (also called “perfunctory behavior”) with sanctions.

Researchers offer several pointers for developing specific and effective contract requirements, including ways to enhance partner accountability (e.g., Romzek and Johnston, 2002; LeRoux, 2007; Malatesta and Smith, 2011), adapting the contract to service characteristics (Warner and Hefetz, 2010), aligning values and institutions (Brown, Potoski, and Van Slyke, 2006), and the need to strike the right balance between incentives and sanctions (Brown, Potoski, and Van Slyke, 2016, 47).

Some scholars are critical of formal mechanisms and propose that parties rely more on means to enhance relationships, such as by building trust. According to this way of thinking, stipulations that parties “shall do this” or “shall complete that” can cast a negative tone over the relationship and even encourage bad behavior. Informal mechanisms can prevent opportunistic behavior (Bertelli and Smith, 2010) and reduce contracting costs for activities such as contract monitoring and third-party enforcement through the court system (Marvel and Marvel, 2007).

The use of alternative informal mechanisms can also guide the collaboration relationship. This is sometimes referred to as relational contracting. Relational contracting has a basis in law (MacNeil, 1985) and has influenced management generally (Poppo and Zenger, 2002) and public management more specifically (Van Slyke, 2007; Bertelli and Smith, 2010). In relational contracting, factors such as trust, reputation, and shared values help to build social capital among parties (Amirkhanyan, Kim, and Lambright, 2012; Feiock and Andrew, 2006; Van Slyke, 2007; Fernandez, 2009). Applying a relational contract framework, Curnin and O'Hara (2019) explore interorganizational collaboration in the nonprofit and public sectors during disaster recovery efforts. Based on interviews with practitioners involved in the recovery following a flooding event, they find that collaboration is reliant not only on established interorganizational structures but also on trusting relationships.

Of all the attention given to informal mechanisms, trust is perhaps the most frequently discussed in the literature. The evidence suggests prior interactions can create trust (Das and Teng, 1998; Bryson et al., 2006; Vangen and Huxham, 2005; Stone et al., 2010), and trust can be built by sharing information and by demonstrating competency through reaching smaller or intermediate goals (Vangen and Huxham, 2005). Small acts of cooperation above and beyond what is explicitly stipulated in a written agreement can encourage goodwill, called tit-for-tat behavior (Brown, Potoski, and Van Slyke, 2016). On the other hand, trust is easily lost when parties perceive others' behavior as uncooperative or unfair. And when one party ceases cooperation in one form or another, other parties are prone to react with their own retaliatory behavior, called shading. The act of one party taking advantage and the other parties reacting is also a form of tit-for-tat.

Several scholars consider the potential interaction between formal mechanisms and relational governance mechanisms (Gong et al., 2007; Gulati and Nickerson, 2008; Hoetker and Mellewigt, 2009; Poppo and Zenger, 2002). After identifying different dimensions of trust, Dyer and Singh (1998) find a positive association between relational mechanisms of network governance and trust and a negative relationship between formal governance mechanisms and the level of inter organizational trust. Tobias-Miersch (2017) finds that trust can explicitly be used as a substitute to replace conventional controls. In contrast, Poppo and Zenger (2002) provide evidence that formal contracts can complement (rather than substitute) relational governance mechanisms as the complexity of transactions increases. Likewise, Lui and Ngo (2004) provide evidence that the use of formal governance mechanisms can encourage the development of trust in a network because they act as safeguards and reduce risk for those involved. Taking a different approach, Inkpen and Curral (2004) propose a temporal model to follow the evolution of trust and control mechanisms in strategic alliances.

In a multiple case study Romzek et al. (2014) explore the interpersonal interactions within collaborative systems, among subsystems, and among organizations and illuminate the informal mechanisms that facilitate collaboration. The researchers find informal interpersonal dynamics nested in combinations of vertical and horizontal ties with mixed administrative authority arrangements derived from both formal and informal accountability relationships.

Research insights about cognition and interpretation have in turn influenced research on collaboration mechanisms. For example, Gray et al. (2015) describe how communication frames are used as mechanisms. Frames are schematics of interpretations that enable individuals to locate, perceive, identify, and label what happens in the world around them (Goffman, 1974, 21). The main idea is that issues are framed differently by different actors, which is central to the communication process. Frames are considered a mechanism of communication because they can define which actors are involved, what problems are addressed, how these problems are defined, and what kind of solutions belong to the set of solutions.

Performance

Since there are many reasons parties decide to collaborate as opposed to going it alone, there are many ways to measure performance. Put another way, whether or not a collaboration can be considered successful depends on the partners' initial goals. The vast literature on collaboration performance suggests there is no simple answer about how to successfully manage collaborative public networks or on how to solve so-called wicked problems (Cristofoli, Meneguzzo, and Riccucci, 2017). The same holds true for public–private partnerships (P3s), which are widely considered a form of collaboration.⁴ Carston Greve and Graeme Hodge, in their extensive work on public private partnerships (P3s), discuss the challenges with evaluating the performance of P3s. Acknowledging the many different motivations for P3s, they analyze both extremes: P3s considered successful and those considered failures. Hodge and Greve (2007) call for strengthening future evaluations and to conduct such assessments away from the “policy cheerleaders” (2007, p. 545; see also, Greve and Hodge, 2007; Greve, 2015; Hodge, Greve, and Biyguante, 2018).

Case study research on a work release program in Illinois highlights some of the typical challenges in determining and measuring collaboration performance. Yung et al. (2016) compared different approaches to program delivery: direct provision by government and a collaborative contract between government and Safer Foundation, a nonprofit organization. The two approaches were implemented with similar specifications in different parts of the city of Chicago. The Illinois Department of Corrections was motivated to collaborate because Safer Foundation had connections within the community to assist prisoners in job placement. The explicit goals of the collaboration included reducing costs, increasing program completion rates, and increasing job placements. The researchers found improvements in completion rates as well as significantly better employment and earnings associated with the government-nonprofit collaboration. However, it was not at all clear that the collaboration reduced costs, as compared to direct provision by government. In implementing the program, the government organizations became so intertwined with Safer Foundation that it was impossible to accurately assign costs. For example, key personnel who were on the government's payroll performed many of the management functions associated with the collaboration, working side by side with management of Safer Foundation. This type of symbiotic relationship is more common than uncommon in collaborations involving the public sector, making cost comparisons virtually impossible. The researchers also evaluated factors associated with the collaboration that are theoretically reasoned to yield positive performance outcomes. They found evidence of mission alignment, sufficiency of resources, and expertise of the nonprofit organization. However, there was no way to determine the relative importance of each factor to the program outcomes. In addition, the researchers found a number of problems associated with the collaboration that caused tension in the relationship, such as ambiguity in the contract document, a lack of operational transparency on the part of Safer Foundation, and a depletion of government expertise. In the end, this case underscores claims researchers have made regarding collaboration success. That is, success is almost always difficult to verify, and subject to differing interpretations and to controversies (e.g., Hodge and Greve, 2007). Going one step further, establishing success through metrics can even undermine collaboration outcomes (Harris and Tayler, 2019; Muller, 2018).

In spite of the complications, however, research on collaboration identifies a number of key factors for collaboration success, with the role of the manager primary among them. According to Davis (2016), the manager is key to mobilizing network participants to perform collaborative work. And Obstefeld, Borgatti, and Davis (2013) propose that the ability to access resources depends on the manager's active involvement and influence in their organization because necessary contacts may be deep inside organizational networks and distant from the locus of action. Research by Van Slyke and Hammonds (2003, p. 146) finds that public management capacity increased as a result of privatization; one can interpret this finding as consistent with calls for treating the manager as significant in collaboration success.

The manager's experience can play a very important role in success. For example, Sytch, Wohlgezogen, and Zajac (2018) find that experience in intraorganizational collaboration may lead to success in complex interorganizational settings. They find that matrix firms are more likely than nonmatrix firms to enter into complex alliances. (See Chapter 8 for discussion of matrix organizational structures.) They reason that managers of matrix firms have greater familiarity with coordination, knowledge sharing, and conflict management challenges in intraorganizational context, which in turn gives them greater confidence in their ability to manage similar challenges in complex alliances.

The literature identifies a wide range of factors that either enhance the prospects of success or threaten success in collaborations. The importance of relationships is frequently emphasized. For example, Bryson, Crosby, and Stone (2006, p. 52) suggest that cross-sector collaborations are likely to be more successful if built on strong relationships with key political and professional constituencies, while Coleman (1990) describes how common linkages can strengthen ties. Looking at a different dimension of relationships, Baker, Gibbons, and Murphy (2002, p. 39) claim it's the potential for future relationships that matter. Specifically, they proffer that informal agreements are sustained by value of future agreements.

The literature also acknowledges the role of partners' rivalries (Park and Ungnson, 2001) and partners' reasons for friction and withdrawal (Greve, Baum, Mitsuhashi, and Rowley, 2010). Collaboration success may therefore be a function of how well interdependent network participants successfully navigate and balance cooperation and competition over time (Hannah and Eisenhardt, 2017). Boyne (2003) observes that more partners or a larger network mean more potential for conflict. Likewise, Beckman et al. (2007) find that multi-partner alliances have disadvantages over dyads because multiple parties increase likelihood for conflict and disagreement; they call the phenomenon relational pluralism. Heidi, Steensma, and Phelps (2014) argue that different partner members will form divisive fault lines, while Sytch and Tatarynowicz (2014) suggest that collaborations lead to subgroups, which in turn isolate some of the initial members from the larger group.

The institutional logic perspective provides a framework for analyzing the interrelationships among organizations and institutions in social systems. This developing literature offers an opportunity to advance our understanding of collaboration relationships. The main idea is that different types of organizations are said to have their own “institutional logic,” which may drive conflict and create challenges for work coordination. Institutional logic is the socially constructed historical pattern of cultural symbols and material practices, assumptions, values, and beliefs by which organizations produce and reproduce their material substance, organize time and space, and provide meaning to their daily activity, all of which complicate coordination across organizational boundaries. For example, hospitals operate according to their own logic as do research institutes, schools, and governance boards. Competing logics mean more challenges for managing relationships, more potential for misunderstandings, and more potential for conflict. The literature includes other terms that refer to the same dynamic. Similarly, Jarzabkowski et al. (2009) notes that “institutional pluralism” can occur at the organizational level, and Kratatz and Bock (2008, p. 244) refer to “multiple logics” that affect coordination of hybrid organizations. On the other hand, organizations with umbrella affiliates may perform better compared with their freestanding counterparts because of the benefits they receive from standardization and economies of scale (Amirkhanyan, Kim, and Lambright, 2008).

Simmel (1955) refers to a great variety of roles and idiosyncratic approaches that collaborative activity can involve. Similarly, recent literature increasingly emphasizes the challenges inherent in collaborations of diverse partners (e.g., Gray and Purdy, 2018). In a similar vein, Seibel (2015) describes collaborative arrangements that are comprised of different formal or informal institutional arrangements owing to overlapping sectoral segments and combinations of governance mechanisms (p. 697). Organizations from different sectors can have conflicting core values (Nicholls and Huybrechts, 2016), and distinct values held by subgroups can create conflict.

Following this line of thought, Bryson et al. (2006) discuss the degree to which members of a collaboration embrace market-based logic, state-based logic, and democracy-based logic as essential conditions to the design and implementation of cross-sector collaboration (p. 50). Likewise, Herranz (2008) uses community-based, entrepreneurial-based, and bureaucratic-based networks to describe different sector-based values and strategic orientations in multi-sector networks.

Of course, as the seminal work of Thompson (1967) reminds us, even the most compatible partners may face challenges in coordination. Obtaining partner cooperation becomes more challenging as partners become more interdependent in performing tasks (Thompson, 1967; Gulati and Singh, 1998). In their case study of four organizations, Stadtler and Karakulak (2020) observe the use of broker organizations to facilitate collaboration. Surprisingly, they found that not all broker organizations improve coordination; specifically, two of the four broker organizations weakened the collaboration by gradually replacing the partners' cross-sector tasks and decision making with unilateral, broker-based ones. This study reveals the processes underlying brokers' role drift and unintended collaborative weakening and those allowing them to maintain their facilitation role. In sum, this study illuminates the boundaries of using broker organizations as a mechanism to facilitate cross-sector collaboration.

This section reviewed the literature on collaborative governance, making certain points clear. First, collaborative activities have become prominent across the globe and in all sectors, resulting in a “stunning evolutionary change in institutional forms of governance” (Alter and Hage, 1993, p. 12). Collaborations are motivated mainly by the need for resources and the need to tackle complex societal problems. Collaborations are generally unstable and thus require proactive planning and attention to mechanisms to foster accountability, reduce conflict, and enhance coordination. Whether collaboration is successful is hard to pin down but the literature offers a range of suggestions and considerations likely to facilitate desirable outcomes.

Ending this section calls for underlining a point made in the beginning of the chapter – a point emphasized by prominent scholars (Hefetz and Warner, 2004, pp. 171–174; Emerson and Nabatchi, 2015). That is, collaborative governance properly considered in the context of government includes a broader conceptualization of performance for public organizations, one that features public values, mutual understanding, and relational contracts, as well as the need to achieve effective coordination and cooperation. Collaborative governance does not dismiss the value of public and private organizations working together, but it does feature the role and importance of public managers in solving complex social problems – in contrast to the foci and emphasis of NPM models, which often assume the superiority of the private sector and its practices. In addition, collaboration in the public sector requires attention to an array of policies, rules, and procedures, many of which are embedded and specified in formal contract documents. The next section discusses what we have learned about optimizing these formal contracts and about a new approach called relational contracts that has garnered recent scholarly attention.

From Formal Conventional (Transactional) Contracts to Formal Relational Contracts

There is no shortage of sources on best practices for government contracts, which are by nature formal and transactional. In 2014, the O'Neill School of Environmental Affairs at Indiana University put together an expert panel to advise state and local governments on government outsourcing. The panel was co-chaired by Stephen Goldsmith, a professor of Harvard University and former mayor of Indianapolis, and Ed Rendell, former mayor of Philadelphia and former governor of Pennsylvania. Well-known scholars on the project included David van Slyke, Jocelyn Johnston, Mildred Warner, Graeme Hodge, Deanna Malatesta, and Sergio Fernandez. The panel pulled together lessons from prominent outsourcing cases that covered the range of activities and involved different government jurisdictions. The panel concluded with a report, offering nine different recommendations supported by on-the-ground experiences and recommending scholarly literature on each recommendation. We were inspired to build upon and update their recommendations by detailing the most overlooked considerations for successful outsourcing. Table 14.1 includes six broad considerations: low cost/ efficiency, performance assessment, desired performance outcomes, enhancing competition via contract specifications, outsourcing's effects on government employees and partner selection. The table explains and advises on each consideration. It also provides theoretical and empirical sources for students who wish to dive deeper into the topic.

Importantly, when public organizations enter into a contract with an external party, certain contract provisions are required. The specific provisions required depend on factors such as the jurisdiction, contract amount, and service type. For example, if the contract amount is over $100,000, federal policies require the formal agreement to include an anti-lobbying certification. The language for the certification is dictated by federal code (44 CFR Part 18). If the contract involves legal services, the contractor is required to notify the government if the attorney of record is disbarred (called a Suspension and Debarment clause). Federal and state-level contracts include a “termination for convenience clause,” providing the government with the right to terminate the contract without cause. These are standard boilerplate provisions that facilitate a degree of necessary accountability but also set a tone for the relationship.

As Frydlinger et al. (2019) explain, it is not uncommon for conventional contracts between two organizations to exceed one hundred pages that detail obligations and outline dozens of metrics to measure performance. Government contracts can be even more onerous given the additional boilerplate clauses that the agencies are required to include. Further, contracts filled with “supplier shall” statements can undermine trust and confidence in the relationship (p. 119). According to this way of thinking, these contracts are traditionally used to protect against the possibility that one party will extract benefits and therefore needs to be held accountable. The formalities and language embodied in these contracts create an adversarial mindset and can lead to a downward spiral of negative tit-for-tat behaviors. Frydlinger et al. (2019) discuss how such a contract undermined the relationship between Federal Express and Dell Computers, bringing them to the breaking point.

What's needed, these scholars argue, is an alternative approach that provides flexibility and trust to accommodate the need for adaptability and cope with uncertainty (Frydlinger et al., 2019, p. 118). The alternative approach is called a formal relational contract. Formal relational contracts specify mutual goals and establish governance structures to keep the parties' expectations and interest aligned over the long term; they are designed to foster trust and collaboration, yet they are also legally enforceable. Formal relational contracts are especially useful for highly complex relationships where parties are interdependent and future events are highly unpredictable.

Frydlinger et al. (2019) outline steps to put formal relational contracts into practice. Although their instructions are intended to guide large for-profit companies, they are just as relevant to government (see Table 14.2). Formal relational contracts do not replace traditional transactional contracts, but they should be part of the contracting toolkit to govern highly complex relationships and demand collaboration and flexibility (p. 125). In the case of public organizations involved in collaborations, including outsourcing, public officials need not (and cannot) abandon the procedures and substantive clauses required by oversight agencies. However, they can supplement the requirements while attenuating the negative tone associated with “supplier shall” clauses. In sum, public organizations can approximate formal relational contracts in undertaking collaborations.