The Current State of Interorganizational Collaboration: Lessons for Human Service Research and Management

The research on interorganizational and intersectoral collaboration in recent years has proliferated, offering an increasing number of lessons learned for both the study and management of human service organizations. Organizational collaboration describes dynamic relationships involving coordinated activity based on mutual goals (Gray, 1989). While collaborative activity in the nonprofit sector has been studied empirically since at least the 1970s, the increasing pace of government privatization has led to a rise in scholarly interest in collaboration beginning in the 1990s. Since practitioner interest in this crucial aspect of management is also on the rise, scholarly efforts continue steadily to the present. Thus, the worlds of academia and practice share a strong mutual desire to understand how successful partnerships can be fostered.

My colleague, Chao Guo (University of Pennsylvania School of Social Policy and Practice), and I have produced the first comprehensive synthesis of this literature. Our methodological strategy was systematic and inclusive, to avoid the biased (e.g., discipline-specific or journal-specific) reviews that unfortunately permeate our field (Gazley & Guo, 2015). Our analysis employed four selection criteria: We included peer-reviewed English-language articles reporting on an empirical (e.g., qualitative, quantitative, or mixed methods) analysis, where the analysis included nonprofit organizations and focused on collaborative activity. This editorial includes highlights of what we found along with implications for future research and practice.

Collaboration is complex

Collaboration is indeed complex, and perhaps that is an obvious statement. But it is worth repeating that the academic study of collaborative organizational activity in any field—human services or others—still reflects efforts to simplify an extraordinarily complex set of behaviors. Some simplification is necessary when modeling complex behavior. But flattening the context for collaboration too far can leave out key dynamics that are not translated to practice. For example, a scholar trained in institutional theories who examines collaborative activity as an entirely organizational phenomenon may overlook the obvious fact that the human beings who lead organizations control their decision-making authority. That scholar might consequently overlook the ways in which an organization’s employees foster or impede collaborative activity. This is also why the older, unidimensional approaches to characterizing collaborative behavior (for instance, as a continuum of increasingly integrated activity—competition, contracting, cooperation, alliances, mergers) are unlikely to capture enough of its multidimensional complexity.

Collaboration occurs on many levels although it is not being studied on many levels

The complexity occurs when collaborative behaviors are manifesting themselves on multiple levels and points in time. Collaboration is not only a human activity, it is also a team and group dynamic. It occurs within and across institutions but also within the partnerships and networks they inhabit. The decision to cooperate with others is also both cognitive and psychological—we trust or we do not trust others to share resources or information—and research suggests that our trust levels are likely to be shaped by prior experiences (Gazley, 2008; Huxham & Vangen, 2005; Thomson & Perry, 2006). Moreover, time shapes collaborative activity—what we observe at any point in time is bound to change, and the adaptive characteristics of collaborative systems may in fact be the most important element in long-term sustainability of a partnership. In other words, it is impossible for scholars to capture all collaborative dynamics and very possible that they are missing key information when observing them via one source of information or at a single point in time.

Yet in our review of more than 540 peer-reviewed journal articles published during the past 40 years, Professor Guo and I found that 60% of the studies relied exclusively on one unit of analysis (individuals, organizations, or partnerships), while nearly all of the remainder engaged just two of these units of analysis and very few used all three together. In addition, only 11% of the studies employed a mixed methods analysis: 22% were quantitative and the remaining two thirds relied on some kind of qualitative analysis (Gazley & Guo, 2015). I should note that we express no preference for one method over another, since all contribute in making the field of collaboration research as rich as it is. And as a scholar of collaborative behavior, I too have been challenged by limited data.

We do conclude, however, that the literature we are producing is unbalanced in at least three additional ways: (1) while the field is generating a rich trove of case studies and finely grained narratives of collaborative experiences, this kind of data is not easily generalizable to broader contexts, (2) authors from different disciplines are not taking each other’s work into account, and (3) very little research features the study of partnerships over time in order to understand how they evolve and adapt (Gray, 1989).

And there are other useful ways to understand this complex landscape. Scholars have promoted inclusive theoretical frameworks that “allow us to develop insights [about organizational relationships] at different layers of visibility and interpretability” (Daellenbach, Davies, & Ashill, 2006, p. 74). Not all theoretical frameworks comfortably coexist and some by nature or design contrast with others. Nonetheless, we tend to see among scholars of collaboration an increasing receptivity to a more integrated and interdisciplinary framework—and often a greater effort to incorporate multiple units of analysis in their empirical work (Guo & Acar, 2005; Vigoda-Gadot, 2003). For example, the study by Kumar, Kant, and Amburgey (2007) examines managerial attitudes toward collaboration as a function of socialization factors, personality traits, organizational culture and other job-related factors, and external environmental events such as political pressure. The fact that this study addresses Indian forest management rather than human services offers evidence that an interdisciplinary exchange of ideas has merit. Closer to home, Alexander’s (2000) longitudinal study of human service collaboration in one Ohio county employed a range of theories to determine that the impetus to collaborate not only is both internally and externally motivated but also includes strategies that evolve over time.

Indeed, Alexander’s reliance on sociological, institutional, and economic theories was mirrored in our broad literature review, where these theories also dominated (e.g., resource dependence and network theories). In fact, scholars can be tempted to view a combination of such theories as an indicator of the study’s “diversity” (see, e.g., Bryson, Crosby, & Stone, 2015). But we have much further to go and more to learn about the attitudes and characteristics of the people behind partnerships. Our synthesis found some fascinating studies that investigate collaboration through the lenses of “conscious psychological type,” “theories of personality” and “gender,” and theories of “stigma” and “fairness” alongside more familiar “role” theories. For example, Goldman and Kahnweiler (2000) use popular personality tests to determine if, in addition to describing personality types, they can also be used to identify “collaborative” profiles. And a recent article by Conner et al. (2015) uses “cultural theory” to test public managers’ motivations to collaborate, an approach that suggests that attitudes about cultural norms such as egalitarianism and authoritarianism also create collaborative predispositions. Overall, more than one third of the entire scholarship on nonprofit intersectoral and cross-sectoral collaboration involved theories that do not include institutional dynamics and thus are less familiar to the management, economics, and policy disciplines.

These disciplines include, notably, systems theories such as population ecology that serve an important role in emphasizing that interorganizational collaboration occurs within a complex and multilayered political, organizational, natural, and social system. The natural sciences, in fact, produce considerable literature on institutional dynamics. Take, for example, Jacobson and Robertson’s (2012) explanation of the role of nonprofit associations in creating “adaptive co-governance frameworks.” They describe successful partnerships as flexible social networks that can be well suited to managing dynamic natural ecosystems. As human service managers know, the human service landscape can also change rapidly. Again, more effort is needed to learn from one another across disciplinary boundaries.

Success depends on many factors

With those points made about the breadth of the research so far, we still have a pretty good idea about what human, organizational, and partnership dynamics matter in successful collaborations. At the human level, as noted, certain traits and experiences (collaborative and other) motivate decision makers to pursue interorganizational partnerships or, occasionally, to avoid them when a past experience was unsuccessful (Foster & Meinhard, 2002; Gazley & Brudney, 2007). At the partnership level, political theories suggest that ideology can motivate or demotivate public officials to collaborate with nonprofits. Economic theories also play a role in observing that collaborators can be motivated by the prospect of greater efficiencies or the opportunity to reduce environmental uncertainties (Gazley, 2008). To sustain a successful partnership, we also know that collaborators must share complementary (although not necessarily the same) goals but that the hard effort of aligning goals pays off in the end in better services (Gazley & Brudney, 2007; Selden, Sowa, & Sandfort, 2006). In other words, partnerships involve shared risk, flexibility, and much more than superficial, “inch-deep and mile-wide” endeavors.

At the organizational level, capacity considerations often drive collaborative decisions, with organizational size taking on more importance than the age of the organization (Grønbjerg & Child, 2004). Based on this research, it is possible that a critical capacity threshold must be met before organizations have the time, staff, or experience to pursue partnerships. This finding raises obvious concerns about how smaller human service organizations can be supported in developing healthy interorganizational partnerships. I recall a personal experience when a large human service organization (where I served on the board) was in a position to invite a smaller organization to collaborate on an IT project, with cost-effectiveness as the mutual goal. I was advised (although I did not entirely agree) to wait for the smaller organization to make the first move, lest our offer be viewed as a hostile takeover. In an era of mergers and service integration, what can the literature tell us about how smaller human service organizations can successfully navigate this territory? Unfortunately most researchers often leave out these smaller organizations because their data may be missing from IRS 990 filings and other public records.

Intersectoral collaborations—especially government-nonprofit partnerships—are also of great interest to the human service field. Here, research is generating the crucial argument that partnerships can put enormous financial pressures on human service organizations. An important national study of human service organizations conducted by the Urban Institute and authored by Boris et al. (2010) reinforced this conclusion. They paint a bleak landscape of delayed government payments and eroded services where providers could not be criticized for wondering if the mutual goal of a social safety net has been abandoned by state governments, hoping the private sector will shoulder its responsibilities. This research also helps to reinforce the idea that collaborative activity is not necessarily a win-win proposition. Success depends on resources, patience, and hard work.

Where do we go next?

The research on collaboration is richer than most nonprofit scholars may realize, especially with respect to the diversity of contexts, geography, and collaborative forms that have been studied around the globe. Certainly the potential for greater cross-national and cross-contextual comparative research is there. But even within the United States, the heavy emphasis on studies of regional social and human service networks suggests an opportunity for more cross-regional and broader comparisons to integrate findings and identify key drivers of collaborative success (Gazley & Guo, 2015).

While the research we analyzed improved over time in terms of its sophistication and its willingness to take on collaboration’s complexities, we still found a discouraging number of unexplained selection choices, overuse of convenience samples, incomplete models, weak data, and missing tests for bias or robustness. But the emerging patterns do define possible future research directions for nonprofit scholars. Questions worth greater exploration include the following: How are conflicts among partners avoided or managed? How are imbalanced power dynamics handled? How can we train and prepare organizational leaders most effectively to succeed in highly collaborative environments?

Scholars also seem to know more about the “why” and “what” questions related to collaboration than the “how” question. In other words, we are generating a good stream of research on antecedents and outcomes but need more on the internal processes of partnership activity (Gazley & Guo, 2015). Vigoda-Gadot (2003), for example, has written about the importance of developing common social norms within partnerships, but how does that occur? To understand internal processes, more mixed methods studies and comparative case studies of collaborative service delivery, observed over time, would serve us well. The comparative element (comparing different service fields, funding contexts, geographic scope) adds the opportunity to understand variation in collaborative processes. The lack of emphasis on processes is a big knowledge gap from a practitioner’s point of view as anyone tasked with building and strengthening partnerships understands. In fact, given the problem of survivorship bias in research generally (e.g., studying the successful cases because they still exist to be studied), there are also many unanswered questions about why partnerships fail.

Collaborative success in the field takes patience; scholars must exhibit an equal level of patience to observe partnerships over time, where perhaps the most important lessons emerge. Indeed, we owe the practitioners who seek to use our research the opportunity to get the most from what we produce.