The Collaboration Game: Solving the Puzzle of Nonprofit Partnership

Nonprofit collaboration is difficult; economics, game theory, and behavioral science offer lessons on how to do it well.
By Jacob Harold
Above the desks of many nonprofit executives you will find these words: “If you want to travel quickly, go alone. If you want to travel far, go together.” And yet, nonprofits often act in isolation against complex problems, spinning reinvented wheels.
Nonprofit collaboration is harder than it sounds.  Economics, game theory, and behavioral science suggest some lessons on how organizations might collaborate more often and more effectively.
Collaboration can help nonprofits achieve greater social impact at lower cost. To understand how, we can cautiously apply three core concepts of microeconomics: division of labor, economies of scale, and network effects.
1. Division of Labor
Over a century after its introduction in 1913, the assembly line—where each worker specializes in a particular task—remains the quintessential example of division of labor. The resulting efficiencies helped drive the Industrial Revolution, with all its positive and negative consequences.
The nonprofit sector would do well to learn from the factory floor. Functional division of labor among nonprofits, where separate organizations serve the same individual in multiple ways, can improve aggregate impact. One organization may be better at providing housing, another at job training. But for this to work, the organizations must have open channels of communication. Otherwise, the client will suffer—think cross-scheduled sessions, contradictory messages, and misaligned timelines. A related set of benefits can arise from the geographic division of labor by mapping to the distribution of need across space.
A factory works because common procedures, thorough communication, and a shared goal underlie workers’ coordinated efforts. Nonprofits similarly in sync can reap the benefits of organizational diversity and independence while still protecting the interests of beneficiaries.
2. Economies of Scale
An organization that can rent a large office space pays less per square foot than a small one subletting cubicles. A national network can train employees efficiently with standardized trainings. Economies of scale matter for organizations operating on thin financial margins in tough economic times.
But more importantly, economies of scale matter because they can increase social impact. Teach for America’s national scale, for example, allows cross-site learning and improved training for new teachers, all in the service of better educating students in under-resourced public schools. Large advocacy groups like the Sierra Club or the National Rifle Association can mobilize people quickly and cost-effectively due to their immense size.
In some cases, small is indeed beautiful and scale can bring very real challenges. But often collaboration enables economies of scale that help the nonprofit community advance both margin and mission.
3. Network Effects
Network effects happen when a product or service gains additional value with each new user. Take Wikipedia: One contributor can provide a handful of entries, but as an open platform the site gets 10 edits every second. Nonprofits can similarly benefit from connecting individuals or organizations. For example, VolunteerMatch is an online platform that leverages network effects to connect citizens to causes they care about.
Organizations can intentionally build network effects through standardization. An often-cited example is Strive Partnership—a collaboration of nonprofits, foundations, government agencies, and corporations working to improve education outcomes in the Cincinnati area—which serves as an example for the broader “collective impact” movement. The various organizations track the same indicators and engage in regular, structured communications to inform a dynamic, community-wide strategy.
The Internet can generalize and magnify the power of network effects, especially with the use of data standards. In the nonprofit sector, the Basic Registry of Identified Global Entities (a registry of unique identifiers for nonprofits around the world) and the Philanthropy Classification System (a new taxonomy for nonprofits and grants) offer immense potential for field-level alignment—and thus field-level learning.
Properly harvested, network effects help enable nonprofit communities to achieve scale of reach while still celebrating the diversity of the network.
These microeconomic effects can help nonprofits achieve more bang for their buck once they’ve entered into a collaboration, but most organizations face serious structural barriers well before they even get to that point. Nonprofits are not alone in having to manage situations where working with others is difficult. During the Cold War, the US military spent millions hiring mathematicians to parse the dark logic of the nuclear standoff. Thinkers like John Von Neumann built the new discipline of game theory to illuminate the terrible decisions facing political leaders. Nonprofits can similarly benefit from game theory to explain—and potentially avoid—seemingly shortsighted and selfish behavior.
The Prisoner’s Dilemma
Consider first the most famous problem in game theory, the “prisoner’s dilemma”: Two people are held in separate cells and asked to declare the other guilty. If neither defects, each faces a light sentence—30 days. If both defect, they each get a heavy sentence—5 years. If one defects and the other does not, the defector is rewarded with freedom and the other gets the heaviest sentence—10 years. As a pair, the obvious choice is for both to keep mum. But as individuals, the incentive is to defect; no matter what the first prisoner believes the second will do, he sees he will get a better deal through declaring the other guilty.
Potential nonprofit collaborators often find themselves in prisoner’s dilemma-type situations. Nonprofits are usually oriented toward the status quo in terms of funding streams, staffing models, programmatic strategies, and organizational culture. Collaboration requires changes to each, and while they may dream of the benefits of collaboration, nonprofits also know it means risking their relative stability and safety.
For example, imagine two human services organizations: One offers sexual education programs in local high schools and another provides services to teenage mothers. Each receives $100,000 a year from the same local community foundation. Both nonprofits know the foundation sees opportunity for greater impact through greater collaboration; in fact, it might provide $300,000 in total funding for a well-designed collaborative effort. But both nonprofits are nervous about losing their existing funding. If the collaboration does not work out, would the community foundation cut their core funding? Might one organization betray the other and tell the community foundation it could provide both types of services on its own for just $150,000?
To distrust, add uncertainty and short-term costs: Even if the executive directors of two organizations want to collaborate, staff members may resist, saying the partnership will force them to restructure the program, re-do internal systems, abandon their unique culture, or—horror of horrors!—admit the weakness of their organization.
From Two to Many
These dynamics are further complicated when more than two organizations are involved. The mathematician John Nash, famously portrayed in the film A Beautiful Mind, formalized a way to think about multi-party prisoner’s dilemmas—a concept later called the “Nash equilibrium.” When a group of players interacts over time, they are prone to settle into specific behaviors. These behaviors reach equilibrium when no individual wants to unilaterally change given the others’ behaviors.
Environmental advocacy groups offer a good example. They use different approaches to achieve similar goals, or what game theory calls “mixed strategies.” The large environmental groups—including radical ones like Greenpeace and more mainstream ones like Environmental Defense Fund—share very similar goals, but have settled into a pattern of isolated operation, fractured strategies, and even betrayal. These organizations would almost certainly accomplish more together with aligned strategies that either aggregated power or capitalized on difference (think good cop-bad cop strategies). Unfortunately, no single organization has an incentive to unilaterally change; it might threaten funding or compromise identity.
But there are solutions. Research in game theory has shown that a long-term view (the “repeated game”) can completely change players’ decision-making. Intentional communication can remove the knowledge barriers that cause sub-optimal outcomes, as in the regular “Green Group” meetings of large environmental organizations. Every partnership is evidence that it is possible to transcend the dilemmas that get in the way of collaboration.
Collaboration can challenge nonprofit leaders’ sense of identity. We would be wise to respect how emotionally and intellectually difficult it can be for practitioners of social change to acknowledge that they cannot succeed alone. So let us look to behavioral science—and to our own experiences—for insights on how we might find a path to climb Collaboration Mountain.
Define the community. It is essential that collaborators find a sense of shared identity. The first step is to identify what the groups have in common: “We are defenders of tropical rainforests” or “we serve the Miami homeless.” Overwhelming evidence suggests that a sense of community plays a central role in human decisions, whether negative (racism) or positive (neighborliness).
Name the weakness. Introductions at Alcoholics Anonymous (AA) meetings begin with, “My name is X, and I am an alcoholic.” This statement puts the speaker in a humble frame of mind, making them more receptive to help. Similarly, nonprofits can open themselves to collaboration by acknowledging their limitations as independent actors. Naming the problem need not create a pessimistic atmosphere; indeed, the next phrase at an AA meeting is often something along the lines of, “ … and I’ve been sober for five years.” Nonprofits can immediately build on naming the weakness by highlighting the opportunity.
Identify the Sherpa. Multi-lateral collaboration greatly benefits from a guide—a person or organization—to manage the process, provide encouragement, and help solidify a shared vision. Without this, the short-term incentives of the individual participants will tend to prevail, and the collaboration will dissolve. Pure top-down management will not work; if nonprofit participants feel a foundation is forcing them to engage, the partnership will not be authentic. Ideally, the guide comes from the community but is not identified with a single set of interests. Like a Sherpa leading climbers up a mountain, they are both participant and leader, bearing costs and reaping shared rewards.
Make explicit the implicit division of labor. Over time, nonprofits tend to differentiate themselves: One organization takes the west side of a town, another the east; or one handles middle-school students, another high-school students. Once potential collaborators acknowledge those differences, they can capitalize on them. It is often necessary to say explicitly what implicit division of labor may have developed over time. Without that openness, participants may hesitate to speak clearly or act decisively, afraid to offend or stereotype.
Demonstrate proof points. One powerful human tendency is to do what we see others doing. Nonprofits are far more likely to collaborate if they see relatable examples of other successful collaborations. It can be a step-wise process of sharing simple examples, building confidence, bringing in new data, and making larger steps possible.
Design collective systems. Formalized systems for knowledge sharing, governance, and external communications not only create value, but also reinforce collective identity and incentivize organizations to remain in the group. This makes it more likely that the collaboration lasts long enough to achieve the desired change.
This six-step approach is neither linear nor immutable. Some collaborations may skip a step or move through them in a different order, but a trip down this path can name, build, and strengthen new communities of organizations bound together by common purpose.
The challenges the nonprofit sector faces are big and they are complex. Economics reminds us that we need collective action to achieve scale. Game theory reminds us how hard it can be to overcome our narrow interests. But society, again and again, has proven its ability to resolve these dilemmas and shown that we can break these unhealthy equilibria. Surely nonprofits can do the same. The stakes could not be higher.

For any query write to:

Get our Newsletter