Bookmark and Share

Work in Progress

The Hewlett Foundation Blog




Collaboration and Diffuse Reciprocity 

May 1, 2014 — By Larry Kramer

This piece is cross-posted from Stanford Social Innovation Review, where it was published last week. -ed.

It is, I believe, generally recognized that collaboration among funders is important, because none of us has the resources to make serious inroads on big problems by ourselves. Yet despite widespread acknowledgment of its value, I’ve been surprised by how difficult it can be to form fruitful collaborations with other funders. Collaboration happens, but less than it could—even though foundations are unregulated and not competing with each other.

The question is, why?

Many people say the fault lies with foundation presidents and board members, who some think discourage collaboration because of ego or to avoid sharing credit. I haven’t encountered this myself and doubt it as an explanation. More plausible—and this is something I have witnessed—is that presidents and boards can make collaboration difficult by becoming overly directive and inflexible about the specifics of where and how to make grants.

A bigger problem—bigger because it inheres in the structure of most foundations—is the agency cost associated with dispersed decision-making. As a practical matter, collaborating on any particular project or program requires buy-in from a foundation’s program officer, program director, president, and board—each with different degrees of knowledge, different levels of commitment to existing strategies, and different incentives to change. Not surprisingly, aligning so many differences can be challenging.

Ongoing collaboration is hard, for instance, when program officers from different foundations agree but must wait months for their respective boards to sign off. A president who wants to support a request from another foundation may be reluctant to override a more knowledgeable and informed program officer who does not want to veer from an ongoing strategy the program officer devised. A program director may discourage the president or board from collaborating on a particular strategy, even one the relevant program officer supports, if this could affect the allocation of resources among strategies within his or her particular program. And so on.

Funders can overcome these obstacles, as we see in collaborations that already happen. But it’s not easy, and I have increasingly come to believe that overcoming these hurdles—whether by persuading foundations to be more flexible about process or by aligning their internal constituencies—is difficult because there is a dearth of what is known in international relations theory as “diffuse reciprocity.”

Nations cooperate with other nations because there is something in it for them. The actions of each are to some extent contingent on the actions of others, with good returned for good and bad for bad. But international relations scholars draw a further, useful distinction between two forms of reciprocity—namely, “specific” and “diffuse.”

Specific reciprocity is what most of us have in mind when we use the word reciprocity. As explained by author and academic Robert Keohane in his book International Institutions and State Power, it refers to “situations in which specified partners exchange items of equivalent value in a strictly delimited sequence. If any obligations exist, they are clearly specified in terms of rights and duties of particular actors.”

Diffuse reciprocity is different. Keohane elaborates: “The definition of equivalence is less precise, one’s partners may be viewed as a group rather than as particular actors, and the sequence of events is less narrowly bounded.” Actors participate and share, he goes on to say, “not because of ensuing rewards from specific actors, but in the interests of continuing satisfactory overall results for the group of which one is a part, as a whole.”

As Keohane’s description indicates, diffuse reciprocity is more like a cultural norm within a community: I do things with and for others without demanding or expecting an immediate payback or return, knowing that you and others will do the same later and that we’ll all be better off in the long run as a result. Diffuse reciprocity is an attitude, a willingness to give without demanding a precise accounting of equivalent benefits for each action, albeit because others in the community do so as well. It’s why nations respect international waters, for example, and why they make trade agreements that extend unconditional “most favored nation” status.

The benefits of adopting and acting on this attitude can be considerable. As the trade example illustrates, short-term sacrifices may be more than offset by longer-term gains that accrue to everyone from cooperating. Consider what this could mean in the context of foundation grantmaking. Foundations work hard to ensure that grants have maximum impact in accomplishing the foundation’s strategic goals. They are, as a result, reluctant to make grants outside their defined strategies and (often highly) specified theories of change—even if that means losing opportunities to collaborate. But what if, in so doing, they are actually accomplishing less? Is it possible that, in seeking to squeeze maximum efficiency from every dollar this way, we achieve marginally greater short-term progress but sacrifice what economists call “gains from trade” that would leave everyone’s strategies more completely realized in the long run?

Equally important, a regular practice of diffuse reciprocity makes other things easier—including finding more and better instances for specific reciprocal exchanges. This is because few exchanges involve perfect equivalence: There is always some looseness in the joints, some likelihood that one party will benefit less or need to depart from usual practices more than others. The potential obstacle presented by such disparities is reduced by the extent to which partners know they will engage in future exchanges that, over time, will smooth over any one party’s particular advantage or disadvantage. The net result, again, is greater benefits over time for everyone as a result of successfully cooperating.

By way of example, I would cite the experience of the Hewlett, Packard, and Irvine Foundations. We have, in the course of many years, developed a relationship in which we regularly partner to help each other, often involving one foundation asking the others to make grants outside their comfort zone. Doing so has, I believe, made all three organizations more effective, and enabled each of us to make greater progress on projects and programs we care about.

Yet such relationships seem rare—something I am attributing to the absence of diffuse reciprocity in the foundation world, which in turn inhibits collaboration.

But how do I know this? After pointing to some examples of successful collaboration, one of my colleagues (who disagrees with me) wrote: “We know that collaboration requires time, trust, and relationship building. How do you factor that into your argument? How do you contend with what are the real and perceived challenges to collaboration?”

These are good questions and difficult to answer. My sense that there is too little collaboration in the foundation world grows out of my experience in academic administration. Collaboration is less necessary in that setting, yet I found it much easier to accomplish there—even when working with schools that were my competitors and rivals. Reflecting on this experience, it is not that “real and perceived challenges to collaboration” among foundations are fabrications; it is that such challenges loom larger than necessary precisely because of the absence of diffuse reciprocity. We’ve made the idea that it’s difficult for foundations to build trust and develop relationships for collaboration into a self-fulfilling prophecy.

I’ll be the first to admit that I can’t prove this as an empirical matter. To do that, we would need to know what the objectively “right” amount of collaboration is, something impossibly difficult to calculate. Instead, I’ll just ask all you funders out there: How many times has someone asked you to partner on something that was roughly but not perfectly within your strategies, and you’ve said no? How often have you asked someone else to partner with you and they’ve said no? Are you satisfied with the resulting level of collaboration, or are you open to the possibility that we are making it too hard by being too inflexible, requiring too much alignment, and demanding too much process?

That is not to say that everyone should always say yes to every request. Resources are limited, and we can’t do everything. One of the biggest obstacles to growing diffuse reciprocity in the foundation world may be a shortage of discretionary funds, which could enable foundations to test partnerships without having to change or abandon existing programs or strategies. But I think the real problem may be cultural. The current norm among funders feels like a presumptive no—no unless the fit is precise, no unless it’s done our way. Perhaps if we all bend a little more, we’ll all achieve a lot more.