[Added August 27, 2014: GiveWell Labs is now known as the Open Philanthropy Project.]

Something I think about a lot is the spectrum from “passive funding” to “active funding.” By “passive funding,” I mean a dynamic in which the funder’s role is to review others’ proposals/ideas/arguments and pick which to fund, and by “active funding,” I mean a dynamic in which the funder’s role is to participate in – or lead – the development of a strategy, and find partners to “implement” it. Active funders, in other words, are participating at some level in “management” of partner organizations, whereas passive funders are merely choosing between plans that other nonprofits have already come up with.

My instinct is generally to try the most “passive” approach that’s feasible. Broadly speaking, it seems that a good partner organization will generally know their field and environment better than we do and therefore be best positioned to design strategy; in addition, I’d expect a project to go better when its implementer has fully bought into the plan as opposed to carrying out what the funder wants. However, (a) this philosophy seems to contrast heavily with how most existing major funders operate; (b) I’ve seen multiple reasons to believe the “active” approach may have more relative merits than we had originally anticipated. This post discusses our observations on this front, and the implications. Note that Good Ventures has played a major role in facilitating and participating in conversations with other major funders, which is where most of our learning on this front comes from.

Brief summary:

• Major funders very frequently take a highly active approach to funding, in some cases going so far as to play a major role in creating partner organizations.
• In the nonprofit world of today, it seems to us that funder interests are major drivers of which ideas that get proposed and fleshed out, and therefore, as a funder, it’s important to express interests rather than trying to be fully “passive.”
• While we still wish to err on the side of being as “passive” as possible, we are recognizing the importance of clearly articulating our values/strategy, and also recognizing that an area can be underfunded even if we can’t easily find shovel-ready funding opportunities in it.

Major funders appear to be highly activeSince we launched GiveWell Labs, we’ve spent a fair amount of time talking to major funders. Unfortunately, in many cases we’ve been able to have these conversations only under condition of confidentiality, but we’ve posted notes from conversations when possible and we can share examples of patterns we’ve observed.

Perhaps the most vivid example of an “active” funder is the Bill and Melinda Gates Foundation, which we also perceive as one of the most generally impressive foundations we’ve interacted with.

• The Gates Foundation has been integral in the creation of many of the organizations it has made major grants to, such as GAVI (which has received several of its largest grants), Alliance for a Green Revolution in Africa, Innovative Vector Control Consortium, Foundation for Innovative New Diagnostics, and PATH Malaria Vaccine Initiative.
• On polio eradication, it describes itself as providing “technical and financial” (not just financial) resources.
• In a conversation with Cynthia Lewis of the Gates Foundation’s Tobacco Initiative, Cynthia emphasized that much of the Foundation’s work is “hands-on,” and also stressed that when going into a new area it may be necessary for a grantmaker to “create” the partner organizations it needs to fulfill specific objectives. When we asked her how one might go about this, she gave an example of an anti-tobacco organization whose team had been largely recruited by a program officer at another foundation. (At the same time, she characterized much of her own work as providing more “hands-off” funding to particularly strong organizations, indicating that there’s a time and place for both styles of grantmaking.) We will soon be publishing conversation notes from another conversation on this specific topic, which will give many more examples of cases in which various foundations were instrumental in the creation of partner organizations.

Other examples of “active funding” can be seen in our notes from discussing drug policy with Open Society Foundation representatives and in CIFF’s description of its approach to funding. An Atlantic Philanthropies document on political philanthropy states, “Once a grantmaker decides to support an advocacy campaign, there are several options for how to manage advocacy-oriented funding. Determine early on whether there is 1) a grantee, or coalition of grantees, that is already actively involved in a campaign or can easily take leadership; 2) a third party group that should be contracted to act as the manager for a new campaign initiated by the funder(s); or 3) the campaign would best be managed directly by foundation staff. Each approach has its own benefits and risks depending upon the issue and the funder(s).” In all three cases, it is clear that foundation staff sometimes intend to restrict funds and/or manage projects quite actively rather than simply providing funding.

Limitations of a passive approachWhen we first started work on GiveWell Labs, we contemplated a highly passive approach in which we would ask major charities – and major foundations – for the “best projects that aren’t already funded.” This approach seemed to face many obstacles. When speaking with major charities, we found that:

• Some groups (such as GAVI and PATH) emphasized the need for unrestricted funds to respond to opportunities as they came up, rather than pointing to specific un-funded opportunities.
• Other groups (such as GAIN and Micronutrient Initiative) listed many possible projects, at a very broad level, and asked us to direct them to the ones we were most interested in.
• In all cases, it’s been a consistent fact that there are no existing detailed writeups on the ideas ready for us to review. Rather, it seems to consistently be the case that detailed writeups on potential projects are developed for (and in collaboration with) potential funders. (Every project proposal we’ve ever seen from a large organization has been for a specific funder).

When speaking with major funders, we were often presented with more specific project proposals, about which more written information was available. However, in these cases, it wasn’t always clear to us that the projects we were looking at needed more funding to go forward. In the case of the PSI project that Good Ventures co-funded with the Gates Foundation, we initially believed there was a concrete funding gap and later changed our view. In the case of the Wellcome Trust, it was made explicit to us that there are rarely specific projects that are outlined and ready to go but not funded.

Without detailed writeups, and without a good deal of context, it’s very difficult to assess a potential project. Asking for a detailed writeup would be, to some extent, indicating and committing interest in the potential project. Because of this dynamic, it’s very difficult to a purely “passive” funder: one needs a sense of which sorts of projects one is most interested in.

This may also help explain the shortcomings of the “maximally passive” approach we took in 2007. (See our July 2012 blog post discussing this.) We had hoped that many charities would fully make the case for their impact to us, and that we would simply choose the best case. But when writeups are usually done for – and in collaboration with – specific funders, this sort of approach isn’t possible, and it’s necessary to have an idea of one’s criteria in order to even decide what to evaluate.

Another approach to “passive” funding is to focus on smaller, startup-like nonprofits rather than large charities and foundations, as Ashoka does. However, even when seeking out groups like this, we see possible advantages to being active:

• There are already multiple groups that take a systematic, “sector-agnostic” approach to funding smaller and more startup-like groups. (Ashoka, Draper Richards Kaplan, Echoing Green; Skoll Foundation at a later developmental stage.) We’re not sure whether we’d have substantial value-added over such groups if we focused on taking a similar approach.
• We’ve seen multiple cases in which a funder’s “active” strategy was crucial in finding – or even creating – a “startup-like” nonprofit. Some of these involved interactions with other funders from which we have not produced public information, but we can discuss cases that involved us directly. These are intended to give examples of how one’s active strategy can be crucial in becoming aware of (or even contributing to the creation of) a “startup-like” nonprofit; we don’t intend these examples to be interpreted as groups we’re actively considering funding.
  • In one case, we spoke to a professor about the Cochrane Collaboration, and after he saw our post on meta-research, he asked whether we’d be interested in a proposal for the creation of a meta-research-oriented group. He explained to us that he had wanted to write up such a proposal for a long time, but had not done so previously, because he couldn’t identify any funders that would likely be interested.
  • Dario Amodei has also been influenced by our discussion of meta-research. As an academic himself, he has considered trying to get involved in improving the system of academia. Our interest in this area has helped give shape to his interests, and has played a role in the creation of Vannevar, a group of scientists dedicated to making scientific research more efficient, collaborative and productive. According to Dario, the knowledge that a funder might be interested in his collected observations on academia – even though the interest was very preliminary and did not come with any funding commitment – has spurred him to conceptualize, and follow through on, a strategy focused on producing the sort of information such a funder might find valuable.

I think the above observations point to a potential disanalogy between the for-profit and nonprofit worlds. In the for-profit world, a business gets capital from investors but ultimately gets its revenue from customers (and if it can’t raise capital, it can often “bootstrap” to the point where it has something more impressive to show funders). In the nonprofit world, funders provide both the capital needed to get started and the ultimate verdict on whether a project succeeded. Thus, in the nonprofit world, it seems that the “what projects are people trying to start?” question and the “what projects do people think major funders will be interested in?” question are tightly integrated. Expressing clear interests (as we did with our meta-research post) can cause new opportunities to emerge; simply looking for “shovel-ready” projects may severely limit a funder’s options.

Another possibility that has occurred to me is that the role of “strategy development” may be structurally more common at foundations/grantors than at charities/grantees. Since foundations ultimately decide which strategies get funded, people who are oriented toward developing strategies may gravitate toward foundations.

ImplicationsIn the nonprofit world of today, it seems to us that funder interests are major drivers of which ideas get proposed and fleshed out, and therefore, as a funder, it’s important to express interests rather than trying to be fully “passive.”

It’s possible, and perhaps desirable, that the nonprofit world of the future will work differently. If funders had greater commitments to transparency, and posted the proposals they chose not to fund, it’s possible that a database of “shovel-ready” proposals could emerge. We certainly intend to share the proposals we don’t fund, when possible, and to encourage others to do the same. But in the world as it exists, it seems important to recognize that articulating preferences and areas of interest is an important thing for a funder to do.

Partly for this reason, we’ve largely de-emphasized the kind of thinking we were doing in 2011 about “sector-agnostic” giving (looking across all issue areas for the most outstanding projects), in favor of the related idea of strategic cause selection. In strategically choosing causes, we want to place some weight on the criterion of room for more funding, and therefore we’d conceptually like to choose causes based partly on what projects are available to fund within them. However, we don’t believe that looking at the current set of “shovel-ready” projects is a reliable way to assess conceptual “room for more funding,” and we believe it may sometimes be fruitful and necessary to express interest in a cause despite not having a good sense of what projects are available within it. We will have to find other – and perhaps rougher – ways of estimating which causes are “overfunded” as opposed to “underfunded.”