Should the Open Philanthropy Project be Recommending More/Larger Grants?

Throughout the post, “we” refers to GiveWell and Good Ventures, who work as partners on the Open Philanthropy Project.

The Open Philanthropy Project has ambitions of influencing very large amounts of giving in the future (hundreds of millions of dollars a year or more). To date, we haven't made nearly enough recommendations to reach this level of giving, and this is not ideal. In a perfect world, we'd be recommending far more giving.

However, our approach is deliberate: we have chosen to prioritize capacity-building (choosing focus areas and hiring/onboarding program staff, in order to lay the groundwork for future grantmaking) over near-term grantmaking. This post discusses the reasons we've done this so far, as well as outlining our plans for ramping up giving in the future.


  • The case for focusing on grantmaking – why we'd like to be recommending far more giving than we currently are.
  • The case for focusing on capacity building – why we've nonetheless chosen the approach we have.
  • Could we do both? – in theory, it should be possible to ramp up grantmaking while continuing to focus primarily on capacity building, but we haven't gone this route and don't plan on doing so.
  • The plan from here – we expect to ramp up giving significantly (though not to its full eventual level) in 2016. We think there is a good chance that within 5-10 years, we will see more giving opportunities than dollars to fund them.

The case for focusing on grantmaking

Good Ventures hopes to give away several billion dollars over the coming decades, which – when accounting for likely investment returns – would imply hundreds of millions of dollars per year in grants for an extended period of time at peak giving. In 2014, Good Ventures gave ~$15 million to GiveWell's top charities and an additional ~$8 million based on Open Philanthropy Project recommendations. In other words, their current level of giving is nowhere near where they hope it will eventually be.

Furthermore, the Open Philanthropy Project hopes eventually to influence many other major donors, which would mean both (a) that we eventually hope to make recommendations summing to an even higher figure than what the previous paragraph implies; (b) that there is an argument for Good Ventures to front-load its giving, in order to help the Open Philanthropy Project establish itself while many of the donors we hope to influence aren't giving at scale yet.

Thus, we (the Open Philanthropy Project) could be recommending far more/larger grants than we currently are, without significantly lowering the amount available for giving in the future. There are a few arguments for doing so:

  • "Interest rate" on accomplishing good. All else equal, we'd rather accomplish good sooner than later, for reasons discussed previously: helping people today can empower them to help themselves and others, causing a "compounding" of good accomplished. It's unclear how the "interest rate" implied here compares to the return Good Ventures can obtain by investing.
  • Laying the groundwork for future grantmaking. Funding today can help promising organizations and fields grow, which could increase the ease of giving productively in the future.
  • The fact that giving opportunities may worsen over time. We expect greater overall levels of wealth, and lower levels of need, in the future; we also expect more philanthropy, as both wealth and inequality grow. "Room for more funding" in our causes of interest may fall dramatically. This point is especially salient to us because we believe many people today have large fortunes (built from the wave of entrepreneurship over the last ~15 years) that they haven't yet started putting serious effort into giving away.
  • Challenges of peak giving. One could imagine that it would be very difficult to give away hundreds of millions of dollars per year effectively, both in terms of maintaining a large enough staff to investigate opportunities and in terms of simply having enough room for more funding. Every dollar that is "saved" (by not being granted) today is, one could argue, simply adding to a future budget where it is less needed and will do less good. When we feel on the fence about a grant, we could ask: "Is this potential grant worse than the last grant we'll recommend in a future year, when giving is in the hundreds of millions of dollars per year?"

The bottom line is that the amount of giving we're recommending is much lower than we'd ideally like it to be. With a fairly small number of staff, we wouldn't be able to recommend much more at our current level of due diligence, but we could recommend a lot more if we made a concerted effort to do so – for example by lowering our standards for due diligence, or by seeking out people and organizations we could ask to regrant funds. We have had many conversations about whether we ought to move in this direction, and we think some of the above arguments present a strong force in favor of doing so.

At some point in the future, we probably will put a concerted effort into increasing the amount of giving we recommend. But we aren't at that point yet, because we think we have a more pressing priority: building capacity.

The case for focusing on capacity-building

For the last couple of years, we've had no target amount of giving at all. Our goals, and our efforts, have revolved around (a) selecting focus areas; (b) hiring people to lead our work in these areas (see our most recent update); (c) most recently, working intensively with new hires and trial hires on their early proposed grant recommendations.

Collectively, we think of these activities as capacity building. If we succeed, the end result will be an expanded team of people who are (a) working on well-chosen focus areas; (b) invested (justifiably) with a great deal of trust and autonomy; (c) capable of finding many great giving opportunities in the areas they're working on.

Ultimately, we think expanding capacity is the best way to dramatically increase the amount of giving we recommend over the long run, while dramatically improving the degree to which that giving is well-informed. A program staffer we're confident in – meaning they share our values and goals for the cause they're working on, and use reasoning we think is sound to make grant recommendations – could give their full attention to a particular focus area and recommend a large amount of giving, while requiring relatively little oversight. As an early sign that this is the case, Chloe Cockburn – our first cause-specific hire – believes she could fairly easily identify tens of millions of dollars' worth of strong giving opportunities in criminal justice reform. Each of these giving opportunities would likely be based on knowledge that dwarfs my own, yet I'll be able to be confident in these recommendations – despite knowing comparatively little about them – if I am confident in Chloe.

Capacity-building is a project with long time horizons and high stakes. In order to attract the sort of program staffers we've been seeking, we've committed to spend multiple years and many millions of dollars in the focus areas we're recruiting for. If we're too quick to hire and too quick to give trust and autonomy, we could end up working for a long time (and recommending a lot of grants) with someone who's ineffective and/or misaligned with our values. That's why we've been choosing to put a lot of up-front effort into selecting focus areas, running job searches, and working intensively with new hires and trial hires. Our hope is to be very thoughtful and careful with our choices on this front, so that we can give program staffers a great deal of autonomy – and the ability to recommend a lot of giving – down the line.

The reason we haven't put much effort into ramping up the amount of giving we recommend is simply that we've preferred to focus our effort on capacity building.

The grantmaking we have done has largely been done with learning and capacity building in mind, which is why we've tended to err on the side – so far – of deeper due diligence on grant recommendations, even though we know that we will eventually need to invest less time per grant recommendation. This has been especially true of early recommendations from new hires and trial hires.

The main reason we feel good about this approach is because we believe capacity-building will ultimately lead to dramatically more and better giving than we could do today (no matter how we tried to do it). Some additional points in favor of this approach:

  • I think most of the points from the previous section are valid, but perhaps less strong than they appear at first. As discussed below, I think we'll be ready to dramatically increase the amount of giving we recommend (in our existing focus areas) fairly soon. I think our focus on capacity-building will ultimately end up having the "cost" of a few extra years during which we weren't recommending as much giving as much as we could; I don't think the issues of "laying the groundwork for future grantmaking" and "worsening giving opportunities over time" are very significant over this time frame.
  • I think the "challenges of peak giving" point is fairly questionable. I think it's quite plausible that we'll be able to make hundreds of millions of dollars' worth of recommendations in a future year, while having all such recommendations be better and more informed than any "borderline" recommendation we might make today. A couple of points in support of this idea:
    • I don't feel confident that any of our "borderline" giving opportunities today are stronger than simply giving more to GiveDirectly, and GiveDirectly is rapidly growing in scale. At "peak giving," it's quite realistic that we'll still be able to hold all grant recommendations to a standard of "needs to be competitive with GiveDirectly, or else the funds would be better spent via GiveDirectly instead." Under such a picture, I don't think any of the grants we recommend will be much less effective than "borderline" grants would be today.
    • Many major foundations give in the range of what we're projecting for "peak giving," and they don't seem to be resorting to desperate measures to give it away; in fact, most seem (even at several-hundred-million-dollar-per-year budgets) to think of their funding as scarce, to be making what they see as very difficult choices, and to confine themselves to gifts with the potential for significant leverage.
    • When I project out what peak giving might look like (more below), my guess is that we will have a number of program staff recommending tens of millions of dollars per year each in grants, and that each staffer will be better-informed about their recommendations than we could be today about "borderline" grant opportunities.
  • I believe it's fairly common for companies to grow relatively slowly early on, as they prioritize building the products, procedures, habits, and core team that will form the foundation for later growth. This has certainly been our experience with our work on GiveWell's top charities (a larger team and more mature operation than the Open Philanthropy Project): early on, Elie and I put a great deal of time into each charity review and each hire, and we feel that this was essential for developing the ability to hire and evaluate people well and grow faster later on. With this in mind, I don't think it's as strange as it might first appear for the Open Philanthropy Project to be investing a lot of effort – at this early stage – into each grant recommendation and each hire, rather than simply growing our team and our grantmaking as fast as we can. Careful, thorough early work need not be in tension with later fast growth; in fact, it can be essential for it. I think this point tends to be intuitive for companies, which are short on both capacity and capital early on. It is less intuitive for major grantmakers (who have large amounts of capital from day one), but still seems true.

Could we do both?

In theory, we could give the bulk of our time and effort to capacity building, while still ramping up near-term grant recommendations by simply lowering our level of due diligence. If we recommended grants based largely on gut feelings (and with little discussion and writing), we might be able to help accomplish a lot of near-term good with fairly little time.

We've discussed this possibility a fair amount, and we aren't sure it's a bad idea. At the moment, however, our thinking is:

  • It's always dangerous to take on a fundamentally new kind of activity (which this sort of gut-based recommendation work would be) and expect it to take little time and attention. In my experience, proliferation of goals and projects almost always leads to proliferation of unexpected issues cropping up, and almost always leads to significant distractions. Staying focused on a contained set of goals has major benefits.
  • More specifically, every grant we recommend poses a variety of potential risks and potential distractions. For example, (a) grants affect others' perceptions of us and our priorities, and any given grant could therefore create communication challenges; (b) grantees generally want clear guidance on their odds of renewal, which can be time consuming to think through and provide. It's possible we could find a way to recommend many more grants while costing very little time, but I think that figuring out how to do this (and reaching internal agreement on where we do and don't expect to spend time) would be a project of its own.
  • We don't think of our gut instincts as a major comparative advantage. We think our ability to be reflective, self-critical and communicative is a major potential comparative advantage, and that our ability to do the kind of "capacity building" described in this post distinguishes us from many high-net-worth individuals who don't have enough capital to build out the kind of staff we're envisioning (but are well-positioned to take a more gut-instinct-based approach).

I have a general heuristic of "when in doubt, say no to everything that isn't a core priority." Right now, our core priority is capacity building, and I have doubts about trying to ramp up our giving recommendations in a way that doesn't contribute to that.

The plan from here

By early 2016, we will have been working intensively with at least three cause-specific program staff for several months. That will probably be a good time to start moving in the direction of giving more autonomy to those we've built confidence in – and along with that, to put more time into thinking about the total grantmaking we'd like them to be responsible for recommending. At that time, I hope to step back and review a variety of ideas we've had for how we might ramp up the amount of giving we recommend. I expect giving recommendations in the relevant areas to go up substantially following this (so in mid- to late-2016), even as we continue to focus on building capacity in some areas where our work is less developed (e.g. scientific research).

I believe that the relatively low amounts of giving we've recommended so far are deceptive, and I expect these amounts to rise sharply in the next year or two. If that doesn't happen, we will rethink the views expressed here.

In the long run, I'm not sure how things will look, but I can imagine some possibilities that suggest it will not be terribly difficult to reach our "peak giving" target. For example:

  • We currently have five focus areas in U.S. policy.
  • Most are relatively thin, young fields, but an exception is criminal justice reform. Chloe Cockburn, our Program Officer for this space, believes she could easily make tens of millions of dollars per year in recommendations (and would still be making hard choices at that margin). (As a side note, other foundations often have program areas that are in this size range.)
  • If, in the future, all five focus areas reach a similar state (relatively built-out field; dedicated Program Officer), our recommendations for U.S. policy could easily exceed $100 million per year.
  • U.S. policy is one of four major categories we're looking at. If the other three categories each reach a similar state, total giving could easily exceed $400 million per year. This would imply a total of ~20 Program Officers each driving grant recommendations of tens of millions of dollars a year.
  • I think this is a substantial underestimate, because I believe that focus areas in the "scientific research" and "global health & development" categories will be larger than in the U.S. policy category (some rough reasoning). In addition, we also see ourselves doing cross-cutting work, such as supporting research on the history of philanthropy, and (in the future) potentially helping to build out the general infrastructure around effective altruism.
  • All of this excludes recommendations to GiveWell's top charities, whose capacity is growing.

All things considered, I have an easy time imagining future recommendations in excess of $1 billion per year, while still feeling that we have a reasonably sized staff, that we are working only on outstanding causes, and that every program staffer has a reasonably strong understanding of every grant they're recommending and is making hard choices between outstanding options. This is far beyond what we're able to do today, though, which is why capacity building is the top priority.