The GiveWell Blog

Exploring policy-oriented philanthropy

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

Over the last few months, I’ve been working on improving our broad understanding of the role that philanthropy can play in influencing and informing public policy. We feel that this is one of the major categories of philanthropy that we’re currently least well suited to understand.

This is the first in a series of posts. This post discusses:

  • Why we think it’s important to explore policy-oriented philanthropy.
  • What we’ve done so far in this exploration.
  • What our key questions have been.

First, a general note: there are cases in which philanthropists are legally constrained from funding certain types of policy-oriented activities, particularly (a) attempting to influence elections and (b) lobbying. (A summary of these constraints is discussed on page 13 of an Atlantic Philanthropies paper on supporting advocacy.) We haven’t yet focused on thoroughly understanding these issues, whose relevance may be limited since most of our audience has freedom to structure its giving as it chooses (i.e., we aren’t managing an endowment that’s locked into a particular organization type). When we refer to “influencing and informing policy,” we mean this statement to broadly encompass a variety of possible activities from lobbying and advocacy to general provision of information and education.

Why explore policy-oriented philanthropy?
From what we’ve seen, it’s very common for major philanthropists to seek some degree of influence on public policy. This sort of work is quite prominent in the previously discussed list of philanthropy’s success stories, and most of the major foundations we’ve spoken to put a fair amount of emphasis on influencing policy.

One argument we’ve heard for focusing on policy is the sheer scale of government as compared to philanthropy. To give a simple example: the U.S. government alone is estimated (by IHME) to spend more on global health aid than all foundations and NGOs (excluding GAVI and GFATM, which draw much of their support from governments) combined; the discrepancy grows if one considers other governments, multilateral organizations such as the World Bank and World Health Organization, etc. We believe this general pattern to hold across many sectors, such that a relatively small (percentage) impact on government spending could justify a huge expenditure in philanthropic terms. Of course, there are many ways in which the importance of policy can go beyond funding. For example, migration to the developed world appears to be an extremely effective poverty reduction measure, but in most cases, migration faces hard policy restrictions.

For nearly every cause we’ve looked into, influencing policy is one possible path to having as much impact as possible per dollar spent, and in some (such as labor mobility) it is the only clear path to impact.

However, influencing policy is unlikely to be something we can analyze using our traditional approach and criteria. Because it is inherently adversarial – advocating for any given policy change likely means advocating against someone else’s preferred policy – there are unlikely to be proven, repeatable interventions with easily quantified expected impact. We’ve been trying to understand what sorts of things philanthropists can do to try to influence policy, and under what circumstances one might expect these to be effective.

Our process for exploring policy-oriented philanthropy
We’ve come to believe that in the early stages of an investigation, when we often don’t know which questions to ask, the ability to have extended, repeated, friendly, low-stakes interactions to get “grounded in the basics” is crucial, and we’ve largely used referrals from existing contacts to get started. Our investigation has included:

Key questions we’ve focused on
While we’ve largely tried to keep our investigation open-ended, we’ve been particularly interested in the following questions:

  • What are the “tools” of policy-oriented philanthropists? What are the activities one can fund that have a chance of influencing/informing policy?
  • What is the track record of policy-oriented philanthropy? With what probability, and on what time frame, can one reasonably expect to have an impact on policy?
  • What are the best opportunities to make a difference within policy-oriented philanthropy today?

Each will be the subject of a future post.


  • There are important normative distinctions to be drawn between policy-oriented philanthropy and “normal” philanthropy. Normal philanthropy involves a personal choice of donating my money, while policy-oriented philanthropy also seeks to influence a governmental decision to force taxpayers to spend their money or share their economic resources. It may make sense to measure the impact of normal philanthropy in terms of lives saved, but perhaps policy-oriented philanthropy should be more properly measured in terms of achieving societal goals of stability, mobility, sustainability, and cohesiveness for the primary benefit of the society being forcibly taxed. In other words, one must ask whether governments should primarily strive to save lives or to benefit the governed.

  • Colin Rust on October 25, 2013 at 8:17 pm said:

    Joe D., I think hardly anyone would dispute that “governments should primarily strive to … benefit the governed.” But that doesn’t remotely mean government should solely do that or that policy-oriented philanthropy can’t or shouldn’t seek to influence for example the magnitude or make up of foreign aid. Like any kind of expenditure in a democracy, some taxpayers will agree with and some will disagree with that, but it’s perfectly legitimate.

  • M. Jahi Chappell on November 4, 2013 at 1:48 pm said:

    I recommend John Kingdon’s classic “Agendas, Alternatives, and Public Policies”; the 2nd edition is available on Amazon. This has been incredibly influential on my work, and I think is one of the better (more accurate) descriptions of policy change processes, though I think it is also somewhat disfavored because it emphasizes the role of a semi-random process. See also the classic, and subsequent work by Cohen, March and Olson on “The Garbage Can Model of Institutional Choice.” Some critiques of these two works (and others like them) support the idea that policy processes do take place in less-random, more orderly patterns in (many? some?) cases. I would argue that the system resembles the simple models of Robert May from the 1970s, showing that a time-lagged simple system of a predator and prey populations can generate deterministic, but completely chaotic (in the technical sense) results. With other values for the constants, the system behaves very “nicely”.

    Kingdon’s “system” has three interacting streams, that undoubtedly see time lags between them (Policy Solutions, Public Sentiment, and Electoral Politics). It could thus actually explain both chaotic (random-seeming) and simple policy processes in principle.

    It doesn’t lend itself to easy suggestions on how to influence policy, but I think it does lend itself to some *obvious* (non-easy) ones.

Comments are closed.