Downgrading confidence in earning to give

Holden from GiveWell recently wrote a post outlining the difficulties they’ve had in finding more interventions that can be funded by small independent donors. The upshot is that

  • Governments fund most good-seeming interventions;
  • Other opportunities get funded by large donors faster than GiveWell can move;
  • Many of the problems are related to logistics, rather than funding.

Reading the comments, it seems the main issue is that there are enough reasonable (evidence-based) large donors that if you can create cost-effective ways to distribute proven interventions, it’s very likely they will get funded.

Here are some other things I’ve noticed recently that decrease my confidence that earning to give to proven interventions is the best option available for effective altruists:

  • GiveWell seems to be having trouble hiring people (their self-evaluation indicates they wanted to have eight employees by the end of 2012, whereas their about page lists that they still have only six), suggesting there’s a shortage of philanthropically minded people with strong enough quantitative skills.
  • The Against Malaria Foundation was founded by a businessman, not a person with lots of previous charitable experience, which suggests that there may be shortages of qualified people in the field of aid charities (or at least starting up such charities).
  • When Toby Ord came to talk to Harvard High-Impact Philanthropy, he mentioned preliminary calculations showing that meta-research could be vastly more effective (a factor of 103 or 106) than direct intervention if enough money would flow into the new intervention. Even if practical considerations take this down 2-5 orders of magnitude it would be clearly the best option.

This has the following implications:

  • Doing finance, or other high-paying highly-skilled careers, might be dominated by doing directly charitable things. If you are competent enough to succeed in such a career, you will probably be able to pay for less than one similarly competent person working directly on altruism. Since human capital seems to be scarcer than actual capital in several highly important areas, the net effect is worse.
  • Altruism advocates should probably be spreading a broader message. Most of the mass-appeal advocacy I’ve seen tries to boil down its message to something easily-digestible like “Want to Give Back? Get a Job on Wall Street”.1 But if we attract new donors with that kind of message, only for them to find that all the good interventions are saturated with funding, it’ll do more harm than good.

Thoughts? Counterarguments?

  1. Not to pick on Will’s article specifically—it’s a great, well-written summary of altruism’s talking points, and extra props to him for getting so much visibility! This is just the first example of this trend that I remembered. 

3 commentscomment


How about finding a way to fund quantitatively-minded people to do philanthropic work? Perhaps part of the problem is that the choice between Wall St and GiveWell as a career is too stark in terms of earning potential, particularly if you are young and might want to have a family in very uncertain economic times. Or set up a program to help mid- or late-career quantitatively minded people to help philanthropic organizations.


Will Crouch


This is a good post! 80k has actually been trying to broaden its message away from just earning to give since it changed its name from High Impact Careers to 80,000 Hours (Nov 2011), for precisely the reasons you give. Unfortunately, Earning to give is such a salient and sexy concept that it takes a life of its own, and its much easier to get media attention on earning to give than on “Work for a Charity Evaluator!” My view is that a certain % of effective altruists should be earning to give, and that it should be thought of as an easy-to-quantify baseline of how much good you can do, so that you can compare other, less easily quantifiable careers against it. But it’s not the best route for everyone.



Nat, I’m not sure what salary GiveWell pays, but they claim (I believe) to have their funding needs met for the short term, which suggests that people are still the bottleneck. A program to help mid- to late-career people would be really cool, and in fact I think such things exist (maybe not focused on effectiveness though), but that would need even more human capital to start up.

Will, that’s interesting—I think when I first ran across 80k, I had heard the EtG message from enough other sources that I pattern-matched 80k onto that general message without paying too much attention. The coordination problem (which percentage does what?) is also a thorny one…