Stanford Social Innovation Review : Informing and inspiring leaders of social change


Individual Giving

What Charity Navigator Gets Wrong About Effective Altruism

Why we can and must try to answer the question: What charities do the most good?

Imagine that you discover a burning building, with no one else around. Through the windows you see that, in one of the bedrooms, a family of five is trapped. Next door, a small garage that belongs to an art museum is also ablaze. The garage has recently been cleared out, but you can see that one painting worth $20,000 still remains. You know that you have time to save either the family or the painting, but it’s unlikely that you can save both. What do you do?

Suppose that you save the painting, letting the family die, justifying your decision with the claim, “I’m just really passionate about the arts.” Your action would clearly be worthy of moral criticism. It’s uncontroversial to think that someone who chooses a painting over the lives of five human beings has made a mistake in their assessment of what’s really of value.

But if that’s correct, what is the morally relevant difference between this case and someone who chooses to donate $20,000 to an art gallery rather than to a charity that would save several strangers’ lives with the same money? This is exactly the situation that philanthropists face. The burning building example is more vivid than the choice between charities, of course. But surely mere psychological salience doesn’t make a moral difference. And if there is no moral difference between these two cases, then isn’t it important for people to highlight just how much good the very best charities can do and encourage people to donate to those charities that do the most good?

The idea of honestly and impartially attempting to work out which charities do the most good, whatever that consists in, has become a core idea of a movement called effective altruism. This movement is embodied through organizations such as GiveWell, which provides charity recommendations; Giving What We Can, which encourages individuals to give 10 percent of their income to the most cost-effective organizations; and 80,000 Hours, which advises people on how to choose a career to maximize their social impact. In general, effective altruism is about dedicating a significant portion of your time or money to helping others, and trying to use that time or money in whatever way makes the most difference.

A recent op-ed by Charity Navigator CEO Ken Berger and a consultant to Charity Navigator, Robert M. Penna, attacks this idea. The authors claim that, while comparing the effectiveness of charities within a cause is a noble pursuit, the attempt to compare the effectiveness of charities across causes is “moralistic” and “elitist.”

I e-mailed to ask for clarification, and they said that the following essentially interpreted their position correctly:

It’s impossible to weigh one person’s interests against another; there are therefore no true statements of the form: “It’s more important to help person X rather than person Y.”

This is a remarkably strong claim. If this is their belief, then they have to reject the idea that the family of five’s interest in continuing to live is weightier and more morally important than the museumgoers’ interest in viewing an additional painting. They would have to reject the claim that saving the family of five, rather than the painting, is the right thing to do. That is a conclusion I cannot condone.

Moreover, if they do claim that one person’s interests cannot be weighed against another’s, then the authors should not take issue merely with effective altruism. Policymakers, healthcare providers, and distributors of foreign aid—all these bodies regularly decide what to do by weighing the interests of different parties against each other. In the emergency room, nurses in triage assess incoming patients: They treat those with life-threatening conditions immediately, make those with moderate conditions wait, and send those with minor ailments home. The reason they do this is so that doctors don’t use up valuable time treating coughs and colds when they could be treating heart attacks. Is this idea really an example of “logic so cold that even Mr. Spock would cringe upon hearing it”? Should we really label the concern to help people by a greater amount rather than a smaller amount “hyper-rationalistic” and “defective”?

Effective altruists typically believe that all parties’ interests should be given equal weight. That means that if, with the same resources, you can help one person by more than you can help another, then you should help the person with the greater need. If a given amount of money can cure either Person A’s pneumonia or Person B’s mild headache, but not both, then you should cure the pneumonia. And if, all else being equal, you can help more people rather than fewer people with a given amount of resources—curing blindness ten times, rather than only once—then you should help the many, rather than the few. Is this really a view held only by “zealots”?

The authors could weaken their claim, and sometimes they do so. Perhaps, even though there are some facts about what outcomes are better than other outcomes, it’s just no-one’s business to go around saying so. This idea is presented most clearly by Robert M. Penna in the comments:

The notion of freedom of conscience is deeply embedded in the American value system. That freedom extends to the right to make choices about what causes or charities to support. … Our essential position is that such a decision is for no one other than the individual to make. Moreover, we do not believe that it is the role of anyone to say to another that his or her cause is not “worthy.”

I agree with the authors that there are certainly cases of individual behavior, such as sexual preference, that are simply up to the individual and that it is moralistic to criticize. But these are cases where others’ interests aren’t at stake. Two people choosing to sleep with each other does not affect anyone else, so others do not have a right to criticize them for doing so. The same is not true of donations. Over a lifetime of giving, the choice to donate to one cause rather than another (or not to donate at all) is the decision to let tens or hundreds of people die, in exchange for, say, one additional opera performance or saving several stray dogs. That’s not to say that saving stray dogs or putting on operas isn’t a good or “worthy” use of money. It’s merely to say that, given the very many problems in the world, there are even better ways that donors can spend money.

Moreover, to attempt to compare charities across causes is not to say that we can always do so. Sometimes, things are just too hard. If the choice was between saving one’s spouse and one’s child from a burning building, I think it would be perfectly reasonable to throw up one’s hands and say that there’s just no correct answer. But the fact that comparing charities across causes is very difficult and sometimes impossible doesn’t mean that we shouldn’t try at all. There are easier cases and there are harder cases, and making progress even on the easy cases means doing a tremendous amount of good that otherwise wouldn’t have happened.

Effective altruism simply takes ideas that are absolutely common sense within the world of economics and public policy, and applies them to the world of individual altruism. And the results of doing this are remarkable. By giving wisely, you can save the lives of dozens of people over your lifetime, improving the education of hundreds more, and giving some of the poorest people on earth a dramatically better quality of life. Far from being “defective” as a pursuit, we think that’s pretty cool.

Tracker Pixel for Entry


  • BY Brian Tomasik

    ON December 3, 2013 12:46 PM

    There may be relevant meta-consequentialist distinctions between the “painting vs. family” case and charitable donations. For instance, while it seems appropriate to endorse a social norm expecting the onlooker to save the family, a social norm expecting similarly strict optimization of charitable giving could have counterproductive consequences, due to being excessively demanding and turning people away from altruism. That said, the general thrust of this thought experiment is clear and should move us toward caring more about triaging our donations.

  • Michelle Hutchinson's avatar

    BY Michelle Hutchinson

    ON December 3, 2013 01:10 PM

    Thanks for this thoughtful piece. Your illustration brings home the impact we can make on people’s lives. I’m grateful to have found people who care so much about helping people, and who inspire me to do likewise.

  • A well-reasoned response to what I perceived to be an unfair op-ed. Thank you, Will!

  • Thanks for this Will.

  • Robert Wiblin's avatar

    BY Robert Wiblin

    ON December 3, 2013 03:38 PM

    Great essay Will. I hope it convinces Ken and Robert to reconsider their ill-conceived position.

  • So in other words, stop any funding for arts and focus just on health and education? And then none on US, and all on DRC or other worst case country? After all, that’s where fire is hottest.

  • Lets take this to its logical conclusion, William, and say that funding for your university is wasteful until there isn’t poverty, disease, or want in the USA (as by your measure, Ivory Towers are like paintings, not worth supporting if there are more impact full options)

  • BY Niel Bowerman

    ON December 4, 2013 01:53 AM

    Hi Wayan, I don’t think it’s unreasonable to suggest that your money would have more impact if you donated it to support anti-malaria or deworming efforts in the DRC rather than to a University.  In fact a large number of Princeton students recently decided the same thing, when they were collectively given several thousands pounds to decide how to spend in a class on philanthropy:

    Having said this I do think there are cases where it is plausible that investment into research rather than global health interventions could have even higher yileds in terms of improving human welfare in the long run, the most obvious example being research into a vaccine for malaria which takes place in Universities.

  • BY Robert Wiblin

    ON December 4, 2013 05:28 AM

    Wayan, the recommendation to give to people in extreme poverty is made given the fact that US universities already have access to enormous sums of money compared to what is available in the DRC. We are trying to work out where extra money should be directed *on the margin*.

    World GDP is $85 trillion - we have plenty of resources to both have universities and the arts, and cure most people of preventable diseases. But if I had to shift $1m to one of these things today, there’s a strong case that I would do more good by expanding provision of basic health services rather than giving more to universities or arts orgnanisations in developed countries, given that they serve the ‘global rich’ and can sustain their services by charging users.

  • BY Norman Olshansky

    ON December 4, 2013 07:12 PM

    When it comes to philanthropy, people do have different values, beliefs, customs, traditions and experiences.  Those who are philanthropic (and that is too small a group) each have their own priorities based on those values, beliefs, etc.

    What is a social, emotional or practical priority for you may not be the same for someone else. Therefore, who are we to tell others who are being philanthropic, what their priorities should be.

    If you are passionate about a cause, use your influence to make a good case for support and be an advocate for those charities you support.  Cultivate and educate those with whom you come in contact with to get them to appreciate and care for this charities.

    We need to appreciate those who do give, especially since they are in such demand.  We are finding that fewer people are contributing to charities even though total giving is up.  We have become more and more dependent upon major donors and philanthropist to support the nonprofit sector. 

    We need to stop criticizing others who give but rather get more people involved in the joy of philanthropy.

  • Neil, Robert, and the silent William,

    Now that I can type on more than a small phone screen, I still find it amazingly cheeky that someone whose career is supported by donations (directly, in fact, with 80,000 Hours), at an institution that has over a £3 billion endowment, much of that via donations, is now saying that all other organizations that are not direct service providers for the poor and sick should be starved of donations.

    Shouldn’t William start at home and demand that Oxford itself be disbanded and its assets sold, with all the proceeds and endowment sent to the poorest of the poor? Or at least start with the arts and humanities departments, eh?

    As someone with a much more tenuous position - no tenure in NGO land - I have to justify my organizations actions every day to continue our funding. And through hard work and real impact, we do. To suggest that our efforts our less deserving than anther’s who is spooning the gruel into the child’s mouth, is so offensive that only someone completely out of touch with development could utter it. Or since dear William is a Research Associate, someone so ignorant of the realities.

    But hey, I’m willing to help further the education of someone already too privileged (another act of ignoring a baby on fire, apparently) and organize a debate on this topic in person or online. Let’s see how this line of thought plays out with a brighter light on its dark undercurrents.

    William, the invitation is open. I’ll even buy the train ticket down to DC from Princeton.

  • BY William MacAskill

    ON December 5, 2013 03:31 PM

    Hi Wayan, and Norman,

    Thanks for your comments. Rob has already summed up some of my sentiments.

    Please remember that saying “X is better than Y” is not saying “X is bad”. I think most charities do terrific work. But that doesn’t mean that all ways are helping are equally good. I think that saving a painting from a fire is a wonderfully brave and important thing to do. But that doesn’t mean that saving a family from a fire isn’t an even more wonderful thing to do - and doesn’t mean that you should be indifferent between saving a family of five and saving a painting.

    Wayan. My career isn’t paid for by donations, but rather at the moment from the Arts and Humanities Research Council UK. You say “William… is now saying that all other organizations that are not direct service providers for the poor and sick should be starved of donations.” but that’s clearly not what I said. As Rob has already stated, the important question for a donor is not “What should the total allocation of philanthropic funds be?” (which isn’t something an individual has any control over”, but rather, “Where should I give, given how everyone else is already donating?” If you do want to continue the discussion in person I’d be more than happy to - my email address is available on-line.

    And, Norman, far from decreasing the amount of donations, the EA movement has inspired many people to give far more than they otherwise would have done. Because the very best charities are able to help people by such a tremendous amount, I’ve pledged to give everything I earn above £20,000 per year. And many others feel similarly: Giving What We Can has raised over $150 million in pledges; GiveWell moves $10 million per year. Much of that, I believe, wouldn’t have happened had some people been working out how to have the most impact.

  • Norman, you wrote “If you are passionate about a cause, use your influence to make a good case for support and be an advocate for those charities you support.”

    I think that is precisely what Will and other effective altruists are doing. It’s just that their cause is generically “to do the most good.” We tend to think of causes as specific, like eradicating malaria, curing cancer, or helping the homeless, but effective altruists seek to simply do the most good with their donations. By definition this entails a process of optimization, which can end up pitting issues against each other using moral standards.

    The main concern I’ve had with effective altruism is that sometimes it seems to be more about the donor’s own desire to maximize his or her effectiveness than it does about a desire to address the most pressing needs of the people who need the most help. For example, in Will’s comment above he says that the important question for a donor is “where should I give, given how everyone else is already given?” That’s really more about maximizing the donor’s incremental impact. One could argue that a “pure” altruist would take a more needs-based approach, focusing on contributing toward addressing problems that have been identified as the most pressing needs (e.g., adequate nutrition, access to safe water, and shelter in developing countries), even if it means that the donor’s incremental impact is very much smaller than it could be in an effectiveness-based approach.

    Hunger kills more people every year than AIDS, malaria, and tuberculosis combined, for example, but very few hunger/nutrition charities are endorsed by EA evaluators. The reasons vary, but in some cases it’s because effectiveness can’t be demonstrated or because those charities are already receiving a lot of funding from other donors. But generically, it’s because these charities can’t ensure the level of effectiveness that effective altruists seek, even if the issues are arguably more pressing than those being addressed by the charities that effective altruists support. That to me makes it feel that effective altruism is mostly about satisfying the donor’s need to maximize the impact of his or her donation. It’s about getting the most bang for your buck so you can feel like you actually made a difference.

    I can understand that point of view, and yet the pursuit of effectiveness feels less altruistic to me than focusing on identifying the greatest needs of needy populations and contributing to meeting those needs, even if one’s own incremental impact is vastly smaller. Some of those needs may require millions of donors donating for many decades in order to achieve results. But if the results are achieved, the combined impact of all those donors would be worth it, no?

  • BY Robert Wiblin

    ON December 6, 2013 01:40 PM

    Brad, I don’t really understand the distinction between maximising one’s own incremental impact, and “identifying the greatest needs of needy populations and contributing [as much as possible] to meeting those needs”. I think what you perceive as a disagreement is actually the result of a miscommunication on our part, because all the effective altruists I know are focussed on doing as much good as possible, *in the eyes of the recipients*. Sometimes this does involve covering ‘fixed costs’, that allow future donors to go on to make a large impact on the margin.

    The reasons charities focussed on hunger (rather than nutrition) are not generally recommended is that there is already large amounts of money available to fund such programs, which makes it harder to find outstanding projects that aren’t already able to go ahead. Of course we would love to identify such projects, and if they would help people a great deal with important needs they have (eating!) with any donations received, direct funding to them.

  • Thanks, Robert—I do think it’s a communication issue, although more likely on my part: I don’t think I’m explaining my point clearly enough.

    I’ll try again to see if I can clarify the distinction I see (which of course may not be a valid distinction—it’s just my perception): In the case of a problem like malnutrition, there may already be large amounts of funding directed toward the issue but the problem itself is so large and so pervasive that it’s conceivable that every additional contribution could still be put to use. “Solving” the problem of malnutrition may require addressing systemic issues that could take decades to turn around. In a case like this, an individual donor’s effectiveness might be so small as to be seen as nil, but the combined impact of thousands or millions of individual donors providing sustained support over decades might be enough to address the problem.

    The fact that you’re focusing on finding “outstanding projects that aren’t already able to go ahead” implies that existing, well-funded projects aren’t worth supporting, at least not if you’re an effective altruist. I’m not convinced that existing well-funded projects aren’t worth supporting, but I am of course willing to be convinced.

  • Add “but you can hear a siren” to the end of the first sentence and change your justification to “I thought someone else was going to rescue the family” and the analogy is closer and the case less clear.

    Wayan, since people who act rationally rather than emotionally will always be in the minority, I can’t see your “logical conclusion” is in danger of becoming reality.  That doesn’t mean it wouldn’t be a good thing if a the percentage of rationality increased.

  • Mark Godwin's avatar

    BY Mark Godwin

    ON December 16, 2013 11:08 AM

    Isn’t the basic conceit of this article a bit disingenuous?  Are our philanthropic choices ever as stark as “saving a family” vs. “saving a painting”?  And doesn’t such a comparison give short shrift to the important work that many arts organizations undertake related to social justice and education?  Not all arts organizations are the Metropolitan Museum of Art and not all arts contributions go to fund paintings.  (That said, I’m a big fan of the Met and I love paintings). 

    I agree that we should use logic to make decisions about philanthropy and I think arts organizations could do a much better job of using data to look at their work critically and to help explain that work to donors.  I’m inclined to support “effective altruism,” but this article illustrates exactly what I don’t like about it—how reductionist and oversimplied it is. 

    In the example you give, the relative merits of Choice A (the family) vs. Choice B (the painting) are very clear.  The real world is murkier.  Not all impact can easily be captured with data (especially when it comes to the arts).  What metrics would use to compare one charity to another?  Please don’t refer me to GiveWell.  The metrics they use seem overly simplistic, and seem to come down to “hey, we’ve looked into these places thoroughly… trust us.”

  • BY Boris Yakubchik

    ON December 24, 2013 10:19 AM

    Re: Mark Godwin

    Hey Mark, thanks for your response. I too support Effective Altruism (EA), though I don’t think your worry about it being “oversimplified” is justified. No article as short as this one can properly describe the (as you point out) “murky” real world. The best one can do in this case is to be “reductionist” to focus the reader’s attention on some salient feature.

    There are many aspects of charity selection, but not all factors are equivalent in their importance. EA draws attention first to effectiveness because if one was to *only* look at one factor, one would have a greater positive impact with respect to any cause s/he cares about.

    You’re right that in the real world it’s not a simple choice of “painting” VS “family”, but if you try to come up with a reasonable explanation of why one should choose to save a painting, I think you’ll end up saying things like “it might inspire many to do good”, at which point you can equivalently say “someone in the family you save might do good”. You point this out when you suggest “arts organizations could do a much better jobs of using data”.

    Maybe I’m too near-sighted, but I don’t think a story about how art needs more support than kids that can be cured of parasitic worms for $0.50 / year each can be made. If you feel there is an argument to make - please make it; if it’s good enough I’m confident people with the EA community will listen.

    I’m all ears.

  • Stella Maris's avatar

    BY Stella Maris

    ON January 7, 2014 04:52 AM

    Nice information. Thanks for sharing this information with me. 


  • Alex Parkinson's avatar

    BY Alex Parkinson, The Conference Board

    ON January 14, 2014 09:00 AM

    You might be interested to read my Q&A with Peter Singer on The Conference Board’s Giving Thoughts blog. He outlines what effective altruism means for corporate philanthropy.

  • Surely the utilitarian imperative obliges saving the painting to the end of selling it and redirecting its $20,000 value to maximally beneficial causes, accepting the obvious truth that $20,000 can save more than five lives (and thus, assuming parity of utility-potential among individuals, can create more utility than saving five lives). If anything, this is a reductio of the whole enterprise.

    I agree, however, that Berger & Penna’s piece was below abysmal in both content and form, collapsing at several points into flat-out mud-slinging.

  • @FL: No, in the analogy, the premise is that it costs about $20,000 to save five lives (which is in the right ballpark, actually, according to e.g.

  • Geoff's avatar

    BY Geoff

    ON June 7, 2014 05:33 AM

    I’m all for using logic to determine the best way to make the most impact with our lives.  This goes beyond deciding which are the best charities to support, which is very important, and also includes:

    - earning the most money we ethically can;
    - managing our careers to maximise future earning potential; 
    - helping people we encounter on a day to day basis
    - spending some of our ‘spare’ time doing high impact volunteer work;
    - managing our finances wisely to maximise investment returns to be able to give more;
    - try and avoid losing large chunks of our net worth by getting sued, not having proper insurance, or getting married to the wrong person and then having to give large amounts to an undeserving ex-spouse, etc. Of course if we give most of our income away to start with then these risks will be less significant;
    - working until we die or as long as we can;
    - maintaining health and fitness so that we live longer to be able to give more, and to be more efficient while we are alive;
    - making an appropriate will so that a reasonable proportion goes to the most effective charities;
    - taking steps to influence others to do the same. 

    I think the average person can do way more good in their life than just saving a few dozen lives.  E.g. if it costs $25 to prevent or cure blindness in developing countries then if they managed to donate $10,000 a year for 40 years they could prevent 16,000 cases of blindness, which given the consequences and risks of blindness (including of death due to poverty caused by blindness) to those directly affected and their dependents is arguably more significant than a few dozen lives.  Or the same amount of money could vaccinate hundreds of thousands, if not millions, of children against preventable diseases.

    The average person might in addition be able to save numerous lives each year by doing volunteer work to help people in crisis.  Maintained over 40 years or more, that could add up to hundreds more lives saved. 

    And one of the best ways of influencing others to be altruistic is to be helpful to them - it might open their eyes.   

    Unfortunately the priorities of most people in the western world are seriously wrong.  It’s an absolute scandal that there are people starving to death while people who could help waste their money on frivolous things.  That’s not being moralistic, it’s just telling it like it is. 

    Ironically, there is a lot of depression in the western world partly caused by people finding their lives to be meaningless, or obesity caused by wasting too much money on food, or cancer caused by people wasting their money on cigarettes, or problems caused by alcohol because money is wasted on alcohol.  If people tried to help people in the developing world as much as they can, they might find that their own lives are a lot better. 

  • @ Geoff

    I agree with your comment.  I would add simply that many religious-based organizations have both provided millions of dollars in “private” donations and countless hours of volunteer efforts, as well, especially when a major disaster strikes.  Typically, the first people on the ground with essential items for survival are volunteers from religious-based organizations.  In the Katrina disaster, truckloads of potable water were on the scene as soon as within hours. In the aftermath, large numbers of individuals from all faiths were clearing away debris and.assisting those hit hardest in an organized and helpful manner.

    This happens all over the world on a daily basis, yet many in the US howl about the tax-exempt status of religious-based organizations.  Certainly, we should be attentive to those that would abuse this exemption and stop the abuse immediately.  We should also be sure that all religious-based organizations are paying their fair share when they spin off profit-making units or divisions of the organization.  But, the value of what these organizations and their members provide in times of greatest need is truly remarkable.

Leave a Comment


Please enter the word you see in the image below: