Is Effective Altruism Killing the Love?

In July 1990, the Australian state of Victoria put a law requiring cyclists to wear helmets into effect (1). More than two decades later, it is unclear whether or not the introduction of the law had a net societal health benefit (2). This might be puzzling when considering that cycling with a helmet on is safer than cycling without it. It prevents head traumas, especially those resulting from accidents at lower speeds. In London, the police started last year to stop cyclists without helmets and to educate them about the benefits of wearing a helmet (3). However, one of the arguments against laws requiring the wearing of bike helmets is that it significantly reduces the number of people that cycle. Hence, there is a good chance that the health costs – increased morbidity due to lack of exercise outweigh the health benefits – less head traumas (2). In the words of Milton: “Easy is the descent into Hell, for it is paved with good intentions.” Might effective altruism have similar unintended consequences?

Effective giving tries to maximize the impact of donations on social good. The opposite of effective giving can, for instance, be described in terms of the identified victim effect. Baby Jessica is the poster child (or poster baby) for this effect. At the age of 18 months, Jessica fell into a well; people donated more than $700,000 for her rescue (she is alive and well). The staggering amount of donations in cases like these appears to be motivated by the strong compassion people feel for an identified victim. Anonymous groups do not evoke such strong compassion (4), even though they might be much more deserving than a single child in a well, and even though the same amount donated to the group might vastly increase social good. So, what happens if we teach people about the identified victim effect?

In a series of studies, Small, Loewenstein, and Slovic (5) look at what happens when people learn about the identified victim effect. Participants either did or did not learn about the identified victim effect and how it influences charitable donations, and then had a chance to donate £5 of their endowment; the charity to which they could donated was promoted via either an identifiable victim or a statistical victim. Participants who did not learn about the identified victim effect donated more than double of the money in the identified victim condition than in the statistical victim condition. In contrast, participants who were educated beforehand did not donate more in the identified victim condition than in the statistical condition. Unfortunately, though, they didn’t donate much at all. In other words, the education had the effect of reducing the overall level of donations by stopping people from giving when faced with a more “emotional” appeal. Educating people did not, in contrast, increase donations in response to more rational appeals.

One reason for the lack of donating in the “educated” condition might be that the charity appeal that offered only statistical information lacked the emotional triggers needed to motivate people to give. Maybe combing statistical information with an identifiable victim—combining empathy with evidence—will pull people towards more anonymous victims. Peter Singer in his TED talk, for instance, provides both a powerful emotional appeal as well as statistics and reason. However, when participants read a donation appeal that has both, an identified victim and statistical information, they give as little as participants who read a donation appeal that has only statistical information (Study 3). Statistics seem to override any emotional appeal. The authors argue that statistics prime a deliberative mode of thinking at the expense of emotional intuitions, and once people are in such a deliberative thinking mode, nothing motivates them to give. And indeed, people who have been primed to rely on their intuitions give when facing an identified victim, but not when facing a statistical victim. In contrast, people primed to deliberate give to neither the identified victim nor the statistical victim (Study 4). In other words, deliberation breeds selfishness, whereas emotional intuitions give rise to pro-social behaviors under certain conditions (see 6 for similar arguments and studies).

If we take these findings serious, then educating people about effective giving might have detrimental effects on the overall amount of donations. Cost-effective charities have to highlight at some point, why it makes more sense to give to them than to less cost-effective charities (e.g., Malaria prevention instead of stopping FGM). In order to do so, they have to appeal to reason rather than emotions, and at least some minimal use of statistics seems hard to prevent. What the presented studies suggest is that most likely, the majority of people respond to such an education by stopping giving to ineffective charities, but not replacing them with effective ones. Considering the millions of people who watched Peter Singer’s TED talk, for instance, such effects might be profound. While this is obviously an empirical question at the end, measuring the net impact of effective giving on social good, it appears reasonable to entertain the possibility that effective altruism might erase the emotions that underlie giving in some people, or at least reduce them. So, what can be done?

The problem is not effective giving itself – effective giving is better than non-effective giving, at least from my point of view – but how to promote it best. What these studies and arguments highlight is that scientists, philosophers and lay people like to think of the brain/mind in dualistic terms – intuition versus reason. It is a handy caricature for sorting the complexity of the mind into two neat categories. Furthermore, most research seems to highlight the problems of intuitions. And surely, there are many instances in which it is good to distrust intuitions in our daily life. Once people are made aware of how their intuitions lead them astray (e.g., in the case of the identified victim effect), they probably start to exclude them from the decision making process, especially when these intuitions are not the strongest to begin with. A rational approach to buying a car is better than an emotional one, but a rational approach to morality might just never get people going.  The challenge is to prime and promote ways of thinking about giving that incorporate both, emotion and reason. And clearly, there are people who are able to do so, and situations in which we all combine emotional oil with rational refinery.


  • Facebook
  • Twitter
  • Reddit

6 Responses to Is Effective Altruism Killing the Love?

  • Cody Fenwick says:

    Thanks for the interesting post, I think these issues are incredibly complex and worth discussing.

    One of the considerations that often gets lost in this discussion is the idea of promoting effective giving vs. promoting certain causes. Ideal marketing campaigns for certain charities or organizations might be different than the ideal promotion of effective giving groups. While the studies you cite have powerful effects, the effects do not apply to everyone, and many of the people who they do not apply to could be very significant contributors to important causes (who perhaps might be lost if promotional materials have mass appeal.)

    The effect might still be a decrease in charitable giving overall–but it’s not obvious to me that this is a bad thing. There are countless cases of charitable endeavors that are seriously harmful, not just neutrally ineffective. Perhaps getting people to be more skeptical of charities overall, even if it reduces total giving, would be beneficial.

    • Andreas Kappes says:

      Cody, thanks for the comment. And I very much agree, it’s complex and marketing campaigns can have different targets. My wife pointed out to me that Bill Gates and Warren Buffet did not lose their interest in giving due to a stats-based approach; rather the opposite. What I would be really interested in is to better understand when and under which conditions these effects occur.; when people’s compassion gets grounded by more rational appeals, and when not. And I agree very much with your second point about seriously harmful charities; but how can we educate people about such charities without them losing interest in giving overall? No easy answers here, I am afraid.

  • Dave Frame says:

    Its hard to get people to give. If you want more of it, you need to send strong positive reinforcement. If your response to people’s giving is anything other than “thank you – you’re awesome” then they will give less. By turning it into a moral and intellectual competition in which a effective altruists lecture people about their obligations to give to the most effective charities (rather than the charities that elicit the strongest emotional responses), effective altruists will undermine people’s already brittle and infrequent departures from self-interested rationality.

    See, for example: http :// (And thanks to Elizabeth Baldwin for first letting me know about this cartoon!)

    PS – I’m still going to give money to cats.

    • Andreas Kappes says:

      Dave, first of all – thanks for the wonderful cartoon (and Elizabeth)! This encapsulates exactly the phenomenon I’m interested in; the conversation, or as you put it: lecture, about the right way of giving kills people’s interest in giving. And maybe this is also why some people like yourself have such strong reactions to these kinds of appeals; they can sense that this undermines their motivations. While I think it is very important to think about these issues and try to figure out the best ways to help other people (and I’m definitely on board with effective giving), the ways that is communicated have to be carefully considered. Your perception of it as a ” moral and intellectual competition”, for instance, seems to indicate that something went wrong.

      P.S. – I hope you donate to effective cats charities 🙂

  • umar syed says:

    Very thought provoking, well written article.

    I think you touched upon a very salient point when you said that people/scientists tend to divide our brain/process-making into two categories: emotional and logical. Charity at its core is predicated on the notion of helping someone with their struggles in a seemingly selfless way. But to encourage effective altruism, it’s almost necessary to encourage rationalizations about cost-effectiveness and how to make the best use of your money. But then the charitable, emotional impulse fades (the one that initiated that cost-effect rationalization). When your initial reason for doing something disappears, it’s very hard to rationalize doing it anyway (If I want to give money to a charity to help a cause, but someone tells me it’s an ineffective charity, my original reason for giving to that charity is bunk, and no amount of logic could re-ignite my desire to give to charity).

    I think one avenue to circumvent this is to find ways to appeal to a person’s selfishness and logic. For example, some charities tout that your donations are tax-deductible, and spending less money on taxes is a strong logical appeal. This way, you aren’t trying to salvage that lost emotional impulse, but rather act on a new logical one.

    • Andreas Kappes says:

      Thanks Umar. If I understand you correctly, then you see no way of reasoning that does not eliminate the initial impulse of giving. I think the really intriguing point implicit in your comment is that this is not true for selfish impulses; selfishness seems to be more ingrained in reasoning. But at other times, our reasoning might just as well ground selfish impulses, right? Giving people reasons on why tax-deduction is maybe not the best way of thinking about children in need might also eliminate such impulses, I could imagine. I would suggest that when reason is directed against impulses, the latter are grounded. But reasons don’t have to necessarily be directed against them, and this is the challenge for effective altruism; making the case for it without deflating charitable impulses.


Subscribe Via Email