Effective Altruism and the Urge to Optimize Help

–By Anton Karpovich & Arwin Wang–

A modern philanthropic movement has recently gained popularity in the young, technical milieu of self-described ‘rationalists’. Called effective altruism (EA), it has been promoted by prominent philosophers like Peter Singer and William MacAskill, and aims to do humanitarian work that empirically benefits most the greatest number of people. In practice, this usually means donating money to help Africa. EA groups such as GiveWell highlight charities that carry out malaria-prevention programs, deworming initiatives, and direct cash-transfer operations such as GiveDirectly, urging readers to ‘give what they can’ to those organizations.

Yet EA has faced substantial criticism from other humanitarians. Some claim that it is the apotheosis of neoliberal commodity-fetishism, making human caring a fungible good in a grotesque attempt to quantify the unquantifiable. Others dispute its insistence on unusual interpretations of utilitarian ethics, its continued support of interventions that were shown to be ineffective, or its encouragement to pursue higher-paying, unethical jobs so that one can donate more to charity. EA is a fraught movement born of a Big Data age. The goal of this commentary will be to look at the issues that surround it, especially in the context of its African aid.

Surely Nobody Asked These Questions Before?

The motivations of EA stand in opposition to those of most other humanitarians. As Liisa Malkki documents extensively in The Urge To Help, aid workers and elderly knitters alike help others primarily to satisfy their own moral and affective needs (4,8). This is not a bad thing if done right. But a focus on the giver often leads to a general ignorance of the needs of the recipient. Interventions take the form of showy emotional appeals and clearly-demonstrable actions. Sponsorship of individual children is facilitated by church mission trips to Africa, complete with gratuitous photo-ops (Friedus 340, 346-347, 394, Bornstein 310-320), and without considering the deleterious second-order effects of such aid. A moralistic campaign to ban coltan extraction led to mass joblessness in the eastern DRC (Chase), and similarly, sentimental messaging of an ‘orphan crisis’ gave rise to a cottage industry of child abduction in Uganda (Cheney and Rotabi 376-380). Unless someone documents these abuses, they generally go unnoticed by the Western aid-giver.

Enter the data-driven EA ethos. Effective altruists seek to mitigate these externalities by empirically assessing the effects of their chosen interventions. If at all possible, this is done by conducting a ‘gold-standard’ randomized controlled trial. GiveDirectly serves as a benchmark for this process (Arjmand). Started by a group of economists (Aizenman), the charity is conducting trials of direct cash donations to the needy in various Kenyan villages, examining different aid strategies to determine which is optimal, with the help of external research partners (Haushofer and Shapiro). To conventional aid workers, such gratuitous experimentation may seem callous or unethical (Aizenman). But the data is essential to optimizing EA’s main success metric—utility provided to the aid recipient. This is an intellectual goal (Karnofsky [a]), quite unlike the practical needs of the giver (cf. Malkki ch. 5) that are centered in most charitable acts.

Motivated solely by the numbers, EA-endorsed charities typically focus on problems dismissed as ‘mere’ by other aid groups that raise awareness of novel or dramatic concerns. For instance, malaria in Africa kills over 400,000 people per year, a number similar to or greater than in other epidemic diseases such as Ebola or HIV/AIDS (Richards, World Health Organization [a,b]). Even in the midst of Liberia’s Ebola outbreak, local residents expressed consternation over the lack of resources to combat malaria (Richards 159). EA logically highlights this underserved need by steering donations to groups such as the Against Malaria Foundation (GiveWell). Yet the movement finds common ground with other humanitarians in their struggles to deal with the political. After World War II, need-based agencies like Oxfam saw that politics often perpetuates injustice and harm (Barnett 273), resulting in a shift in activism recapitulated by GiveWell as they consider whether to endorse restrictions on toxic pesticide sales (Arjmand). Per their analysis, small political campaigns can have large effects on global welfare, but must be done carefully as their impacts are uncertain and hard to empirically analyze (Arjmand). But EA, unlike Oxfam, seems to ignore the moral capital of neutrality jealously guarded by the ICRC and MSF (Barnett 276, Malkki ch. 6). In many instances, the access needed to do effective aid work is only granted if locals see humanitarians as ‘above politics’ (179, Barnett 276). Whether GiveWell will be harmed by foregoing this remains to be seen.

However, logic on occasion still yields to affect in EA. Some of its interventions, such as deworming campaigns in West Africa, are not actually supported by the evidence. While GiveWell heavily promoted multiple charities of this type, a Cochrane meta-analysis suggests there is next to no indication that reducing parasitic worm burden improves childhood nutrition, academic achievement, or lifespan (Taylor-Robinson et al., Deaton). Clearly, an awareness of rationality does not necessarily defend EA givers against biases. As with all humans, they are slow to acknowledge contrary evidence that questions existing modes of thought (Anderson 93). The end result is an embedding of affective aid into a discourse of logic and reason, satisfying the psychic need (cf. Malkki 4, 8) of optimizing aid. EA eases the rationalist’s first-world conscience by ensuring one is doing the most good they can, as in Singer’s 10% pledge (paragraph 2). As I discuss in the next section, such an urge demands quantification, often done in economic terms.

The World, Commodified: EA as “Neoliberal” Praxis

To date, many attacks on EA have painted it as the repugnant outgrowth of the modern variety of technocratic, international-consensus capitalism, known by the left as ‘neoliberalism’ (Friedus 401, Hearn 281, Eikenberry and Mirabella 43). As with most snarl words, this term is rather nebulous, being used to describe a range of disparate outlooks from those of Margaret Thatcher to Ta-Nehisi Coates (West). But many self-described neoliberals have endorsed EA principles and causes, as evidenced by a large recent fundraiser for the Against Malaria Foundation (reddit.com/r/neoliberal). And the kernel of truth in critiques of neoliberalism, especially as applied to EA, is that it judges success and optimality primarily in economically quantifiable terms, rejecting human emotion in the distribution of aid (Karnofsky [a], Eikenberry and Mirabella 44).

A constrained example is instructive. There has been much recent debate over the influence—positive or negative—of used clothing donations on African lives. Economists like Garth Frazer conclude, based on empirical factors, that clothing imports are severe burdens on local producers, but benefit the consumer by lowering clothing costs and increasing choice (1764-1766). Sociologists, however, emphasize the intangible, transformative cultural effects of this clothing—salaula–in terms of local desires and discourses (Hansen 14). Both perspectives are valuable, yet incomplete; EA takes the former, influencing its humanitarian proscriptions.

MacAskill frames his metric of good in terms of the Quality-adjusted Life Year (QALY). This is a widely-accepted measure in clinical contexts, where one unit of utility is assigned to a person spending a year in perfect health, with points deducted based on the severity of any illness (34). Yet this measure is individual-focused, failing to capture cultural or community health, or any of Hansen (et al.)’s transformative discursive relationships. Such complications may well outweigh the immediate boons of ‘standard’ EA initiatives, such as GiveDirectly’s cash transfers. Community jealousy (Aizenman), mistrust of first-world motivations, and the erosion of state frameworks (Hearn 294-295, Acemoglu) have all been cited as negative societal outcomes of such actions, not included in its standard statistical evaluations of effectiveness. Indeed, “how much more valuable is [it] to save the life of a one-year-old than to send a six-year-old to school?” (Deaton, Acemoglu, paragraph 3).

Assessing these issues requires, as in the Ebola outbreak, a “people’s science” where local knowledge of cultural matters is integrated into large-scale policy judgments (Richards 148-149). But the distance of most EA interventions makes this difficult. Extreme absolute poverty and loss of QALYs are concentrated in sub-Saharan Africa, affecting the majority of the population, instead of less than 3% as in the US and Europe (Roser and Ortiz-Ospina [a]). As life satisfaction scales logarithmically with income (Ortiz-Ospina and Roser [b]),  EA by its own logic should not aid the developed world. This is very much unlike other agencies, including the Red Cross, that frequently organize relief to victims of local natural disasters, or dispatch ‘volunteer friends’ to contact the lonely (Malkki 147). To them, local action is a way of strengthening community bonds and being more responsive to particular socio-cultural needs. EA has no yardstick to measure such externalities, leading its opponents to call for a return of charity to domesticity (Deaton, Eikenberry and Mirabella 45-46).

Similarly, EA’s focus on the economic encourages a controversial practice–”earning to give”. This is done by pursuing a higher-paying career solely to donate more to effective charities (Cowen). In certain situations, theorists like MacAskill claim that the benefits of giving volunteer labor (e.g. by practicing medicine in the Red Cross) are less than the monetary good provided by giving away a large fraction of a six-figure income (77). His website, 80000 Hours, offers ‘rational’ career advice based on this principle. But to critics, that thinking again ignores the intangible externalities of vocation. Most observers see a life spent doing good works as intrinsically more ethical than one of accumulating wealth (Acemoglu); even MacAskill’s vote for the greatest person ever to live is not Bill Gates, but Soviet virologist Victor Zhdanov—a man who gave bold ideas and public health expertise to eradicate smallpox, not money (ch. 5). Also, a world in which the rich are the superstars of aid approximates the era of the robber barons, captured in Carnegie’s Gospel of Wealth, with the same undue, unaccountable influence on local and global institutions that weakens the social fabric (Saunders-Hastings, Eikenberry and Mirabella 45). If African poverty is a political, postcolonial question, reinforced by “neoliberal” power structures and concentrated wealth, then ‘earning to give’ and aid-by-economic calculus are at odds to human welfare (46, Friedus 401). Perhaps that is why it is so alienating.

Assessment of the economic criticisms of EA ultimately depends on one’s priors. If human worth cannot—or should not—be measured in dollars, or if the Washington consensus is oppressive and needs to be torn up root and branch, then the movement and its policies make no sense (Eikenberry and Mirabella 46). But if reason is the watchword, enabled by dispassionate analyses and gradual reform of the policies and actions that will do the most good, EA becomes much more appealing (Reich, Singer). That latter choice also leads into the inherent utilitarianism of the movement.

Utilitarianism With an Axe to Grind

EA’s moral framework almost universally takes as a baseline assumption an unusual and perhaps naive form of utilitarianism. To Singer and MacAskill, doing moral arithmetic on QALYs is straightforward and necessary (MacAskill 34). Yet this mere addition leaves huge theoretical and practical blind spots, especially in issues of fairness and justice (Skelton 140, Gabriel 459). Effective altruists do not dismiss justice altogether; it is, after all, often a good way to increase utility. But it is not a primary quantity to be maximized. This leads to EA celebrating policies, such as sweatshop labor, that raise living standards even while increasing inequity in ways deemed grotesque by other humanitarians (461). Such an idea reflects the peculiar ethical mismatches between effective and ordinary altruism, as well as its biases toward market fundamentalism discussed in the previous section.

A more damning objection is the utilitarian ‘Repugnant Conclusion’. In any system that uses a utility function to sum well-being across a population, a paradox can be constructed in which adding a large number of miserable, mediocre lives to the population would improve group utility, despite this being absurd, illogical, and contrary to common sense (Arrhenius et al.) Such an issue may seem purely theoretical. Yet because of the fringe beliefs shared by many at the core of EA—drawn from a homogeneous subset of itinerant computer scientists obsessing over ‘friendly AI’, such as those in the Machine Intelligence Research Institute (MIRI)—it has troubling effects in practice (Matthews).

The greatest fear of EA AI theorists such as Nick Bostrom is the total extinction of humanity. To them, that calamity would not only extinguish the 7 billion human lives currently in existence; it would also eliminate 1052 healthy lives in the future, made possible once the ‘AI Singularity’ allows interstellar travel and computer simulation of minds (Bostrom [a] 19, Matthews). With such vast numbers, the Repugnant Conclusion kicks in, dictating not only that minimizing the chance of this catastrophe is the most important altruistic work possible, but that all other humanitarian issues are “rounding errors” in comparison (Bostrom [a] 19, 26, Matthews). And to thinkers in this vein, ‘unfriendly AI’ is the single most probable cause of such an event. Hence the only logical outcome is to give as much money as one can to MIRI and related groups such as the Future of Life Institute, which have already received at least $10 million from Silicon Valley personalities like Peter Thiel and Elon Musk. (Matthews, Bostrom [a] 26, [b], Kastrenakes).

However, to everyone outside the EA bubble and even some within it, such ‘altruism’ is patently nonsensical and ineffective. GiveWell’s evaluation of the Singularity Institute was quite negative, stating that its mission was poorly justified and likely to backfire (Karnofsky [b]). Such fringe beliefs also serve to alienate outsiders who may be sympathetic to EA’s goals (Matthews). Most humanitarians are driven by their professional ethics (Malkki 51-52) to solve actual, currently existing problems, such as malaria or global poverty. They do not seek to participate in an AI death cult driven by dire universal threats of infinitesimal probability (Bostrom [c]), Matthews).

But improper utility-driven evaluations can manifest even in EA’s saner circles, such as when altruists categorically reject the domesticity of aid (Malkki 78). Kathy Forth recently made a convincing case that working to reduce instances of sexual assault in the EA community would benefit “diversity, job satisfaction and productivity”, as well as “saving lives [and reducing] suffering” in a manner fully consistent with EA principles (Forth, paragraph 8). But the predominant response to those ideas was dismissive. Commentators either minimized the well-documented psychological harms resulting from sexual trauma, or attacked the rigor and impartiality of ‘feminist researchers’ in the social sciences, upon which Forth’s argument was based (Marcus_N, e102). They (mis)used utility and concerns about the quality of evidence to justify a focus on inoffensive, international problems such as deworming. This seems to reflect a preexisting skepticism of women’s experiences and domestic concerns, as well as failures of empathy among the predominant demographic of EA—white male Bay Area programmers (e102; Forth, “Impact” paragraph 16; Matthews).

Abuse of utilitarianism also plays out in the African context. Rwanda’s longtime president-cum-dictator Paul Kagame has attracted large amounts of international development aid by appealing to bean-counters of effectiveness, visibly steering basic necessities such as bed nets and healthcare to rural communities (Deaton). Yet this help is then used as leverage to support a one-party state. Rwandan dissidents are routinely arrested and killed, sometimes simply because of criticizing problematic aid initiatives like a badly-implemented plan to replace thatched roofs (Deaton, Sundaram). Nor do the regime’s externalities count in Western eyes. Rwanda funds itself with sales of Congolese coltan, seized by its military to the detriment of the local poor (Chase). It is all too easy for ulterior motives to shape the moral calculus of utility—and if the World Bank falls prey, what can small, poorly resourced EA groups like GiveWell do (Deaton, Sundaram)? The answer seems to be ‘not much’.


Effective altruism is a novel and interesting take on the urge to help. It strives to be rational and logic-driven, standing in contrast to the usual moralistic tendencies of humanitarianism through quantitative optimization of charity. If done right, it can provide biting empirical critiques of the showy but useless forms of aid too often directed toward Africa. The movement’s meta-charities like GiveWell then direct resources to more practical ends. But to err is human, and EA is no exception. Too often, practitioners use its logic to justify their own prejudices—towards ineffective deworming programs, free-market fundamentalism, or even fantastical extrapolations of ‘rationalist’ beliefs that claim AI-induced doom is the only important issue facing humanity. Despite EA’s unfeeling edifice, its members have unique fears and codes of professional ethics that define it as a humanitarian movement just like any other. Such human factors will ironically remain important to its growth, even in an increasingly globally interconnected and data-driven 21st century.

This post may have been edited by admin for clarity and length.

Additional References

Acemoglu, Daron. Response to “The Logic of Effective Altruism.” Boston Review, 1 Jul. 2015, http://bostonreview.net/forum/logic-effective-altruism/daron-acemoglu-response-effective-altruism.


Aizenman, Nurith. “An Experiment Gives Cash Aid To The Poor. Is That Ethical?” NPR Goats and Soda, 13 Sep. 2017, https://www.npr.org/sections/goatsandsoda/2017/09/13/542261863/cash-aid-changed-this-family-s-life-so-why-is-their-government-skeptical.


Anderson, Craig. “Abstract and Concrete Data in the Perseverance of Social Theories: When Weak Data Lead to Unshakeable Beliefs.” Journal of Experimental Social Psychology, vol. 19, 1983, pp. 93-108. doi:10.1016/0022-1031(83)90031-8.


Arjmand, Isabel. “Considering policy advocacy organizations: Why GiveWell made a grant to the Centre for Pesticide Suicide Prevention.” GiveWell, 22 Mar. 2018, https://blog.givewell.org/2018/03/22/considering-policy-advocacy-organizations-givewell-made-grant-centre-pesticide-suicide-prevention/.


Arrhenius, Gustaf, Jesper Ryberg, and Torbjörn Tännsjö. “The Repugnant Conclusion.” The Stanford Encyclopedia of Philosophy, ed. Edward N. Zalta, Spring 2017, https://plato.stanford.edu/archives/spr2017/entries/repugnant-conclusion/.


Bostrom, Nick. “Existential Risk Prevention as Global Priority.” Global Policy, vol. 4., no. 1., pp. 15-31. doi:10.1111/1758-5899.12002.


——-, “This Oxford professor thinks artificial intelligence will destroy us all.” By Dylan Matthews, Vox, 19 Aug. 2014. https://www.vox.com/2014/8/19/6031367/oxford-nick-bostrom-artificial-intelligence-superintelligence.


——-, “Pascal’s Mugging.” Analysis vol. 69, no. 3., pp. 443-445. doi:10.1093/analys/anp06.


Cowen, Tyler. “Effective Altruism: Where Charity and Rationality Meet.” The Upshot, New York Times, 14 Aug. 2015. https://www.nytimes.com/2015/08/16/upshot/effective-altruism-where-charity-and-rationality-meet.html.


Deaton, Angus. Response to “The Logic of Effective Altruism.” Boston Review, 1 Jul. 2015, http://bostonreview.net/forum/logic-effective-altruism/angus-deaton-response-effective-altruism.


e102. Comment on Kathy Forth, “An Exploration of Sexual Violence Reduction for Effective Altruism Potential.” Effective Altruism Forum, 13 Nov. 2017, 4:26 a.m., http://effective-altruism.com/ea/1gy/an_exploration_of_sexual_violence_reduction_for/cle.


Eikenberry, Angela, and Roseanne Mirabella. “Extreme Philanthrophy: Philanthrocapitalism, Effective Altruism, and the Discourse of Neoliberalism.” PS: Political Science & Politics, vol. 51, no. 1, 2018, pp. 43-47. doi:10.1017/S1049096517001378.


Forth, Kathy. “An Exploration of Sexual Violence Reduction for Effective Altruism Potential.” Effective Altruism Forum, 11 Nov. 2017. http://effective-altruism.com/ea/1gy/an_exploration_of_sexual_violence_reduction_for/.


“fundraiser/2017/againstmalaria.” Reddit.com/r/neoliberal, modified 18 Apr. 2018, https://www.reddit.com/r/neoliberal/wiki/fundraiser/2017/againstmalaria.


Gabriel, Iason. “Effective Altruism and its Critics.” Journal of Applied Philosophy, vol. 34, no. 4, 2017, pp. 457-473. https://doi.org/10.1111/japp.12176.


Haushofer, Johannes, and Jeremy Shapiro. “The Short-term Impact of Unconditional Cash Transfers to the Poor: Experimental Evidence from Kenya.” The Quarterly Journal of Economics, vol. 131, no. 6, 2016, pp. 1973-2042. doi:10.1093/qje/qjw025.


Karnofsky, Holden. “Effective Altruism.” GiveWell, 13 Aug. 2013. https://blog.givewell.org/2013/08/13/effective-altruism/.


——-, “Thoughts on the Singularity Institute (SI).” Less Wrong, 11 May 2012. https://www.lesswrong.com/posts/6SGqkCgHuNr7d4yJm/thoughts-on-the-singularity-institute-si.


Kastrenakes, Jacob. “Elon Musk donates $10 million to keep robots from murdering you.” The Verge, 15 Jan. 2015. https://www.theverge.com/2015/1/15/7551687/elon-musk-donates-10-million-to-prevent-killer-ai.


MacAskill, William. Doing Good Better: How Effective Altruism Can Help You Help Others, Do Work That Matters, and Make Smarter Choices About Giving Back. Penguin, 2016.


Marcus_N. Comment on Kathy Forth, “An Exploration of Sexual Violence Reduction for Effective Altruism Potential.” Effective Altruism Forum, 14 Nov. 2017, 8:02 a.m., http://effective-altruism.com/ea/1gy/an_exploration_of_sexual_violence_reduction_for/cm4.


Matthews, Dylan. “I spent a weekend at Google talking with nerds about charity. I came away…worried.” Vox, 10 Aug. 2015. https://www.vox.com/2015/8/10/9124145/effective-altruism-global-ai.


Reich, Rob. Response to “The Logic of Effective Altruism.” Boston Review, 1 Jul. 2015, http://bostonreview.net/forum/logic-effective-altruism/rob-reich-response-effective-altruism.


Roser, Max, and Esteban Ortiz-Ospina. “Global Extreme Poverty.” Our World In Data, modified 27 Mar. 2017, https://ourworldindata.org/extreme-poverty.


——-, “Happiness and Life Satisfaction.” Our World In Data, modified May 2017. https://ourworldindata.org/happiness-and-life-satisfaction.


Saunders-Hastings, Emma. Response to “The Logic of Effective Altruism.” Boston Review, 1 Jul. 2015. http://bostonreview.net/forum/peter-singer-logic-effective-altruism.


Singer, Peter. “The Logic of Effective Altruism.” Boston Review, 6 Jul. 2015. http://bostonreview.net/forum/logic-effective-altruism/emma-saunders-hastings-response-effective-altruism.


Skelton, Anthony. “The Ethical Principles of Effective Altruism.” Journal of Global Ethics, vol. 12, no. 2, 2016, pp. 137-146. doi:10.1080/17449626.2016.1193552.


Sundaram, Anjan. “Rwanda: The Darling Tyrant.” Politico, Mar./Apr. 2014, https://www.politico.com/magazine/story/2014/02/rwanda-paul-kagame-americas-darling-tyrant-103963.


Taylor-Robinson, D. C., N. Maayan, K. Soares-Weiser, S. Donegan, and P. Garner. “Deworming drugs for soil-transmitted intestinal worms in children: effects on nutritional indicators, haemoglobin, and school performance.” Cochrane Database of Systematic Reviews, vol. 7, no. CD000371, 2015. doi:10.1002/14651858.CD000371.pub6.


“Top Charities.” GiveWell, accessed 20 Apr. 2018, https://www.givewell.org/charities/top-charities.


West, Cornel. Editorial. “Ta-Nehisi Coates is the neoliberal face of the black freedom struggle.” The Guardian, 17 Dec. 2017. https://www.theguardian.com/commentisfree/2017/dec/17/ta-nehisi-coates-neoliberal-black-struggle-cornel-west.


World Health Organization, “HIV/AIDS”. 15 Feb. 2018, http://www.who.int/news-room/fact-sheets/detail/hiv-aids.


——-, “Malaria”. 20 Apr. 2018, http://www.who.int/en/news-room/fact-sheets/detail/malaria.

Leave a Reply

Your email address will not be published. Required fields are marked *