Loading
GIVING IS BIGUSAYour browser does not support the element. business. In 2023 Americans alone handed $557bn to charities, according to the Giving annual report. So identifying which charities are the most efficient in terms of good done per dollar given is important. GiveWell, a charity evaluator, tries to do just this, and currently recommends giving to four worthy organisations. How is this recommendation put together, and how good is it?Determining which charities get more bang for their buck comes with challenges. One is data. Any rigorous assessment of efficiency requires someone to catalogue both money spent and outputs achieved. It also requires data on how outputs (such as teacher training) translate into outcomes (such as learning).Any effort to assess how different charities fare relative to each other must also grapple with the fact that they seek to do different good things: some to cure blindness, others to preserve natural parks. To compare them means these goods must be compared, too—a moral judgment with no correct answer.Despite these difficulties, outfits like GiveWell argue that with sufficiently good data, and reasonable assumptions about moral considerations, it is possible to try to identify the most efficient ways to give. GiveWell uses a mix of academic scholarship, impact evaluations, site visits, reviews of financial documents, interviews with experts and other data to identify its top charities. To compare charities doing different things, it uses a system of moral weights. For instance, doubling the consumption of 100 people is valued roughly the same as averting the death of one person in their 30s. Averting the deaths of young children is valued most highly.No surprise, then, that GiveWell’s four top charities all focus entirely or largely on saving children’s lives. Two focus on preventing malaria, which kills 600,000 people, mostly children under five, every year: the Malaria Consortium delivers preventative medicine, at a cost of $7 per prevented infection; the Against Malaria Foundation delivers bednets, at about $5 per net. The other two give vitamins and vaccines: Helen Keller Intl delivers vitamin A supplements (about $2 per child per year); New Incentives gives cash handouts for child vaccinations ($155 for a full course).How efficient are they? According to GiveWell’s calculations, the number of children’s lives saved by its four favoured charities ranges from 1.6 to 3.1 per $10,000 donated—a solid return on investment.But how does this stack up against other approaches to giving? A natural comparison would be with the practice of simply handing over money to the very poor. This is also the comparison favoured by GiveWell. GiveDirectly, a charity that despite the name is not related to GiveWell, does just that: for every dollar donated to it, 80 cents ends up in a poor person’s pocket. Recipients then use it as they see fit, with studies showing rising incomes, better health and lives saved as a result.GiveWell argues that its top picks win out. According to its own calculations, and using its moral weights, its four favoured charities provide between 3.7 and 5.8 times the benefit of GiveDirectly’s unconditional cash transfers, per dollar given. Several past external reviews have found such GiveWell estimates to be reasonable.GiveWell’s analysis might make sense, but those who contest its rankings prefer to focus on its priorities. GiveWell’s moral weights heavily prioritise saving lives over other outcomes. In addition to averting deaths, the top charities do also help many more people avoid terrible, non-lethal disease. But if you care about literacy or political rights as a good in itself, then you would apply a different set of moral weights to charities.GiveWell also does not give any weight to the preferences of those in need; some might rather have more cash in their pockets than better health. Those preferences may be better assessed by smaller local organisations, and better met by simply handing over cash. The debate about which approach is best will go on. Data alone, as GiveWell admits, cannot provide the answer. But it is a good start.
