Singer notes that an unstated presupposition in the earlier chapters entailed the possibility of rectifying extreme suffering. Now he asks the practical question of whether this really is possible. He relies on the work of Karnofsky and Hassenfield, who did significant research into charities they were interested in funding. Karnofsky and Hassenfield inquired into many charities that were not able to provide them with detailed financial information, which they found shocking; individuals and organizations had been funding agencies without any insight into whether those agencies were effective. They eventually realized that it wasn’t that the agencies were being cagey about their data but rather that they didn’t even have data. Singer discusses the downsides of focusing on administrative expenses as if agency effectiveness were a function of this. Karnofsky and Hassenfield’s institute GiveWell “is a nonprofit dedicated to improving the transparency and effectiveness of charities” and therefore concerned with finding the exact information Singer needs for answering whether a modest sum can save a life (84).

Singer discusses historical examples of global aid that were massively successful, like the 1967 campaign by the World Health Organization to eliminate smallpox. Determining whether these campaigns were as effective as possible is another question, and Singer appreciates the complexity of the issue.