The Life You Can Save: Acting Now to End World Poverty

Peter Singer

Book cover

A very quick read and a compelling argument. Singer argues that middle-to-upper-class people in developed countries (and upper class people in developing countries) have a moral obligation to give significantly more than we do to help the poorest people in the developing world. Although it is easy (and fair) to argue over exactly how much should be required of us, Singer pretty convincingly argues that, using any reasonable standard, the number should be much higher than it currently is. Singer’s “reasonable standard” is essentially that, at minimum, we should be giving as long as it doesn’t make a significant difference to our quality of life. Although many Americans do give a reasonable amount to charity, very little of it is directed to the poorest people in the world (with most going to our own religious, educational, and cultural institutions). The reason for focusing on the global poor is not merely that they have a worse quality of life than the poor in our own countries (although that is true), but also that interventions can be made much more cheaply for them, so that any given amount can go much farther in alleviating human suffering.

Singer’s discussion of the relevant considerations is both straightforward and fascinating. One of the biggest takeaways for me is that, really within the last 20 years or so, it has become much more feasible for us to verify that money given is truly (a) getting where it’s meant to go, and (b) having a meaningful positive impact on people’s lives. This is due to the enhanced level of analysis and transparency championed most prominently by GiveWell, as well as by the proliferation of randomized controlled trials championed by organizations such as Innovations for Poverty Action. Our current cultural ideas about giving are inherited from (say) the 18th through 20th centuries, when this type of verification was not possible. The reasonable result of this information shortage was a focus on giving “close to home” and “in the community,” where it was easier to verify impact. A less obvious corollary is that this local focus effectively lowers the equilibrium expected level of giving, because the gap in life circumstances between the rich and the poor in a single community, along with the associated effectiveness multiplier, is by and large going to be much smaller than the gap between the global rich and global poor. (Singer allows that it may be reasonable for us to value our own well-being more highly than that of others, but not by very large multiples.) In addition, issues of desert (whatever one may think of them overall) are much less plausibly raised in a global context than in a local context–one can hardly argue that someone in a developing country who lacks access to clean drinking water just isn’t working hard enough.

Indeed, Singer’s focus in this book is less on changing individual minds (though that is important), and more on shaping the cultural mores around giving so that a higher level is expected of the well-off. I think he takes an admirably pragmatic view on this matter. One problem with the moral issue he raises is that the obligation can feel practically unlimited, and therefore overwhelming–I could give away virtually all of the money I make and still be very much better off than the people it would be going to help. Singer recognizes that ultimately the “right” level is something that each of us will have to decide individually. Instead, he proposes a societal standard that is significantly higher than the status quo, yet low enough that it is difficult to argue credibly that it’s too high. (In other words, it’s difficult to argue that the amounts he proposes would meaningfully lower the givers’ well-being–in fact, given what we know about the psychology of giving, it’s quite plausible to argue that they would raise it.) It’s roughly as follows (using marginal brackets as with taxes, such that an increase in gross income across a bracket boundary never results in a decrease in take-home income):
-0% of income if you’re below the poverty line, then
-a sliding scale of 1% to 5% of your income between the poverty line and about $100k, then
-5% up to $150k, then
-10% up to about $400k
-and several higher brackets above that

Note that all the way up to $400k, which covers the vast majority of people I know (maybe everyone??), this standard is less stringent than the traditional “tithe” standard–the key distinction being that it is much more specifically targeted toward effectively alleviating the maximum amount of human suffering. Singer based his standard brackets on a rough back-of-the-envelope calculation of the total amount needed to end extreme poverty (as articulated in the UN development goals), spread across the wealthy population of the planet, with some judgment to set the brackets at levels that seemed reasonable.

While it’s easy to pick at any one premise of Singer’s argument, I find it pretty difficult to argue with his general conclusion. For myself, I plan to at least meet Singer’s standard for my own income for this year, as well as to try and “backfill” some of what I would have given if I had been meeting it since starting to work. I’ll see what that feels like, and think about whether I should set a higher personal standard.

I encourage others reading this to do the same, especially if your income is in the 6-figure range (or more!). If my paraphrase isn’t totally compelling, pick up Singer’s book and read it in a weekend. You can check out his website ( to find a list of 10-20 organizations that meet his standards of (a) working on behalf of the poorest people in the world, and (b) having produced strong evidence of their effectiveness. Some of them really do save lives; others do non-life-saving things that can nonetheless have a profound impact on quality of life, such as surgery for blindness or for obstetric fistulas. At the beginning of reading about this general topic, I was really focused on just how much it costs to actually save a life. It’s obviously very difficult to estimate this precisely, but the order of magnitude for the most cost effective interventions (e.g. against malaria) is somewhere north of $1k and south of $10k. But the more I read about it, the less I focused on that specific outcome. Some of the quality-of-life interventions are hugely important and much more easy to price out–for example, something like $350 pays for a cataract surgery. If you gave $3500 (about what Singer recommends for a gross income of $90k), you could know with pretty high certainty that your gift restored sight to 10 people. To me that is even more compelling than a statistical one-ish life saved (although that is also pretty cool too!).

My Goodreads rating: 4 stars