Enemies of the bad, enemies of the good?
— Blog Post — 4 min read
Apparently, while I have been in Sitapur, a lot of people in the U.S. have started paying attention to the deaths of so-called “invisible children”: child soldiers in Uganda. I don’t know much about child soldiering in Uganda or elsewhere (if I wanted to, I might ask my friend Tricia’s boss Chris Blattman). Nevertheless, this is going to be a blog post about what is apparently called the “Stop Kony” movement. But I warn you: I’m going to take the long way around the fence.
Consequentialism, in my rough formulation, is the notion that the goodness or badness of a potential course of action depends on its reasonably expected consequences for suffering and joy (as opposed, for example, to what a universal set of rules tell you, or to a notion of what a good person would do). Too briefly: do what will be best.
Increasingly, economists are recognizing deliberation costs – the costs of thinking something through – and they have long written about transaction costs, costs of gathering information, and costs of changing prices. So, economists sometimes permit that it is optimal not to optimize. Long ago, Herbert Simon famously proposed that we instead satisfice: that we choose something good enough.
Unfortunately, sometimes we consequenialists miss this point. If your goal is to do the best thing for the balance of suffering over joy, you can get stuck deep in the mud while you try to sort out exactly which thing is best.
My sister Allie read a story about the importance of pulling drowning children out of ponds at Diane and my wedding. But, in Cosmopolitanism, we have Kwame Anthony Appiah (you may have read my praise of his writing on honor in an early blog post) critiquing this fable, ostensibly from within its own consequentialist logic. “Saving the child may be preventing something bad; but not saving the child might, for all we know, prevent something worse. After all, shouldn’t I be busy about saving those hundreds of thousands of starving children? And wouldn’t selling my suit raise a few hundred dollars? And wouldn’t ruining it mean I couldn’t raise those dollars? The principle [of optimal consequentialism] says that, if this kid right here has to drown for me to save my suit for sale so I can save, say, ninety other children, so be it; though it also leaves me free to let the ninety die, if I can find something worse to prevent.”
For a person who likes to spend time with economists, it’s easy to spot optimization taken too far. Neither an economist nor a philosopher is necessary to invoke the old advice not to let the perfect be an enemy of the good.
But, these days, development economists – flush with apparently precise results from scientific experimentation – often talk about cost-effectiveness, about finding the best thing to do. I am included. I’ve been telling everyone I see that my initial estimates are that the Total Sanitation Campaign prevented an infant death at an average of $2,000 to $3,000, close to as inexpensive as these numbers realistically get. A few years ago at J-PAL, I helped write briefcases with bar charts comparing the estimated cost-effectiveness of one intervention to another. Spend money on deworming, not scholarships: you get much more bang for your buck.
And that has brought me back to stopping Kony. I found out about this campaign when I got my periodic email update from GiveWell.org , a “charity evaluator” aiming to “find the best charities we can” (disclosure: I think they're broadly on to something; I’ve visited their offices and offered to help). Their blog contained a post that I could see myself writing, depending on the day: while stopping Kony would accomplish much good, trying to stop Kony may or may not, and instead trying to fight malaria -- using proven strategies – would almost certainly save many, many more lives.
Commenters were predictably irate. How dare they criticize such a successful humanitarian movement? “I am offended by this... I was deeply moved by the video and have posted it on my FB page.” (I would love to get some irate comments around here, but all comments ever did was send spam to Avinash’s email, so we turned them off. Suggestions?) Many of the comments missed GiveWell’s point, accusing them of wanting publicity for “their cause.” They are not a malaria organization; they analyze research in an attempt to figure out where donations would do the most good, and try to point people in that direction.
Now, it’s almost surely a good idea to discourage contributions that actually do more harm than good, or that are complete wastes of money. There are a lot of these. But, in this big, disaggregated world, full of deliberation and transaction costs and wacky decision-making, it seems to me to be an open – and largely empirical – question when and how to discourage things that help in favor of things that help more.
For starters, as surmountable as I think it is, there is the Appiah problem: where does it end? The post’s headline claims that “The worst killer of invisible children is not Joseph Kony,” but it isn’t malaria, either. Diarrheal disease killed over twice as many children in 2000 as malaria did – over 2 million for diarrhea, compared with 915,000 – and has very bad effects on the children who survive. But I don’t begrudge somebody with a good plan to save the 915,000.
The bigger issue -- as Henry Sidgwick recognized in the 19th century and other consequentialists have, too – is that, people being people, there are tricky ethical implications of what we say as well as what we do. Should we discourage one good anti-poverty investment (whether or not stopping Kony is one) in favor of a seemingly better investment? It depends.
- Is fighting malaria really an opportunity cost of stopping Kony? Are they alternatives? Could people be persuaded to move money from one to the other?
- Is the movement to stop Kony generating new money to help poor children and new interest? If so, it is sticky? By that I mean, do people come for the child soldiers and stay for the malaria? I doubt there is much of a lasting, general effect on interest in poor children, but I would be interested to know.
- What responsibility do we people in the poor children industry have – individually and collectively – for the conversation and rhetoric that people outside of our industry hear? There is a lot of nonsense. And despite everybody’s claims of effectiveness, a lot of people think that money for the poor is money down the rat hole. How much do people really even think about what works when they decide whether to give? Is this important?
Since this is supposed to be a blog about research, I’ll end with the suggestion that I think there are important practical and theoretical questions here. Maybe not the best, but very good ones. If only I had a comment section to hear about them!