This post is my contribution to the December edition of Figuring Good Out, the effective altruism blogging carnival. This month’s topic is “blind spots.”
Thanks to Ruthie for sparking the discussion that led to this post and for reading over a draft.
My favorite “effective altruist blind spot” story goes like this.
Last August, Ruthie and I attended the Effective Altruism Summit reception. Ruthie still didn’t have a good sense of the community at that point, so she was asking other attendees about it.
“What kind of people are in your EA group?” she asked him.
“Oh, all different kinds!” he replied. “Mathematicians, and economists, and philosophers, and computer scientists…”
It didn’t seem to occur to the fellow that these were all basically the same kind of person.
I’m actually not going to talk about diversity as a blind spot of effective altruism, though. I don’t think it is, per se: while we’re not very diverse right now, I don’t think it’s for want of caring. Plenty of people I’ve talked to agree that the movement would be better off as less of a monoculture. The blind spot is subtler than that.
It starts with the fact that few of these people feel like they have levers to do anything about it. They’re all busy getting good grades, or earning to give, or climbing the academic totem pole, or something else that sucks up all their time and doesn’t put them in a very good position to reach out to people different from themselves.
This sounds pretty reasonable for each individual person. But looking at the big picture, it seems like almost nobody is taking active steps to interest people outside the traditional EA demographics. And on the margin, more of that seems a lot more exciting than better grades or higher expected income.
I think that people are (correctly) asking themselves the question, “would I be better at expanding EA’s diversity than other EA-folk?” But then I suspect that they’re comparing themselves to hypothetical EA-folk modeled on their most extroverted friends, with the widest social circles. And they decide that some more extroverted EA-folk should be the ones to take this on.
But precisely because effective altruism isn’t very diverse right now, the movement doesn’t actually have those incredibly extroverted socialite people! Just a lot of very studious philosophers and mathematicians who are busy with college or work. And that’s where our coordination problems come from. People should definitely be thinking about their comparative advantage—but they need to make sure they’re comparing to the right set of people.
The Centre for Effective Altruism is hiring for five positions right now.
I know that they’ve previously mentioned not being able to hire as fast as they want to. I don’t know if this current round will go similarly or not—things seemed to be looking up recently.
I do know that CEA is offering a $1,000 referral bonus, which is about 4-5% of their typical starting salary. That’s pretty beefy for a referral bonus from a nonprofit with an already-wide media reach.
This is surprising: it suggests that perhaps not enough aspiring effective altruists want to work for the Centre for Effective Altruism. I can think of a couple possible reasons for this:
Everyone thinks that replaceability means their marginal contribution to CEA is zero.
Everyone thinks that the work being done by the effective altruism movement’s most central organization is, against all odds, not very effective.
Everyone thinks that although the work they would do at CEA is valuable, it wouldn’t be their comparative advantage, so they should let someone else do it.
I don’t think (1) is very likely, because CEA already has a track record of being unable to find as much talent as they want; people should realize by now that working for them is not replaceable. Reason (2) seems doubtful as well; people generally seem to support CEA’s mission in other ways, e.g. by donating money or treating its employees as high-status.
I think the last reason is the most likely. But if so many people are following their comparative advantage that CEA can’t fill the positions they need, at least some of those people must be wrong!
When Nir Eyal first asked me to help run Harvard College Effective Altruism (then Harvard High-Impact Philanthropy), I remember being a little puzzled. Shouldn’t there be someone else more qualified to do this? Because I was pretty sure I’d be terrible at it.
Well, I was half right. “HHIP” spent roughly its first year flailing around in utter disarray. We organized a couple talks and dinners, but only in the loosest sense of the word “organize” (and “we”). There was no continuity and nothing to give people a reason to actually go to our events. Probably the only thing that saved our attendance was my inability ever to shut up about effective altruism, which at least got some people to go to one talk to see what I was blabbing about.
Eventually, though, things smoothed out, largely thanks to John Sturm starting to help with the organizing. Now HCEA is up to a board of six or eight people—some of whom might even have actual organizational skills. Their events are getting so numerous that at one point they had trouble making sure none of them overlapped.
I certainly wasn’t the best fit for running HCEA. Many other student groups, started by more proficient leaders, have had a much easier time getting started. But I’m still glad that I ignored my comparative advantage and worked on HCEA. It was important that I was there at the beginning to get it limping along so that John, and later our other champion organizers, could find the group and join it.
I’m not trying to argue against the whole edifice of comparative advantage. For one thing, that would be pretty hypocritical of me, as someone who chose earning-to-give in the software industry over a trial period with GiveWell or the chance to apply to CEA. More broadly, I think decisions driven by comparative advantage are often really good—certainly better than decisions driven by uninformed gut feeling, status-seeking, random guessing, or other common decision-making modi operandi.
But I do think we need a more nuanced understanding of the comparative advantage heuristic, where it succeeds and where it fails or causes coordination problems. I haven’t totally clarified for myself how to think about comparitive advantage. But hopefully these case studies are at least helpful in getting started.