You need to take a stance on population ethics to assess the impact of some charitable interventions

I went to an EAish event tonight. It was put on jointly by the London Giving What We Can group and by Ambitious Impact (the rebrand of Charity Entrepreneurship).

There was a Q&A with a few founders of smaller charities. One of them was LAFIYA, which basically does contraceptive distribution in Nigeria. They were went through CE around three years ago and are now scaling. There was an interesting point in the Q&A where one of the founders said that in regards to population ethics, her charity didn’t take a position and tried to stay away from such a complex ethical topic. I asked a follow up question and asked explicitly how they would evaluate an intervention where giving out contraceptives makes a family 50 QUALYs* better off but prevents the creation of a child who’s life would be on net be 60 QUALYs. I didn’t get a clear answer.

A lot of charitable causes don’t require that you answer complex ethical questions. Some do. I feel like contraceptive delivery is one of the latter. It feel like you can do one of three things

1. Take a position on population ethics and then run the numbers. E.g: My friend Miri also runs a CE kickstarted charity doing contraceptive delivery in Nigeria. She’s a negative utilitarian. Her position on not creating future people is that it doesn’t matter. Hence she can just straight up take the trade of +50 QUALY for the family at the expense of a 60QUALY life not being created.
2. Take some kind of mixed approach where you account for the moral uncertainty. The most conservative way to do this is to run the numbers under both scenarios (creating lives doesn’t matter, creating lives is equally important to making existing people better off) and only do interventions which would be net positive under both sets of assumptions. You can also take a more mixed approach where you assign different credences to each view and do some kind of credence multiplication to give you an expected value-ish moral impact estimate. (There’s a lot more complexity here and inter-theoretic comparison is actually really tricky, but let’s leave that can unopened for now)
3. Pretend that you don’t have to take a position, implicitly take the “Creating net-positive lives doesn’t morally matter” approach.

Now, it’s somewhat predictable that a single charity or founder may do this. Some reasons for this

• Not everyone is a moral philosopher. It may not be obvious that the population ethics question matters here.
• There’s a strong incentive for founders to not look too hard at moral questions where the non-default answer could drastically worsen their cost-effectivness estimates.

I think a stronger negative signal about EA epistemics is that the CEA research paperassessing the impact of LAFIYA’s interventions made the same oversight. I gave it a quick read. I may have missed something, but nowhere did I find any mention of the QUALY cost of the lives prevented by distributing contraceptives.

I guess my closing thoughts here are pretty simple. I think there are a wide range of reasonable moral positions to take here. I think that it’s even possible to take the “potential lives matter” position and still think that contraceptive distribution is good even in the initial example I gave. e.g: Because you believe that women should be allowed to choose whether to bear children, regardless of whether given them that choice would lead to higher or lower world utility. That being said, the fact that this question is not addressed by the charity itself or a CEA research paper leaves a bad impression for me.

Evidence Paper:

*QUALY = Quality Adjusted Life Year