As you can tell, I'm a fan of GiveWell and Giving What We Can because they're the only people I know of evaluating charities' effectiveness. One problem I have with their method is that it's very measurement-focused, so any type of work that isn't easily measurable just doesn't get considered. People like to point out that the GiveWell approach would have had us medicating polio forever rather than funding research for a cure.
I think there may be causes it makes sense to fund even if we're not sure how effective they will be. Here are some types of charities I've seen advocated by people who are serious about effective donation. In no particular order:
I've seen arguments that if we care about reducing suffering, we should think about the planet's most numerous beings that can experience pain – that is, animals. The charities I know of in this focus on convincing people to stop using animal products or researching how to grow artificial meat. I put some weight on animals' suffering, but less than on people's, and I don't know how to compare the two. So my hunch is I should probably eat less meat, but that this is not the best place for my donation.
Direct health work
The charities recommended by GiveWell and Giving What We Can are all doing public health work in developing countries. As far as definitely saving lives for cheap, this is as provable as it gets.
There's an argument that public health work speeds economic development, because people can do everything better when they're not sick.
I think there's a good argument that development (putting in roads, improving water systems, etc.) is more helpful than more direct medical interventions. This has the appeal of the whole “teach a man to fish” thing.
However, it's hard to tell how effective this is. And the charities that do it, like Oxfam, also do a lot of other stuff that's probably less effective, like disaster relief. So you can't really fund it by itself.
Some people are most concerned about risks that might wipe out the entire human population. This might be things like asteroid strikes, nuclear war, pandemics, or super-smart computers taking over everything. This last one, called a "technical singularity", sounded very far-fetched to me at first, but I've learned more about it and I do think it's a real possibility. Some of these things (like nuclear war) aren't necessarily “existential” in that some people might survive them, but would still be very bad.
There are some sensible measures that have been put in place, like seed banks and telescopes to look for asteroids. There's certainly more we could do. The problem with this type of work is that it's hard to guess how likely the risks are, and it's hard to know how effective we might be at preventing them.
The people I know who fund artificial intelligence research believe a friendly and super-intelligent computer would be especially good because it could not only help us solve current problems, but could also make life much better for people. Their argument goes: “A small chance of a really, really good future is still worth funding.” A counterargument goes, “We have no idea how likely any of this is, so it's better to fund something we understand better."
I don't know how to think about this.
A vaccine for malaria or HIV would be pretty awesome.
Even on a smaller scale, research is essential to knowing how programs are going. For example, there's been a lot of excitement about microfinance, but how well does it work? How could we do it better? Having data on this sort of thing helps charities know what to do and helps donors know what to fund. Innovations for Poverty Action and Poverty Action Lab are the two organizations that I know of.
GiveWell and some existential risk charities could also be considered research charities.
Activism and politics
Changing policies has huge potential to help people who need it. In developed countries, everything from crop subsidies to wars have life-or-death consequences for people around the world. But it's so expensive to swing an election or hire lobbyists that it's unclear whether this is a good use of charity dollars.
A fairly technical look at this question
I've heard a few serious donors say we should save money in case a really excellent cause becomes clear. I find giving is a habit that's pretty easy to maintain once you're in the groove, and I don't want to hoard money only to find I've grown unwilling to give it up once a golden opportunity comes along. So I'm not keen on this one. But I could see donating some to keep in practice and saving/investing some money.
Some more thoughts on this
I can donate money to the best causes I can find, but if I convince other people to do the same, that's even better! (Obviously, that's my goal on this site). Organizations like GiveWell, 80,000 Hours, and Giving What We Can are devoted to getting people to give more, and more effectively. I'm glad they exist, but I'm not sure what they would do with additional funding.
Thinking about these causes is hard. I don't know for sure which is best, and I don't think anyone else really does either. But even thinking about this stuff is moving us in the right direction. Recently, parting at the subway after an evening of discussing optimal charity, a friend's exhausted summation was “People should think and be nice.” And I do believe more thinking and more trying to help will lead us in the right direction. If one of these options, or a new option, becomes more clearly a good choice, we'll be readier to recognize that and pursue it. We should be willing to change our minds.
In the meantime, I intend to make a donation sometime in January. I'm most strongly considering a direct healthcare charity (Against Malaria Foundation) or a research charity (Poverty Action Labs). Do you have thoughts about why I should pick one of these, or something else?