Comments
Nice post. I like this presentation of the idea from Slate too:
Giving to either agency is a choice attached to a clear moral judgment. When you give $100 to CARE, you assert that CARE is worthier than the cancer society. Having made that judgment, you are morally bound to apply it to your next $100 donation. Giving $100 to the cancer society tomorrow means admitting that you were wrong to give $100 to CARE today.
You might protest that you diversify because you don’t know enough to make a firm judgment about where your money will do the most good. But that argument won’t fly. Your contribution to CARE says that in your best (though possibly flawed) judgment, and in view of the (admittedly incomplete) information at your disposal, CARE is worthier than the cancer society. If that’s your best judgment when you shell out your first $100, it should be your best judgment when you shell out your second $100.
When it comes to managing your personal portfolio, economists will tell you to diversify. When it comes to handling the rest of your life, we give you exactly the same advice. It’s a bad idea to spend all your leisure time playing golf; you’ll probably be happier if you occasionally watch movies or go sailing or talk to your children.
So why is charity different? Here’s the reason: An investment in Microsoft can make a serious dent in the problem of adding some high-tech stocks to your portfolio; now it’s time to move on to other investment goals. Two hours on the golf course makes a serious dent in the problem of getting some exercise; maybe it’s time to see what else in life is worthy of attention. But no matter how much you give to CARE, you will never make a serious dent in the problem of starving children. The problem is just too big; behind every starving child is another equally deserving child.
That is not to say that charity is futile. If you save one starving child, you have done a wonderful thing, regardless of how many starving children remain. It is precisely because charity is so effective that we should think seriously about where to target it, and then stay focused once the target is chosen.
People constantly ignore my good advice by contributing to the American Heart Association, the American Cancer Society, CARE, and public radio all in the same year–as if they were thinking, “OK, I think I’ve pretty much wrapped up the problem of heart disease; now let’s see what I can do about cancer.” But such delusions of grandeur can’t be very common. So there has to be some other reason why people diversify their giving.
I think I know what that reason is. You give to charity because you care about the recipients, or you give to charity because it makes you feel good to give. If you care about the recipients, you’ll pick the worthiest and “bullet” (concentrate) your efforts. But if you care about your own sense of satisfaction, you’ll enjoy pointing to 10 different charities and saying, “I gave to all those!”
Here’s a thought experiment for charitable diversifiers. Suppose you plan to give $100 to CARE today and $100 to the American Cancer Society tomorrow. Suppose I mention that I plan to give $100 to CARE today myself. Do you say, “Oh, then I can skip my CARE contribution and go directly on to the American Cancer Society?” I bet not.
But if my $100 contribution to CARE does not stop you from making CARE your first priority, then why should your $100 contribution to CARE (today) stop you from making CARE your first priority tomorrow? Apparently you believe that your $100 is somehow more effective or more important than my $100. That’s either a delusion of grandeur or an elevation of your own desire for satisfaction above the recipients’ need for food.
We have been told on reasonably high authority that true charity vaunteth not itself; it is not puffed up. You can puff yourself up with thank-you notes from a dozen organizations, or you can be truly charitable by concentrating your efforts where you believe they will do the most good.
Early in this [20th - this was written in 1997] century, the eminent economist Alfred Marshall offered this advice to his colleagues: When confronted with an economic problem, first translate into mathematics, then solve the problem, then translate back into English and burn the mathematics. I am a devotee of Marshall’s and frequently follow his advice. But in this instance, I want to experiment with a slight deviation: Rather than burn the mathematics, I will make it available as a link.
I propose to establish the following proposition: If your charitable contributions are small relative to the size of the charities, and if you care only about the recipients (as opposed to caring, say, about how many accolades you receive), then you will bullet all your contributions on a single charity. That’s basically a mathematical proposition, which I have translated into English in this column. If you want to see exactly what was gained or lost in translation (and if you remember enough of your freshman calculus to read the original), then click here.
Even if we rule out difference-making risk aversion, it can still make sense for small donors to diversify
Also, stochastic dominance doesn't mean you can never make any difference-making risk averse decisions; it just means you should rule out stochastically dominated options first. Or, you can be difference-making risk averse with respect to the quantile distributions (quantile functions applied to the same uniform distribution over [0,1]), which is not sensitive to statewise dependence.
Furthermore, if you're sympathetic to statewise difference-making risk in the first place, this motivation itself may directly be in tension with stochastic dominance before we even get to any contradictions, because it can mean you care about statewise dependence, to which stochastic dominance is not sensitive. So, maybe you should actually just reject stochastic dominance. I'm pretty inclined to keep stochastic dominance, though, since statewise dependence seems weird (maybe even metaphysically dubious) to me.
In past years, I believed that donating to many causes was suboptimal, and was happy to just send money to Givewell's Top Charities fund. But I've diversified my donations this year, partly due to 2., 3. and 4. Some other considerations:
7. From the charity's perspective, a diversified donor base might provide more year-over-year stability. A charity should be happier to have 100 donors paying $1k a year than 1 donor paying $100k, in terms of how beholden it is to its donors.
8. Relatedly, a small charity might have a easier time fundraising if they can use a broad donor base as evidence to larger funders about the impact of their work.
9. Wisdom of the crowds/why capitalism is so good: there's a lot of knowledge held in individual donors' heads about which charities are doing the best work; diversifying allows for more granular feedback/bits of information flow in the overall charitable system.
If I want to deal with my normative uncertainty in the same way as my empirical uncertainty (so by maximising expected choiceworthiness) then I suppose my anti-diversification argument above still holds? My instinct under normative uncertainty is to pick a single option that works somewhat well under multiple moral views that I have credence in, rather than pick multiple options.
Other points seem fair, although don't really convince me for my personal giving. In practice I just give everything to the LTFF.
It was more that DMRA preferences can lead to choosing a stochastically-dominated choice, which then reduces my confidence in holding the DMRA preferences in the first place.
Yes, I think the argument would probably hold under MEC (ignoring indirect reasons like those I gave), although I think MEC is a pretty bad approach among alternatives:
I also think your instinct to look for a single option that does well across views is at odds with most approaches to normative uncertainty in the literature, including MEC, and I think a pretty reasonable requirement for a good approach to normative uncertainty. Suppose you have two moral views, A and B, each with 50% weight, and 3 options with the following moral values per unit of resources, where the first entry of each pair is the moral value under A, and the second is under B (not assuming A and B are using the same moral units here):
Picking just option 1 or just option 2 means causing net harm on either A or B, but option 3 does well on both A and B. However, picking just option 3 is strictly worse than 50% option 1 + 50% option 2, which has value (1.5, 1.5).
And we shouldn't be surprised to find ourselves in situations where mixed options beat single options that do well across views, because when you optimize for A, you don't typically expect this to be worse than what optimization for B can easily make up for, and vice versa. For example, corporate campaigns seem more cost-effective at reducing farmed animal suffering than GiveWell interventions are at causing it, because the former are chosen specifically to minimize farmed animal suffering, while GiveWell interventions are not chosen to maximize farmed animal suffering.
Furthermore, assuming constant marginal returns, MEC would never recommend mixed options (except for indirect reasons), unless the numbers really did line up nicely so that options 1 and 2 had the exact same expected choiceworthiness, and even then, it would be indifferent between pure and mixed options. It would be an extraordinarily unlikely coincidence for two options to have the exact same expected choiceworthiness for a rational Bayesian with precise probabilities.
It isn't obvious to me this is relevant. In your example I suspect I would be indifferent between putting everything towards option 1, putting everything towards option 2, or any mix between the two.
I think just picking 1 or 2 conflicts with wanting to "pick a single option that works somewhat well under multiple moral views that I have credence in".
I can make it a bit worse by making the numbers more similar:
Picking only 1 does about as much harm on B as good 2 would do, and picking only 2 does about as much harm on A as good 1 would do. It seems pretty bad/unfair to me to screw over the other view this way, and a mixed strategy just seems better, unless you have justified intertheoretic comparisons.
Also, you might be assuming that the plausible intertheoretic comparisons all agree that 1 is better than 2, or all agree that 2 is better than 1. If there's disagreement, you need a way to resolve that. And, I think you should just give substantial weight to the possibility that no intertheoretic comparisons are right in many cases, so that 1 and 2 are just incomparable. OTOH, while they might avoid these problems, variance voting and other statistical normalization methods can break down with infinities or violate the independence of irrelevant alternatives.
Ah right. Yeah I'm not really sure I should have worded it that way. I meant that as a sort of heuristic one can use to choose a preferred option under normative uncertainty using an MEC approach.
For example I tend to like AI alignment work because it seems very robust to moral views I have some non-negligible credence in (totalism, person-affecting views, symmetric views, suffering-focused views and more). So using an MEC approach, AI alignment work will score very well indeed for me. Something like reducing extinction risk from engineered pathogens scores less well for me under MEC because it (arguably) only scores very well on one of those moral views (totalism). So I'd rather give my full philanthropic budget to AI alignment rather than give any to risks from engineered pathogens. (EDIT: I realise this means there may be better giving opportunities for me than giving to LTFF which will give across different longtermist approaches)
So "pick a single option that works somewhat well under multiple moral views that I have credence in" is a heuristic, and admittedly not a good one given that one can think up a large number of counterexamples e.g. when things get a bit fanatical.
Ya, I think it can be an okay heuristic.
I guess this is getting pretty specific, but if you thought
then something like splitting between that other extinction risk work and s-risk orgs might look unambiguously better than AI alignment across the moral views you have non-negligible credence in, maybe even by consensus across approaches to moral uncertainty.
FWIW, many grants are smaller than annual GWWC pledge values, so you may in fact sometimes be pretty directly supporting multiple interventions at once. I calculated the median grant size as $8,241.50 for the LTFF in Q4 2022 (based on https://funds.effectivealtruism.org/grants ). EDIT: Marginal returns do probably pretty quickly diminish for most grantees of the LTFF, contrary to the post's assumptions.
I'd also guess ACE's Recommended Charity Fund and GiveWell's Top Charity Fund would regrant marginal donations to multiple charities.
Same for me, but the post's conclusion seems stronger than would be warranted just based on this, even if we ignore non-DMRA reasons.
Yeah this is a relevant point to me.