Just wanted to note that this does not mean that you should enter any donor lottery if you have economies of scale. (Not that anyone is saying this.) For example, if a terrorist group needs $100,000 to launch a devastating attack, but won't be able to do anything with their current amount of $10,000, you probably shouldn't enter a donor lottery with them.
I think you have a typo in your post title.
It was meant as mediocre word play on the idiom 'dining with the devil' and 'donating'.
[Cross-posted from https://www.col-ex.org/posts/devil-donor-lottery/]
Effective altruists have proposed and promoted donor lotteries. Briefly, in a donor lottery, donors pool money for charitable contribution. They're given lottery tickets in proportion to their contributions. The winner of the lottery gets to decide where the pool of charitable funds should be donated. For example, two people each set aside $1,000 for charitable contribution. They then make an agreement and the winner of the coin flip gets to donate the whole $2,000 to the charity of their choosing.
The primary claimed advantage to these lotteries is economies of scale. It probably doesn't make sense to spend 100 hours picking the absolutely perfect charity when you're only disbursing $1000 as a solo donor. It probably does make sense to spend the time to make the decision carefully when disbursing $100,000 as the winner of a donor lottery. Other economies of scale include things like: improved chance of response when asking charities for more information, can fund projects with threshold cost (i.e. we can't even start the project unless we receive at least $XX,000 dollars).
Though the original proposals all highlight that the primary benefit of donor lotteries is economies of scale, I've heard many in the EA community concerned about alignment of other donors: "What if I don't trust others in the lottery to do thorough research?", "What if others in the lottery prioritize different cause areas?".
Though early advocates address these concerns briefly, subsequent discussions suggests it might be useful to address these concerns more vividly.
I contend you should be willing to enter a donor lottery with the devil (this is the weirdly honorable Faustian devil that respects all his compacts). To see this, we just proceed systematically.
Suppose you and the devil each contribute $10,000 dollars. This is your life savings while the devil merely tosses in a carbuncular carbuncle from his pit of gems and anguish. You flip a fair coin.
Suppose you enter the lottery and win: You've doubled your money. Because we suppose economies of scale at relatively small amounts like this, the good you can do is more than doubled versus abstaining from the lottery and donating the initial $10,000. On the other side of the bet, the devil is forced to bankroll evil with a mere $X trillion - $10,000. Given the devil's vast wealth, his plans are unconstrained by funding and the loss of $10,000 represents the evaporation of a single drop in an endless sea of agony. He goes on to do almost exactly as much evil as he would have done without the lottery.
Suppose you enter the lottery and lose: You've lost all your money and do no good whatsoever. The devil laughs in triumph, but his heart isn't really in it. Because, again, the money means nothing to him. He's working on higher levels in Maslow's hierarchy of needs.
Finally, suppose you abstain from the lottery: You spend your $10,000 dollars to do good, the devil spends his $X trillion to do evil, and the world hangs in the balance.
Now we can calculate the expected value. The devil does just as much evil in each of the three outcomes. If you abstain from dealing with the devil, you do X good. If you enter the lottery, there's a 50% chance you do > 2X good and a 50% chance you do 0 good. The expected value of entering the lottery is then > X.
Obviously, this isn't actually the decision that confronts you. But it highlights that the crucial consideration for donor lotteries is economies of scale. You should only worry about your fellow lottery participants insofar as you think they are significantly misaligned with you and have significantly greater economies of scale than you do. If either of these factors is small, you have little to worry about in this regard.
I am not sure about whether your usage of economies of scale already covers this, but it seems to make sense to highlight, that what matters is the marginal difference of the money for you and your adversary. If doing evil is a lot more efficient at low scales (Think of distributing highly addictive drugs among vurnerable populations vs. Distributing Malaria nets), your adversary could be hitting diminishing returns already, while your marginal returns increase, and the lottery might still be not be worth it.
Yup, I hope the examples make that clear, but the other descriptions could do more to highlight that we're interested in the margin.