Creating a donor-advised fund lottery

Summary: In a previous post I discussed the construction of charity lotteries, which let donors who think that the effectiveness of their donations has increasing returns to scale convert small donations into a small chance of donations large enough to exploit scale economies. However, transaction and coordination costs pose a barrier to individual users: there are scale economies to setting up charity lotteries. Effective altruists looking for projects could set up a charity lottery with low transaction costs using a donor-advised fund to provide easy access to small donors.




In a previous post, I discussed ways individuals could use lotteries, derivatives, and giving clubs  to exploit economies of scale in charity.

There are fixed costs for individual donors to find donation targets: it is hard to justify spending 1000 hours on research for a $1000 donation. Publishing the results of one's research increases its value, but some personal research is not transferable, e.g. reading enough GiveWell material to come to trust GiveWell's judgments.

There are also donation opportunities that require minimum scale, such as funding the creation of a new startup organization or new position.

If one believes that these economies of scale would on balance let you do more than n times as much good if you were donating n times as much, then you would benefit from a charity lottery where each dollar you spend buys a 1/n chance of being able to direct $n to the charity of your choice.

The transactions costs of buying a chance of a large donation

As discussed in the earlier post, using commercial gaming companies to trade a cash stake for a chance of a larger one involves a substantial 'haircut' due to gaming company profits and, more importantly, taxes. Haircuts are especially large in many cases for government monopoly lotteries.

Structuring bets using securities such as derivatives requires nontrivial initial effort to set up, which may easily defeat the purpose for small donors.

I previously mentioned some setups to avoid haircuts:

For example, say that A is donating $500,000 to a fairly scalable intervention like Against Malaria Foundation or GiveDirectly, while B is donating $1,000 but thinks her best option requires at least $100,000 to take advantage of economies of scale. Then they can use a random number generator and have B donate to A's charity, in exchange for A donating $100,000 to B's with probability 1/100. 

A group of small donors might also contribute to a common donation pool, with each dollar contributed to the pool buying one "ticket" in a lottery to select the donor who gets to choose the allocation of the entire pool. The accumulation of funds could be managed with a crowdfunding coordination mechanism along the lines of Indiegogo.

A donor club where several donors use a random number generator to decide which of them will get to choose where the whole club will donate has startup, search, and coordination costs.

My suspicion is that to make the practice widespread a simple turnkey solution in a regranting institution is required.

Creating a low-friction method

If an effective altruist wanted to make charity lotteries accessible, and market research indicated demand, they could try the following:

1. Partner with an existing nonprofit (or, less attractively, create one).
2. Arrange for easy online donations earmarked for the charity lottery.
3. Over the course of each year, accept donations to the account.
4. If the nonprofit is not itself a donor-advised fund, it can place the donations in a DAF.
5. At the end of the year, use trusted public sources of randomness to perform a lottery weighted by each donor's proportional contribution to the funds in the DAF at the time.
6. Invite the winner to recommend a qualifying charity for the funds amassed in that year (possibly after a substantial delay for research).

Once this organization exists, recommend using it whenever someone suggests something like the following:

This expenditure is also pretty lumpy, and I don’t expect them to get all their donations from small individual donations, so it seems to me that donating 1/50th of the cost of a program manager isn’t as good as 1/50th of the value of a program manager.

An alternative would be to fund organizations like Open Philanthropy (such as by displacing its donations to grantees) and let them act as charitable venture capitalists, but if something like this took off it could enable more distinctive crowdfunded projects.