Should donors make commitments about future donations?

By Owen Cotton-Barratt @ 2016-08-30T14:16 (+18)

As donors, we have choices about which charities to give to. We tend to think these choices are important. We also have choices about whether to give today or save to give in the future.

If you come down on the side of giving now, one question is: should you also make commitments today about where you will give next year?

On the face of it, this seems like a pretty bad idea. By waiting until next year, you maintain option value. By then, it wouldn’t be too surprising if you had new information which changed your view about where could best make use of donations in that year. You always have the option of giving to the original charity, so making a commitment seems to be dominated by not doing so.

Despite this, I think it sometimes makes sense for donors to make multi-year commitments. I think this is particularly true when the value of money is diminishing noticeably over time, which is precisely the type of effect which can drive us to prefer giving now over giving later, and is often claimed to be true of EA organisations. The reason for this is that a commitment not only locks in the future money flow, but lets the charity know that it is locked in. That in turn allows the charity to plan around it.

Here are two things which I think are often believed about EA orgs:

1. The haste consideration means that resources (including money) are significantly more valuable this year than next year.
2. It is good practice for them to keep around 12 months reserves, in order to reduce attention being pulled by financial uncertainty, and allow them to make long-term plans.

If these are both true, then a certain commitment of money in 12 months would be significantly more valuable than just giving the money in 12 months time. Since it is certain, the charity could effectively treat it as part of its reserves, freeing up money from the reserves to spend today. But money spent today is significantly more valuable than money that would be spent in 12 months.

What’s going on here? By making a commitment, you’re effectively borrowing from your future self in order to fund more giving today. You borrow at the same rate the charity could save at for its reserves. If you already thought that giving now was better than saving to give later, this makes such commitments look extremely attractive, at least up to the point where the charities have their future reserves effectively covered by commitments.

It’s actually not as good as that. A certain commitment would reduce financial uncertainty for the charity, but increase financial uncertainty for you as an individual. Particularly if you are donating out of your income, if you were to lose your job, you could find that meeting the commitment required impoverishing yourself, or to be outright impossible.

Because of this, charities can’t treat commitments as certain, only likely. And this somewhat reduces their value for planning, as the point of the commitments substituting for reserves in the first place was to reduce financial uncertainty. Still, a good faith commitment can reduce total uncertainty, if the reasons that it would fail aren’t too correlated with the reasons other sources of income would dry up. For example, a donor getting sick or losing their job would probably be uncorrelated, but deciding to renege on the commitment because other giving opportunities look better could be correlated with other donors deciding not to give.

This uncertainty could potentially be reduced even further, if someone was willing to underwrite others’ commitments. Such an underwriting could provide even more of a help to planning (and therefore move spending forward in time), with little expected outlay.

If this is a good idea, how far should it be taken? Would five-year commitments be better than one-year ones? Ultimately there is a balance, as planning horizons are only so long. After one or two years the force of the argument based on keeping reserves is diminished. There could still be benefits in some cases if a longer commitment could allow confident planning of a longer-term project.

There is also a question about what proportion of future donations individual -- and the community more broadly -- should commit. The costs of committing the first half of future donations are smaller than the costs of committing the second half, while the benefits are typically at least as large for the first half. The costs are smaller for two reasons. First, because the increase in personal financial uncertainty matters less. It is less likely (and perhaps significantly less likely) that you will not be able to maintain at least half of current giving than that you will not be able to maintain current giving levels. Second, because it is extremely valuable that we keep some funds open and flexible to be able to respond to excellent giving opportunities that arise. This is crucial at the level of the community as a whole, but can also apply at the level of individual donors, to reduce friction if those individuals are the ones who discover the giving opportunities.

Given all of this, here is my current view. Donors should be particularly excited about making commitments if they give to charities whose opportunities for creating value are shrinking significantly over time. For small donors, the overhead costs of having conversations about how serious commitments are may be higher than the value the commitment generates. They should therefore not attempt to make hard commitments. They could still create extra value by unilaterally making soft commitments to charities about some fraction of their donations for next year, at the time they make their donations. Larger donors could create substantial value by making commitments about a fraction of their future donations. They could assess how much this would help the organisations they are donating to by having a conversation with them, which would also help the recipients of the commitments to gauge how much they can trust the commitments. I think that the basic argument for this conclusion is moderately solid, but perhaps I have missed or underweighted some important consideration. I’d be interested in thoughts.

Disclaimer: I am employed by the Future of Humanity Institute, and have previously been employed by the Centre for Effective Altruism. The views expressed are my own, not my employer's.

Acknowledgements: this post was inspired by conversations at and after EA Global. Thanks to Kerry Vaughan for suggesting I turn it into a post. Thanks also to Michael Page and Stefan Schubert for helpful comments on a draft. Errors of analysis remain my own.