Is legacy fundraising actually higher leverage?

post by Benjamin_Todd · 2015-12-16T00:22:46.723Z · score: 4 (14 votes) · EA · GW · Legacy · 17 comments

It's often been pointed out that legacy fundraising (asking people to make a commitment to give in their will) has a much higher return than other types of fundraising.

On average, $1 spent fundraising raises about $4.

$1 spent on legacy fundraising, however, raises about $30.

This is taken to be a reason to focus more on legacy fundraising.

However, it just struck me that this is wrong. (Apologies if this point has already been made elsewhere).

With legacy donations, you only get the money a long way in the future. If you persuade someone who's 40 with a life expectancy of 80, you'll get it in about 40 years. If you persuade someone who's 20, you'll get it in about 60 years, or perhaps longer. With normal fundraising, you get the money pretty fast - often nearly immediately, or otherwise over a couple of years.

Money in the future is less valuable, so legacy commitments are less valuable.

Let's quantify the effect. Rather than getting a legacy commitment, you could raise money now and invest it in the stock market. If you do that, you could grow it at about 5% per year (real returns).

$1 invested in short-run fundraising generates $4.

Then $4 invested at 5% for 40 years will be worth $28.

And that's about $30 - what you would have got from the legacy commitment.

So it looks like the extra returns of legacy fundraising are fully explained by the fact that you have to wait a long time for the money. It isn't actually a more attractive method of fundraising.


Moreover, there's reasons individual effective altruists and organisations might want to use a discount rate even higher than 5%. If you do that, or if you raise legacy commitments from people under 40, legacy fundraising is going to be substantially less attractive than regular fundraising.


Comments sorted by top scores.

comment by ClaireZabel · 2015-12-16T01:52:21.551Z · score: 9 (9 votes) · EA(p) · GW(p)

I've seen this discussed before, but not formally laid-out.

Do people who legacy fundraise not regularly, uh, strategically engage with more senior individuals in order to generate donations in the nearer future?

comment by Owen_Cotton-Barratt · 2015-12-16T06:40:37.978Z · score: 3 (3 votes) · EA(p) · GW(p)

Right. I'd be surprised if legacy fundraisers spent much time with people under 60, let alone 40. This isn't just about generating money in the nearer future, but also that the older contingent will have more attention for their legacy.

comment by David_Moss · 2015-12-16T12:43:22.463Z · score: 5 (5 votes) · EA(p) · GW(p)

Yeh there are a few stats in Charity Science's shallow review of the area. One study found that 87% of money from bequests came from people who wrote their will in their 70s/80s, and 76% from people writing in their 80s.

comment by Benjamin_Todd · 2015-12-16T14:35:35.604Z · score: 0 (0 votes) · EA(p) · GW(p)

That's useful data. Almost all EAs are under 35 though, so it doesn't help if we want to pursue legacy fundraising.

comment by MalcolmOcean · 2015-12-16T15:08:48.353Z · score: 7 (7 votes) · EA(p) · GW(p)

Hmm... I'll gesture back at the "Effective Giving vs Effective Altruism" thing, and say that maybe while EAs qua "identify as part of EA movement and comment on the EA forum and hang out with other EAs" might be under 35, we might be able to find lots of candidate Effective Givers who are part of a totally different demographic.

comment by David_Moss · 2015-12-16T16:31:50.869Z · score: 4 (4 votes) · EA(p) · GW(p)

Yes and in the review Charity Science distinguish between "usual EA channels" which they note, by default, target mostly young people and "GiveWell, the Life You Can Save, and the direct charities" who are not uniformly young.

comment by Benjamin_Todd · 2015-12-16T22:39:53.261Z · score: 0 (0 votes) · EA(p) · GW(p)

I agree, should have made the comment more moderate. I just mean to say it seems quite a bit harder to design a major fundraising campaign around EA ideas for 70 year olds.

comment by Larks · 2015-12-17T00:02:28.189Z · score: 1 (1 votes) · EA(p) · GW(p)

This also reduces the risk that they will later alter their will.

comment by David_Moss · 2015-12-16T08:18:38.874Z · score: 1 (1 votes) · EA(p) · GW(p)

Yes, that's absolutely right, and one of the first things to bear in mind when thinking (or commenting) about legacy fundraising.

comment by Benjamin_Todd · 2015-12-16T14:34:17.331Z · score: 2 (6 votes) · EA(p) · GW(p)

Hi David,

This comment felt quite snarky to me. I did apologise in case this point had been made before.

I agree the fact that the money comes a long way into the way is obvious, but I've never seen a quantitative examination of it before.

Looking at Charity Science's review, I don't see one. Glancing at the calculation of expected returns of EA legacy fundraising, it looks like no discounting was done at all, so the 18:1 expected ratio is overstated by at least several fold. I expect if they added discounting, it wouldn't look much more attractive than other fundraising methods, and a whole project was launched on the basis of this report. The report also mentions that young people contribute a significant fraction of the EV, but I expect this would go away if discounting was included properly.

comment by David_Moss · 2015-12-16T16:24:21.620Z · score: 4 (4 votes) · EA(p) · GW(p)

My comment was agreeing with Claire's observation that "people who legacy fundraise... strategically engage with more senior individuals in order to generate donations in the nearer future"- nothing to do whether whether your "point had been made before" because it's directly referring to Claire's point.

I'll address these new points you raise about Charity Science's projected ratio in a separate comment if no-one from Charity Science gets there first since they're quite distinct from the point above (after all, you don't mention Charity Science in the OP).

comment by Benjamin_Todd · 2015-12-16T22:36:44.423Z · score: 2 (2 votes) · EA(p) · GW(p)

Sorry I misunderstood. I noticed several downvotes for the post so interpreted your comment as directed at the main post.

comment by Tom_Ash · 2015-12-16T07:59:13.031Z · score: 2 (2 votes) · EA(p) · GW(p)

Is the following another way of making a similar point?

  • 2015 $ raised now could be invested with above-inflation annual interest and then given in 2055.

  • $1000 given in 2015 does more good than $1000 (inflation-adjusted) given in 2055, partly because the former has a 'social rate of return' (which is plausibly greater than 5%).

  • You have to accept this or else accept that it'd be better to invest 2015 $ raised now and only give the money in 2055. (Absent some other strong reason not to, like thinking that there will be fewer low-hanging fruit then).

comment by Tom_Ash · 2015-12-16T17:31:48.567Z · score: 0 (0 votes) · EA(p) · GW(p)

Explications of 'social rate of return':

"Improving health today has significant flow-on effects to the future, by ensuring people and their children are healthier, and better able to learn, work and make their countries prosperous. The earlier in time these health improvements are achieved, the more people benefit from these flow-on effects."

"Interestingly, just as the returns on your investment would compound over time, donating to a charity now may also yield compounding benefits. In the case of donations to global poverty charities, the benefits that accrue to those immediately helped by a particular intervention (e.g. those freed from intestinal worms by ones donation to a deworming charity) also compound at a social rate of return. The idea here is that the direct benefits of the intervention (e.g. improvement of an individuals health) lead to societal knock-on benefits that compound over time (e.g. increased productivity and likelihood to reproduce, which may themselves have beneficial economic and non-economic spill-over effects on the rest of society)."

comment by Owen_Cotton-Barratt · 2015-12-16T06:38:27.359Z · score: 0 (0 votes) · EA(p) · GW(p)

Do you know if they actually work out the ratio this way? (protected legacies from people who sign up)

I'd have guessed it was calculated by comparing one year's legacies received with outgoings. This has some other issues, like being an average rather than marginal effectiveness.

comment by Ervin · 2015-12-16T19:25:16.239Z · score: 1 (1 votes) · EA(p) · GW(p)

By 'they' do you mean Charity Science? Is that who Ben Todd's talking about?

comment by Owen_Cotton-Barratt · 2015-12-16T20:54:45.814Z · score: 0 (0 votes) · EA(p) · GW(p)

I meant whoever produced figures for the ratios. I guess I was envisioning the Institute of Fundraising, but only because I recently saw some figures from them.

Edit: It seems Charity Science have an overview of this linked in another comment. They have a summary of other studies, but I don't know the methodology for these other studies. They have a Fermi estimate themselves (and they don't discount), but the error bars are large and I think this is more of a sanity check on the other figures -- it's not clear discounting should change the overall conclusion too much, except perhaps pushing against a focus on young people.