Effective Altruism Forum
EA Forum

Comment Permalink

Peter WildefordJun 1 20157

I'll mention a few of my hangups:

What is the quality of a marginal GWWC member?

I'm personally not convinced yet that GWWC can drive quality new members so cheaply. Asking for 10% is quite a high bar to clear, so I'm a bit astounded (and skeptical) GWWC has apparently been doing so well at recruiting people to pledge this. Especially with the cost per new member going down so low, I'm nervous that the new members are pledging a lot more than they'll actually deliver.

It's really great then that GWWC provides data on the actual total number of dollars donated to top charities, and even greater that this number comes counterfactually adjusted.

However, I see this number is primarily driven by one high net worth donor, and I don't know how sustainable it is to expect to repeat this. Taking out that high net worth donor reduces GWWC to a ~1:2.3 ratio on money actually donated (counterfactually adjusted), which is still good and worth donating to (how often do you get 230% returns?), but lower than I'd hope and makes me nervous.

I notice parallels to the animal activism movement where I also spend a good deal of time -- people there talk all the time about how they can spend just like $15 (or some other ridiculously small number) and get someone to go vegetarian for 4+ years. Just as I think people are rightfully skeptical of that number, I'd be skeptical of GWWC's ability to convince someone to donate 10% for 4+ years based on just $230 of costs. (But I also see evidence against that makes it look like maybe some special people are just intrinsically willing to become EAs when pushed and will then donate a lot...)

What is the value of GWWC's marginal activities?

As I said, I'd still donate on a 1:2.3 ratio, but my core argument is that I'm nervous that marginal member recruitment will be done at diminishing marginal returns and the ratio could drop below 1:1. I may have done a poor job engaging with the arguments you've already made (so feel free to re-refer me). Something I've learned at Charity Science is that networks only run so deep and it can be easy to run out of the low-hanging fruit when trying to achieve scale.

Reasons I may be wrong to be skeptical

Of course, while I do think talk is cheap and am skeptical that many of those who pledge will actually end up giving (just like many people who pledge to go veg don't stick), it is definitely incorrect for me to assume that the value of all these pledges is 0. So I would need to re-weigh for that.

And I guess the size of the ask is so large that it can make up for a high failure rate -- say someone sticks for four years and earns $40K, that's $16K, which is enough that 52 other members could fail to give a penny and GWWC would still break even.

Conclusion

Ultimately, though, I think GWWC is a good donation choice, but it's a very high bar to clear to be "the absolute best possible place I could give money to, including saving to donate later".

And, for what it's worth, I'm confused why GWWC hasn't met it's fundraising threshold yet. I thought the EA community loved meta-charity...

Michelle_HutchinsonJun 11 20152

Benjamin_Todd

Jun 2 2015

Peter, thanks for writing out your reasons - it's really useful. I agree the one large donor point is worrying, but you shouldn't set the donations given by the large donor to zero. There's a reasonable chance GWWC recruits another large donor in the next couple of years and you should take account of that. In fact, because donor size generally follows a power law, the top donor determines the mean, so if you ignore the top couple of large donors, your expectation value is always going to be too low. The even bigger point is that I think it's extremely conservative to set all future donations to zero. Yes, talk is cheap, so drop out rates could be much higher than we imagine, but as your example shows, to get to the point where the value of future pledges is zero requires extreme pessimism. Like on the order of 98% dropping out within a couple of years. There's no evidence people are dropping out at anywhere near that rate. Finally, you're only considering downwards adjustments. What probability would you put on GWWC 'optimistic' scenario in the evaluation doc? I wouldn't be super surprised if the leverage ratio turns out to be over 100. Let's say that's a 10% probability. Then the expected leverage ratio is already at least 10, even if I ignore all the other scenarios. Then there's also the impact that results from GWWC that isn't just money moved (e.g. EA movement building), and the fact that funding now also lays the groundwork for long-term growth, which is where most of the expected value lies.

jayd

Jun 1 2015

To expand on this, the difference between core GWWC (and various layers of the core) and marginal GWWC appears to be a fundamental issue here. In response to a long thread in the other post were people were worrying about GWWC's expansion at some point hitting lower than 1:1 fundraising ratios, Peter put it succinctly: And I expanded:

See in context

Giving What We Can needs your help!

by Michelle_Hutchinson

May 26 20152 min read 35

5

Giving What We Can

Frontpage

We're halfway through Giving What We Can's spring fundraising period, but so far we've only raised 25% of our 2015 target, let alone our stretch goal (which would mean that we could focus on our core activities rather than on fundraising for a whole year).

Thank you so much to everyone who has supported us so far! It means a lot to us, and it will mean a lot to the charities that we support.

If you've considered giving to Giving What We Can, now is a great time to do it. Fundraising takes a considerable amount of time and energy. The longer it takes for us to reach our fundraising targets, the more time we have to spend on fundraising activities, and the less time we can devote to inspiring donations to our recommended charities. With our members contributing an average of over $60,000 in lifetime donations to top charities (time discounted and counter-factually adjusted), every member that we don't convince to join up is around 11,000 bednets that won't be distributed, nearly 80,000 people that won't be treated for Schistosomiasis, or 20 lives that won't be saved.

If we fall short of our fundraising goal, we will not be able to offer a long-term position to our current Director of Communications. He has invaluable experience as a Communications Director for several high-profile Australian politicians, which has given him skills in web-development, public relations, graphic design, public speaking and social media. Amongst the things he has already achieved in his three months with us are: he has automated the book-keeping on our Trust (saving huge amounts of time and minimising errors), very much improved our published materials, written a press release and planned a media push to capitalise on our getting to 1000 members and Peter Singer’s book release in the UK. His wide variety of skills mean that there are a large number of projects he would be capable of doing which would increase our member growth, and we are keen for him to test a number of these. But his first, if we can keep him on, is to optimise our website to make the most of the increased attention effective altruism will be generating over the summer and turn that into people actually donating 10% of their incomes to the most effective causes. In the past we have had trouble finding someone with such a broad set of crucial skills. Combined with how swiftly and well he has integrated into our team, it would be a massive loss to have to let him go and later down the line need to try to recruit a replacement.

We believe that Giving What We Can is one of the most effective places you can donate - we calculate that for every dollar we spend on our operations, around $60 is donated to effective charities. This is a truly amazing return on investment.

You can donate at our CauseVox page or the CEA website.

If you'd like to ask any questions about any of this, please feel free to get in touch with us!

Best wishes from Michelle and the Giving What We Can team

5 Reactions

More posts like this

Comments35

Sorted by

New & upvoted

Click to highlight new comments since: Today at 8:04 PM

Ben KuhnMay 27 20158

Out of curiosity, how many folks have held off donating to GWWC in order to see whether their fundraiser hits its goal without their donations (so that their donation would have been replaceable/had no counterfactual impact)? I'm wondering if we have some kind of coordination problem here.

[pollid:3]

(disclaimer: have not donated yet because I've been planning to take a more thorough look at GWWC, but probably also partly because I'm secretly hoping they'll hit their goal and I won't have to)

Robert_WiblinMay 27 20158

I think this is an understandable thing to do and I could imagine doing the same. But people should know that it comes at the cost that we have to put in more time - perhaps a month of staff time - in order to eventually reach our goal. In addition, there's the stress and uncertainty it creates for us.

This isn't a good reason to give if you actually think your donations would be much more valuable elsewhere, but it is a reason not to always hold out until the last minute.

Ben KuhnMay 28 20150

Yeah, I agree (and sorry if it sounded like I didn't)! But I suspect that this is a part of the coordination problem: each individual donor thinks that for them to hold off won't save GWWC very much time, but it could save them potentially having no counterfactual impact.

Owen Cotton-BarrattMay 28 20155

Thinking that they might have no counterfactual impact sounds like a mistake -- not increasing the amount GWWC raises means that other people didn't give to GWWC, and those are likely to be highly strategic and impact-oriented donors who would probably seek out a different valuable giving opportunity otherwise.

You can reasonably think that you have reduced counterfactual impact, if you think that your personal alternative donation target will beat that of the average displaced donor. By the same token you could reasonably give to GWWC in the hope that this would displace donations, and that those donations would be better-researched than your giving otherwise would be.

Benjamin_ToddMay 30 20151

My guess is that GWWC's marginal donors would likely give to other EA orgs or GiveWell recommended charities, so to the extent you're getting replacement (not likely 100%), it's just going to the marginal EA projects instead.

NekoinentrJun 1 20150

Good point, that sounds plausible - I attempted to capture it in my poll in reply to Ben's but it has so far not reached many people who actually gave to GWWC.

NekoinentrMay 27 20151

Oh I didn't know you could do polls! Testing them out with one with a fuller set of options:

[pollid:4]

LarksMay 28 20151

My main reason for not donating was the thought that my donation wouldn't raise GWWC's counterfactual spending

There are two possible ways of cashing this out. Either of

My donation would cause others to donate less
My donation would cause CEA central to transfer resources away from GWWC

but only one of these is what Ben was talking about.

NekoinentrMay 28 20151

It'd be good to hear from more GWWC donors. So far "not being sufficiently convinced" is in the double digits (perhaps predictable given most Effective Altruists appear to donate to the charities that GiveWell thinks do the most good, and this is the most plausible explanation of that). But only 1 person has voted for something else, "I donated money that wouldn't have otherwise have gone to help anyone else". Not that I didn't want to know people's reasons for not donating and wouldn't welcome more about these, but I was equally curious to understand why people did donate.

Robert_WiblinMay 28 20152

We would be very interested to hear from the people who are not convinced! Is this a cause selection issue, lack of confidence that we can drive up member numbers cheaply, or something else?

Peter WildefordJun 1 20157

I'll mention a few of my hangups:

What is the quality of a marginal GWWC member?

It's really great then that GWWC provides data on the actual total number of dollars donated to top charities, and even greater that this number comes counterfactually adjusted.

What is the value of GWWC's marginal activities?

Reasons I may be wrong to be skeptical

Conclusion

Ultimately, though, I think GWWC is a good donation choice, but it's a very high bar to clear to be "the absolute best possible place I could give money to, including saving to donate later".

And, for what it's worth, I'm confused why GWWC hasn't met it's fundraising threshold yet. I thought the EA community loved meta-charity...

Benjamin_ToddJun 2 20153

Peter, thanks for writing out your reasons - it's really useful.

I agree the one large donor point is worrying, but you shouldn't set the donations given by the large donor to zero. There's a reasonable chance GWWC recruits another large donor in the next couple of years and you should take account of that.

In fact, because donor size generally follows a power law, the top donor determines the mean, so if you ignore the top couple of large donors, your expectation value is always going to be too low.

The even bigger point is that I think it's extremely conservative to set all future donations to zero. Yes, talk is cheap, so drop out rates could be much higher than we imagine, but as your example shows, to get to the point where the value of future pledges is zero requires extreme pessimism. Like on the order of 98% dropping out within a couple of years. There's no evidence people are dropping out at anywhere near that rate.

Finally, you're only considering downwards adjustments. What probability would you put on GWWC 'optimistic' scenario in the evaluation doc? I wouldn't be super surprised if the leverage ratio turns out to be over 100. Let's say that's a 10% probability. Then the expected leverage ratio is already at least 10, even if I ignore all the other scenarios. Then there's also the impact that results from GWWC that isn't just money moved (e.g. EA movement building), and the fact that funding now also lays the groundwork for long-term growth, which is where most of the expected value lies.

Benjamin_ToddJun 2 20155

Two other points.

I'm personally not worried about "networks only run so deep" is a concern. In the early days it's true GWWC spread in large part through the networks of the staff, but I don't think this has been true for the last couple of years.

Finally, there's an element of GWWC can't win going on here. You say that the cost per member has got so cheap you're skeptical that it's real, so not convinced to donate. But I imagine if the cost per member had instead gone up, you'd be saying it's diminishing returns, and also not convinced!

Michelle_HutchinsonJun 11 20152

Thanks very much for your comments Peter, it's very useful to know what people's hesitations are. And thanks Ben, for answering! I just had a couple of quick things to add: With regard to the one large donor - having one such actually makes it more likely that we'd attract others than it would have been otherwise. It doesn't just provide evidence that it's possible - it provides us with opportunities for meeting others in similar networks. Indeed, that has already happened. That impact is then partly due to the original large donor rather than separate things we do. But if attracting that original donor was impact down to GWWC, then presumably the impact of other donors coming partly through their help is similarly so. It feels like you're taking that 1:2.3 number at face value, but it's worth bearing in mind that it leaves out a large number of our routes to impact, including, for example, any good done by donations to non-top charities, any good done by people giving to more effective charities than they would have otherwise, and any donations by non-members other than the one. And of course any money given by members in the future. In terms of whether returns are diminishing or increasing, I think the third impact assessment on the fundraising prospectus is probably the most informative thing to read. It seems like at the moment we're experiencing economies of scale rather than diminishing returns. That could be due to things like it being hard to be a very early adopter of something like giving 10%, but it being much easier to make the leap when there are already a whole bunch of people doing it; it could be because it's easier now we've built up a bunch of infrastructure like a good website. Either way, it does seem to be the case.

jaydJun 1 20152

I agree that GWWC's ratio is probably above 1 with a good deal of confidence (though I haven't done the formal math to evaluate how extreme that statement is). But I think the more compelling argument is that expansion funding on the margin may not have a ratio above 1.

And I expanded:

Yes, there's a huge, huge difference between the impact of GWWC existing as a place where people who wanted to pledge 10% of their income to help those living in global poverty could join others in publicly doing so, and the impact of its marginal funded activities now. GWWC existed as that place before The Centre for Effective Altruism was founded around it as an organisation with donors and a budget supporting paid employees. If minimal resources were spent on creating the basic infrastructure, and I don't know if that's so, but if so, then it had a mega high impact ratio. But it seems wrong to use that to justify keeping on spending more money on more employees doing more marginal projects until the impact from the original resource gets "used up."

Benjamin_ToddJun 2 20150

I replied in the other thread: http://effective-altruism.com/ea/hz/please_support_giving_what_we_can_this_spring/

jaydJun 2 20150

That reply:

Ok this is all fair. I think, however, that a big fraction of the historical impact is due to on-going activity, of the kind that could continue, rather than being all due to the 'set up' generating the stream. And that would mean the historical ratio is a reasonable guide to the future.

This can be hard to see from the outside, but if you look at where new pledgers are coming from, it's often new press coverage or student group activity. Many also only take the pledge after being nudged by someone in person, even if they had heard about GWWC some time before, so there's an important role just talking to lots of people about the pledge. These kinds of activities can be scaled much further.

Continues at http://effective-altruism.com/ea/hz/please_support_giving_what_we_can_this_spring/3vw

zdgroffMay 27 20154

I'm thinking of switching a ~200/mo. donation from SCI to GWWC. Do you prefer a lump-sum donation to a monthly one?

Also, it seems your claim is that movement-building charities and effective giving promotion are under-funded relative to direct charities. Any sense of what ratio would be optimal?

Robert_WiblinMay 27 20155

Hi zdgroff,

Yes I think the claim is that on the current margin and point in time movement building is under-resourced.

Of course this can't hold forever but a) effective altruism is quite new b) it's growing quite quickly c) most people still give to charities that help directly d) we have evidence from our past work that we are earning a good fundraising multiplier. (More on this: http://effective-altruism.com/ea/is/how_valuable_is_movement_growth/.)

If it were me and I were giving under £10,000, I would just find the best option wouldn't split my donations. This also allows me to monitor what my donations are accomplishing more closely. However, I know many donors, even relatively small ones, do split and would be interested to hear their rationale.

arrowindMay 28 20151

Yes I think the claim is that on the current margin and point in time movement building is under-resourced.

Do you think this applies to anything or anyone besides CEA and the people involved in it?

Robert_WiblinMay 28 20150

Yes absolutely, the consideration would apply to anyone who had good reasons to think they would get many more people taking effective altruist actions.

However to my knowledge we are currently the group most focussed on movement building (though Global Priorities Project is mostly research so it doesn't stand out there, and 80,000 Hours is something like 50/50).

Robert_WiblinMay 28 20150

Surprised to get downvotes on this. Is the thinking the GW is movement building through its excellent work, or I'm neglecting Leverage Research, or something else?

arrowindMay 28 20153

OK I'll own up. I downvoted in a blip of initial irritation that you hadn't answered my question, just talking more about CEA, making it look like your argument for funding movement-building might (to be direct) be primarily motivated by self-interest as one of GWWC's salaried Directors. I've now retracted the downvote given I've clarified with this comment though it would still be good to see the argument applied to funding things other than further growing CEA.

Robert_WiblinMay 29 20152

Ah OK. I was trying to acknowledge that GWWC wasn't unique in this regard and it would be totally understandable if someone bought the overall argument but then decided to give to another group doing similar work.

When I said 'we' I meant CEA as a whole, because as Ryan points out, we have a bunch of projects where the priority is movement building. Most other groups besides CS and CFAR have more of a research focus.

RyanCareyMay 29 20150

I would say that EA Outreach (by CEA), EA Ventures (by CEA), and Paul Christiano's Certificates of Impact would be the main candidates other than GWWC.

Benjamin_ToddMay 30 20150

Obviously I'm biased, but I think 80,000 Hours has one of the best track record of creating new members of the EA community, and we have scalable ways to continue doing that at the margin, so I think we should at least be a candidate for best EA movement building project. EAO and EAV might turn out to be even better, and there's strong reasons in favor of them, but the evidence is much weaker so far.

arrowindMay 31 20151

Again this is just a list of CEA organizations and a pitch for continually growing funding. But I'll go with that and ask you to expand on the track record: what are the best examples of someone trying to make the strongest case for and strongest case against it? When I've heard people make the case against, it's that people counselled (at least in the northeast megalopolis and London) have been EA's before.

Benjamin_ToddMay 31 20151

Sorry, I was just responding to Ryan's comment, not addressing your overall point.

You're correct that some people we coach are already EAs, but the majority of coachees and especially the 25k monthly website readers either (i) know about EA but aren't actively involved or especially well linked into the community (ii) don't know much about EA. But many people we coach end up active in the community. There's a clear mechanism for this: we make introductions, talk about EA with them, persuade them of the key ideas, and help them figure out how they can be more EA in their career. I'd recommend reading through some of our last evaluation to see the types of career changes people made, many of which involve becoming "more EA". https://80000hours.org/2014/05/plan-change-analysis-and-cost-effectiveness/

Since hardly any recent graduates making career decisions already know about effective altruism, there's huge room of expansion. And talking about career decisions is a great platform for discussing EA ideas, because it's a very big decision but existing advice is so bad.

There's also many people involved with effective altruism but who only donate income and don't know how to apply effective altruism to their career. 80k provides motivation and information to people in this category, making the EA community more effective. Many of our past plan changers are in this category e.g. Peter Hurford. https://80000hours.org/career-guide/member-stories/peter-hurford/

One reflection of this is that we've catalysed the development of 8 new EA non-profits (5 of which have full-time staff) which likely wouldn't have been created if 80k hadn't existed. https://80000hours.org/2015/04/10-new-organisations-founded-due-to-80000-hours/

I think there's a lot more that could be done to improve the career choices of existing EAs. Although our general frameworks are good, the next stage of our research is to investigate all the specific neglected paths EAs could follow e.g. policy careers, what to do within international development, various types of academic research and so on. This is becoming more and more pressing as earning to give looks less important on the margin.

Major arguments against:

You can always argue much of our movement building impact would have happened anyway, just later. That's generally true, but it applies to every EA movement building organisation. Since 80k has been one of the main contributors to growing the community, 80k is causing a portion of the impact. Also speeding up growth is still highly valuable.
You may just think there's some other project at the margin which offers much higher returns for getting new people involved, like investing in EA student groups.
And you could think it's just very hard to improve an EAs career choices, so more research is also not helpful.

Disclosure: I work at 80,000 Hours.

RyanCareyMay 28 20150

Maybe Charity Science?

Robert_WiblinMay 28 20151

Ah of course, yes CS and GWWC are working on more or less the same problem (indeed we make grants to them for UK donors!).

arrowindMay 28 20153

Looking at Charity Science, they do talk about spreading the word about evidence-based charities but reading between the lines they appear to be quite different from movement building/GWWC in that they focus on fundraising, with 'spreading the word' perhaps partly a more acceptable face to present to the fundraised-from. And I couldn't quickly see any references to the effective altruism movement on their website, so I don't think they'd be a good choice for someone following your argument for the absolute priority of movement-building.

Robert_WiblinMay 29 20152

If the problem is framed as 'money isn't going to effective charities' then we look very similar.

But I agree that my impression (admitted from limited evidence) is that CS is more focused on moving money immediately relative to winning hearts and minds for the long term than GWWC or EAO.

zdgroffMay 27 20151

I was just curious about the ratio and whether you had given thought to what the optimal level is, not asking for donation-splitting purposes. FWIW, I've heard the argument against splitting small donations. I do split them, but admit it's probably not the best thing to do. I'm just indecisive and get fulfillment out of supporting both causes, so I've permitted myself this irrational behavior. My guess is there may be others like me, and maybe there are some who split donations because they think we should split our donations the way we would like overall donations to be split.

Robert_WiblinMay 27 20151

Oh and sorry I missed part of the question - yes monthly is a great way to give. :)

DaleAug 11 20150

In the document you suggest using a 3.5% discount rate. Is this a real or nominal discount rate?