In the EA movement people will sometimes talk about charitable funds as if they are a new idea. For example, the recent Giving What We Can post " why we recommend using expert-led charitable funds" ( forum discussion) opens with:

Funds are a relatively new way for donors to coordinate their giving to maximise their impact.

There's a bit of a cute response that would be fun to write, except that it isn't quite true:

Donating through a fund that is able to put more time and effort into evaluating charitable options is not a new idea, and is very natural if you're thinking along EA lines. In fact, the idea goes back at to the last time people tried to invent effective altruism, 150 years ago. While that effort has been through a few names over time, at this point you know it as the United Way.

While this does have elements of truth, it implies historical motivations were more similar to current ones than they really were. As usual, history is much more nuanced.

The United Way is not a single grantmaker which will direct your money to wherever they think it will do the most good. Instead, it's a collection of local organizations, each responsible for their immediate community. This organizational structure reflects one of the larger differences between the Scientific Charity Movement and effective altruism: the former had a strong local focus, while EAs push for spending money wherever the impact is largest. This structure also comes from how the United Way grew out of the Community Chest movement, where each city had their own Community Chest fund.

Coming from a modern EA perspective you might guess that the motivation was to improve the allocation of money within a city, even if not between cities: instead of individuals needing to assess competing community charities, a central fund could weigh their needs and make informed tradeoffs. But reading about the development of Community Chests (ex: Burleson 1926, Todd 1932 and discussion, Seeley 1957) the core problem Community Chests were solving was duplicative fundraising.

For example, Todd (1932) writes about a "federation of givers who were tired of miscellaneous appeals, and banded together for common defense" and "this motive of protection against miscellaneous solicitation on the one hand, and against serving in a constant succession of philanthropic 'drives' on the other, is still dominant". Instead of competing, charities could band together and more efficiently (and less annoyingly!) fundraise jointly. This also let these campaigns fill a role similar to the war effort fundraising of the 1910s, where they were widely seen as legitimate and democratic, a safe way for wealthy residents and business owners to contribute.

I do see a small amount of discussion on how centralized funding might improve allocation, but it's pretty minor compared to fundraising scale, and it's mostly seen as a potential opportunity, not something currently done well:

One of the arguments frequently offered for organizing a community chest is that it would give the directors of social agencies time to consider the proper spending of their resources instead of having to put in all their time in raising money. That is to say, the competition would be shifted from finance to standards of performance. Not all communities have by any means reached such an ideal level. ... The process of accommodation frequently takes the form of compromise. In order to secure, as large a constituent membership as possible, the promoters of the community chest will admit agencies whose standards may not measure up to the best levels of social work in the community. (Todd, 1932)


The Community Chest with its centralization of finance has a unique opportunity to set up machinery for gathering the information concerning the community and concerning the work of the agencies, which would provide a more adequate basis for the social work of the community. I am quite aware that this is no new idea. A number of the chests have already made some excellent beginnings in setting up these research bureaus, under one form or another. And there is rather general lip service to the importance of fact-finding. Perhaps the thirteen or fourteen years which represent the period of development of the Community Chest movement, is too short a time in which to expect the creation of elaborate research departments. (North, 1932)

Other places where we see a similar kind of centralized allocation are large multi-focal charities (Catholic Charities, Oxfam, IRC, Red Cross, etc). These are different in structure in that a lot of their funding goes to running their own programs instead of regranting, though they do some of both.

The history of charitable funding is complex, and I've only touched on it here. But I do think EAs should be cautious about talking about charitable funds as if they're something we invented. Instead, I think they're better viewed as another iteration in a long history of fundraising and allocation.

Comment via: facebook, mastodon





More posts like this

Sorted by Click to highlight new comments since:

Super interesting history here - thanks for sharing this research! 

You note at the end that large, multi-focal charities are different because "a lot of their funding goes to running their own programs instead of regranting, though they do some of both". It's worth noting that when you give to funds like the GiveWell Unrestricted Fund, you support GiveWell's operations. In fact, about 2% of GiveWell's overall spend last year was on internal ops. While 2% is a tiny number compared to organizations like Oxfam or IRC, it's still worth noting. I am in no way criticizing the $11M GiveWell spent on their operations last year - GiveWell needs this money to offer their services and conduct research. 

Another complicating factor is that GiveWell chose last year to begin "rolling over" funds. In practice, this meant that $66M donated to the Top Charities Fund in 2021 did not get donated in 2021. Again, I am not criticizing this decision, but it means that when you give to the fund, your donation is not necessarily immediately directed to an NGO. To your point, the same is/was certainly also true of United Way or the Community Chest. But folks should know that when they give to the Top Charities Fund, they cannot expect their donation to reach the bank account of a charity as quickly as if they had donated directly.

Neither of these factors was mentioned in Giving What We Can's post as potential reasons against giving to a fund, but in my opinion, GiveWell shouldn't be reluctant to share these details. I think that if anything, GiveWell explaining their internal ops and roll over funding strategy helps demystify the new EA incarnation of the charitable fund. Instead of billing it as something entirely novel, GiveWell et al. should bill it as a vast improvement on previous efforts to pool and direct charitable donations more effectively.

The analogy here would be if GiveWell, in addition to making grants to other charities, also had their own on the ground bed net distribution program.

Yes, exactly. One might even wonder if, because a GiveWell recommendation generates SO much funding, select on the ground charities have conformed to the point that we might consider GiveWell (via their standards/eval criteria) an essential part of their operation model. (I don't have the info/experience to support this claim - just interesting food for thought...)

Only GiveWell's Unrestricted Fund supports GiveWell's operations

Oops - I realized I understood this incorrectly and have edited my comment. Thank you for the clarification.

Curated and popular this week
Relevant opportunities