Feb 28, 2017
This post was written by Kerry Vaughan, with contributions by Larissa Hesketh-Rowe and Tara Mac Aulay. Thank you to Nick Beckstead and Holden Karnofsky for providing early feedback and edits.
This post is a follow-up to Will’s post introducing a project we’re calling the Effective Altruism Funds (EA Funds). This post provides a more detailed explanation of the project and why we think it may be among the highest impact donation options for many individual donors.
We only want to focus on the Effective Altruism Funds if the community believes it will improve the effectiveness of their donations and that it will provide substantial value to the EA community. Accordingly, we plan to run the project for the next 3 months and then reassess whether the project should continue and if so, in what form. The main way we will assess if the funds provide value to our community is total recurring donations to the EA Funds and community feedback.
If you think it is plausible that donating to EA Funds is more effective than your alternatives, the best way to signal support for the project is to make a donation through the beta version of the project. Donations to EA Funds are tax-deductible for both UK and US donors. A link to the beta version is provided below.
To provide feedback on the project, please fill out this short survey .
If you have an idea for a fund or a fund manager, please fill out this short survey .
EA Funds are mutual funds for giving effectively within core EA cause areas. When you donate to a fund, you specify the cause area, and pool your donation with many like-minded donors. Cause area experts then decide how to best allocate the pooled donations to the most promising giving opportunities they can find.
Using EA Funds involves three basic steps:
We don’t expect EA Funds to be the highest impact donation option for all donors. However, we only want to devote significant time to the project if it is plausibly a higher-impact donations alternative for many individuals in the EA community.
Below are four arguments for why EA Funds might be higher-impact than the alternatives available to many donors. We expand on each argument below.
Donations to EA Funds could be at least as good as Open Phil’s last dollar in the following four ways:
Open Phil fund managers will treat money donated through EA Funds as additional funding allocated to the cause area, so donors who have a strong preference for some cause areas over other can potentially increase the total amount of funding allocated to that cause area.
While Open Phil program officers have considerable freedom in their grant making, there may be additional giving opportunities that are not a good fit for Open Phil. EA Funds gives Fund Managers a pool of funding they can make grants from with more brand separation from Open Phil, possibly allowing them to counter risk-aversion. Fund Managers might also encounter giving opportunities that are particularly time-sensitive, or very small, where the need to go through Holden and Cari is prohibitive.
Open Phil currently tries to set an upper limit on the proportion of an organization’s budget they will provide, in order to avoid dependence on a single funder. In the case where EA Funds generates recurring donations from a large number of donors, Fund Managers may be able to fully fund an organization already identified, saving the organization from spending additional time raising funds from many small donors individually. Donors who sign up for recurring donations give a strong indication of their desire to continue funding this cause area, which increases the amount of resources that Open Phil can allocate to any individual organization in a given year.
While Fund Managers will attempt to avoid funging donations between individual donors to EA Funds and Open Phil, we think the issue of funging deserves more thorough exploration. In our post introducing the basic idea of the EA Funds, some commenters expressed concern than money given through EA Funds would funge against the Open Philanthropy Project (Open Phil) since the fund managers are all Open Phil employees:
>It seems strange to have the funds run by people who also direct money from on behalf of big grant-making organizations. Under what circumstances will the money end up going somewhere different? ... [T]he current incarnation seems to be basically equivalent to just giving GiveWell or OPP money with a cause-based restriction. -Larks
While the Fund Managers will continue operations as normal within Open Phil, even if the Fund Managers were unable to find any more promising opportunities, and donations through EA Funds did end up funging with Open Phil, this might not necessarily be a bad thing. Getting funged by Open Phil means your donation is at least as good as Open Phil’s last dollar. As Carl explains :
>In principle one one could donate to the donor-advised fund (DAF) of Open Phil, directly increasing its ultimate donation capacity. At the moment, this doesn't seem to be set up, but one could instead donate to something that Open Phil is donating to (inframarginal), and request that it 'funge' you by reducing its own donation to that charity by the corresponding amount, increasing the reserves of Good Ventures and other Open Phil backers accordingly. So the marginal Open Phil/Good Ventures dollar sets a minimum standard for risk-neutral donors: if you don't expect to do better than Open Phil, just arrange to get 'funged'.
To determine whether getting funged by Open Phil is better or worse than the alternative we need to know how good Open Phil’s last dollar would be.
Open Phil’s current view on the value of their last dollar is:
>... I have very low confidence in my working view on how good the “last dollar” is likely to be, and I expect my view to change quite a bit in the future. On balance, our very tentative, unstable guess is that the “last dollar” has higher expected value than gifts to GiveWell’s top charities today.
Carl’s view on the topic is:
>... [M]y expectation for the 'last dollar' of Open Phil’s portfolio is exceptionally high relative to the general world of charity. Among other things, I think even after diminishing returns some combination of scientific research (e.g. gene drives to eradicate vector-borne diseases), policy work (e.g. on foreign aid or science policy), nonhuman animals, global catastrophic risks (potential risks from AI, biosecurity, nuclear risk), and others put the expected value of the 'last dollar' for Good Ventures higher than for GiveWell's top charities.
Thus, it seems quite plausible (although uncertain) that funging with Open Phil is higher impact in expectation than the lowest-cost alternative available now, namely, donating to GiveWell-recommended charities.
In cases where the fund managers are Open Phil Program Officers (which is the case for all of the funds during the three month trial period but could change in the future), funging against Open Phil is a plausible lower bound on how good donations through EA Funds will be. If fund managers are unable to find something that beats Open Phil’s last dollar in expectation they could simply donate to Open Phil’s existing grantees such that Open Phil can reduce their funding to those organizations by a similar amount. Donors who believe that Open Phil’s last dollar will do more good than their current giving may find EA Funds an attractive option for this reason.
EA Funds allows individual donors to pool their donations with others who share their world-view and values, then delegate some of the remaining empirical research and decision-making to people with a comparative advantage in allocating money.To make an effective donation, individual donors must try to answer all of the following questions, given their values:
Time spent answering these questions is time not spent doing other worthwhile activities, and so many donors will wish to spend marginal time investigating cause selection, while deferring the empirical research to those with a comparative advantage in doing so. In addition, it is particularly time consuming to answer questions (3) and (4), as answering these questions requires timely information, and knowledge of other funders’ intentions. Program Officers at major foundations have a particular comparative advantage at answering (2) and (3) since they spend virtually all of their working time thinking about how to allocate money to achieve the most good in any given cause area. These people are also particularly well suited to answer (4) as they have up-to-the-minute information about the largest funder in the space. EA Funds are designed to allow donors to focus their research efforts on (1), the area where differences in world-view or personal values may lead to the biggest differences in donation targets.
The initial Far Future and Effective Altruism Community funds will be managed by Nick Beckstead, a Program Officer at Open Phil who has helped advise a large private donor on donation opportunities for several years. The fund Nick manages was an early funder of CSER, FLI, Charity Entrepreneurship and Founder’s Pledge. A list of Nick’s past funding is available on the EA Funds website.
We think this represents a strong track record although Open Phil’s recent involvement in these areas may make it harder for the fund to find promising opportunities in the future.
Donors can give to the DAF directly by filling out this form and waiting for Nick to contact you. If you give directly the minimum contribution is $5,000. If you give via the EA Funds there is no minimum contribution and you can give directly online via credit/debit card, ACH, or PayPal. Nick's preference is that donors use the EA Funds to contribute.
*Disclaimer : Nick Beckstead is a trustee of CEA. CEA has been a large recipient of the EA Giving Group DAFs funding in the past and is a potential future recipient of money allocated to the Movement Building fund.*
As well as the increasing the effectiveness of individuals’ donations in the short term, we believe that the existence of, and further development of EA Funds as a project might have significant longer-term benefits, which you could see as a positive externality of using the funds.
Longer term, if EA Funds continues beyond the three month trial, it might increase the incentives for researchers to explore new areas for potential donations. This would be both because the EA Funds would cause money to be available for allocation based on the research, and because in the future we hope to encourage new fund managers to create new funds with different focus areas than the current options.
The ability of the EA community to discover new ideas for funding has been one source of the community’s value in the past. As Carl notes:
>[I]n past years I have recommended donation targets other than contributing to the Open Phil grant pool to people asking my advice, generally in the areas of reducing potential existential risk from future developments in artificial intelligence, and developing institutions in effective altruism. These recommendations were for areas where several of the above factors applied: organizational risks, reputational/communication problems, staff bottlenecks, and interactions with broader worldviews. In subsequent years OpenPhil did enter the areas, but the early grants were able to fund time-sensitive opportunities such as seed and growth funding.
For example, many areas that were of interest to the effective altruism (EA) community subsequently became focus areas for Open Phil. Examples include: immigration policy, existential risks (especially risks from advanced AI), farm animal welfare, and effective altruism. It seems plausible that the EA community could continue to serve as a source of ideas for very large funders in the future.
A future version of EA Funds might also have lower barrier to entry than, say, getting a job as a Program Officer at Open Phil. This would allow more people to try out in-depth charity research to see if they should specialize in it. It also provides a clear feedback mechanism for this kind of research, namely, whether the EA community chooses to donate to your fund.
As Will notes in the initial post, it might be possible in the future to use the EA Funds infrastructure to initial a number of interesting, related projects. Two examples are moral trades and coordinating donor lotteries .
That are likely other interesting experiments that can be conducted with the donation infrastructure required for the EA Funds once it gets to scale. Supporting the EA Funds would make it more likely to go beyond the trial phase and reach the scale necessary to conduct these experiments. The EA Funds concept need not remain static as we scale and experiment more in the future.
By virtue of their size, foundations and other large donors gain a number of benefits that are not available to individual donors:
Concentrating more of the funding done by EAs through the EA Funds and putting more funding into the hands of individual fund managers can help individual donors gain access to some of these benefits of size that are generally not available to individual donors.
EA Funds can help solve certain donor coordination problems that arise between strategic, impact-focused donors. We discuss two types of coordination problems and how EA Funds can help solve them below.
Problem: Funders have an incentive to be the last funder in an organization's fundraising round as this makes it more likely that the donation is not trading off against funding that could have been acquired elsewhere. Some degree of this probably makes sense, but it may distort incentives or falsely signal a lack of support for an organization if used too frequently.
Solution: EA Funds coordinates more of the funding through individual Fund Managers to create a single, large funder. Large funders generally do not suffer from the last funder in problem for two reasons. First, large funders generally only have to worry about other large funders for their donations to be crowded out. Since there are few large funders, this makes coordination much easier. Second, larger funder can request much more information from funding recipients. This makes it easier for large funders to get a sense of the funding landscape for the recipient and to then coordination with other funders.
Problem: Standard practice among individual donors in the EA community is to provide projects with around one year of funding and then re-evaluate when they ask for funding again. This makes sense for funders as it allows them to easily pivot to funding more valuable projects later. Yet, for project leaders this means that they don’t know whether funding will be available next year. The uncertainty can cause project leaders to be more risk-averse than they would otherwise have needed to be.
Solution: Large funders solve this problem by providing multi-year grants tied to specific outcomes. This allows large funders to pull funding from underperforming organizations. It also allows funding recipients to predict their funding income multiple years into the future and to plan accordingly. Designing grants is time-intensive for funders and funding recipients. This time is worth it at large funding sizes, but probably not worth it at smaller sizes.
We want the EA community to use EA Funds if and only if doing so is the highest-impact donation option in expectation. We anticipate that EA Funds will not be the highest-impact donation option for a number of donors, especially donors moving large amounts of money and spending considerable time determining where to allocate their donations.
We discuss three considerations against donating to EA funds below.
The beta version of EA Funds has four funds all of which are managed by Open Phil Program Officers. Given that Open Phil is already one of the largest funders in these areas, concentrating more money in the hands of Open Phil staff may have several negative effects.
We see the initial selection of funds and fund managers as the MVP version of EA Funds. In the future we aim to create a wider variety of funds with a wider variety of fund managers.
We chose Open Phil Program Officers as the initial fund managers for three reasons:
We don’t yet know how much money might be donated through EA Funds. If the amount of money turns out to be small, a highly skilled fund manager might be better off spending relatively little time allocating the fund in favor of spending time on other activities. This fact could create two detrimental effects.
First, the fund manager might feel obligated to spend considerable time allocating the funds even if that would be better spent on other activities.
Second, donors might be incentivized to wait to see how large the fund will be before donating because they anticipate that the fund manager will spend more time allocating larger donation pools. This creates a donor coordination problem that we would rather avoid.
However, Open Phil’s Program Officers have reason to investigate these spaces independent of the amount of money raised through the EA Funds. If the EA Funds raises little money, they can spend little additional time allocating the EA Funds’ money but still utilize their deep subject-matter expertise in making the allocation. This reduces the chance that the EA Funds causes fund managers to use their time ineffectively and it means that the lower bound of the quality of the donations is likely to be high enough to justify donations even without knowing the eventual size of the fund.
We found the idea that Open Phil’s last dollar could be a lower bound on the impact of donations (as discussed by Carl Shulman ) compelling. Working with Open Phil’s program officers seems like one of the best ways to make this the lower bound in practice.
Unsurprisingly, Open Phil has been able to hire some exceptional Program Officers. When we thought about the ideal candidates for managing the funds we found that many of the best candidates were Open Phil employees.
Some donors may choose to donate elsewhere if they find that they have specific disagreements with the worldviews of the fund managers.
All fund managers have histories of past donations made through Open Phil although these may not be representative of future donations. Prospective donors should look at the fund manager’s past donations (we have included these on the EA Fund website on the page for each fund) and the reasoning for their donations to determine if they have specific disagreements with the fund managers.
Behind the scenes, the structure of EA Funds operates similar to a donor-advised fund (DAF) with CEA operating as the DAF. Donations go to CEA and are put aside until the Fund Manager recommends a donation. CEA’s trustees must then approve the donation before the donation is sent to the recipient. Just as in a DAF, making a donation through EA Funds surrenders ownership of the donation to CEA with the understanding that the fund manager will have advisory privileges over how the fund is used.
We’ve opted for this structure over having donors give to a DAF for each fund for two reasons: 1) we believe this structure will allow us to reduce the fees associated with giving to the EA Funds by avoiding the administration fees that DAFs charge and by allowing us to negotiate lower processing fees with payment processors; 2) We believe that this structure will allow us to create a more seamless donor experience in the long-run which may allow us to move money into EA causes from those not involved in the EA community.
However, this setup requires placing trust in CEA to donate the funds as intended. In addition, there may be unexpected edge cases that would require CEAs discretion to resolve. Some examples include: fund managers resigning, fund managers ending their term with money remaining in the fund, fund managers making recommendations that are not in keeping with the description of the fund etc. CEA is unlikely to have planned for all edge cases and may need to exercise discretion to resolve unforeseen circumstances. Donors should make sure they are comfortable with this before donating.
While we think our selection of fund managers is ideal for the beta version of EA Funds, it is not optimal in the long run. If we decide to proceed with the EA Funds project after the three month trial, our aim would be to have 50% or less of the Fund Managers be Open Phil Program Officers (although they may manage more than 50% of the money donated). Some things we might consider to accomplish this include: adding funds in new cause areas, adding funds in the existing cause areas that represent a different approach to solving the problem, having multiple people manage some of the initial funds, or through some other approach.
If you have ideas for new funds or new Fund Managers, please fill out this form .
We think EA Funds may be among the highest impact donation options for many donors although the case is far from certain. We plan to use community feedback and engagement with EA Funds to determine whether we should allocate substantial resources to the project in the future.
If you think EA Funds is not among the highest-impact options for you, we would be interested in hearing why in the comments below or our feedback form . We would also be interested in hearing whether you think your case will generalize to others.
If you think EA Funds is among the highest impact options for you, the best thing to do is to make a donation using the beta version of the project at the link below. We would also appreciate any comments on how you plan to use it or why you think it would be a good option for you. However, our evaluation of the potential for the project will depend much more on whether people actually donate through the beta version than it will on whether people expressing support for the project in the abstract.
*P.S. We are planning to take advantage of YC’s launch infrastructure to generate some initial press and users for EA Funds. We will likely do this regardless of community feedback because we think the time cost is justified by of 1) seeing how the concept is received outside of the EA community and 2) learning about YC’s process for gaining media attention.*