31

EA Funds - An update from CEA

CEA sees the EA Funds platform as a key piece of community infrastructure, not only for the convenience offered to donors who wish to give to one of the four managed Funds, but also for its ability to take tax deductible donations in the UK and US (the countries where the Centre for Effective Altruism is a registered non-profit) and regrant them to many of the organizations in the EA space.

We are now dedicating the resources required to improve and build on the early success of the platform (though we recognize this has not been the case throughout the timeline of the project). This post will go into detail on our plans for further deepening the value the EA Funds provide to the EA community, identify mistakes we have made along the way and provide a general update on our thinking regarding the future of EA Funds.

Initial Context

EA Funds was built during the early months of 2017, during CEA’s time as a participant in the Y Combinator startup accelerator. The platform was built as a means to facilitate donations in the EA community. This was built up from (amongst other things) experience with the Giving What We Can Trust, which performed a similar role, but which lacked some of EA Funds’ key features, and had a smaller subset of possible grantees.

In particular, we were motivated by the idea that people often set up their donations on an ongoing, recurring basis (which is good, because it removes trivial inconveniences  to donating), but don’t often review them. We were also interested in increasing specialization of donation decisions, such that a smaller number of more expert people could decide the best ways to spend money, rather than everyone duplicating the effort of in-depth research.

To a large extent, this has been very successful. To date we have regranted (or are currently in the process of regranting) more than $5 million of EA-aligned money to around 50 organizations. We’ve been able to extend the platform to provide additional payment-related services (for example, the donor lottery run at the beginning of the year), and – in the wake of the crypto boom at the end of 2017 – made it straightforward for donors to donate large cryptocurrency windfalls tax-efficiently.

However, the project has always existed alongside other CEA projects, and earlier in 2018 (and given the platform had been operating stably for close to a year), we made the choice to deprioritize active work on EA Funds in favour of devoting more staff resources to other projects (where active work includes technical work to improve the user experience, operations work to e.g. bring on new grantee organizations or to check in on a regular basis with Fund managers). EA Funds has been operational and has been making regrants for the entire period, but it hasn’t been the primary focus of our technical or operations staff.

Our intention has always been to come back to EA Funds, and when we made the choice to prioritize other projects in Q1, we expected to that this would happen later in the year. Furthermore, at the most recent CEA team retreat (mid-June), we reviewed how EA Funds was operating, and decided that we were ready to re-commence work on the platform in order to ensure that it remained a trusted and reliable donation option. As such, we made the choice that we would begin again in earnest to make some overdue improvements to the operation of the platform, and have been doing so since the beginning of Q3. In other words, we’ve been aware of the same problems that have recently been raised in community discussion and working on appropriate solutions.

We regret not communicating about this situation more openly, and in retrospect recognize that waiting until we have resolved all the issues and had more concrete, positive announcements to make has led to a situation where some in the community began developing reservations about how the platform was being run. We plan to be more open about what we are working on with the platform going forward.

Requested improvements

The two main things that community members have asked for during the period EA Funds has been operating have been:

  • Easy visibility of current Fund balances
  • A more consistent regranting schedule

The former is discussed in detail below under the heading Updates to Fund Management, the latter under Operational Updates.

Other features we’re working on at the moment:

  • Better integration between payroll giving providers and EA Funds (so that soon donations created through a payroll giving provider will be visible in your EA Funds dashboard)
  • An automated system to let Fund managers and grantees know their current account balance on a quarterly basis, to prompt more regular regranting

Updates to Fund management

We’re currently in the process of finalizing some changes to how the Funds are managed. The most significant announcement is that Nick Beckstead will step down as the manager of the EA Community and Long-Term Future Funds.

Nick Beckstead has provided the following statement, explaining his reasons for stepping down as a manager of these Funds:

At the time I drafted my April update, I told CEA that I was interested in stepping down as a manager of the EA Community and Long-Term Future Funds. My reasons are as follows.

The original premise of me as a fund manager was that I would fund the grants that I wanted to make but couldn’t fund through Open Phil, and that this wouldn’t require a substantial additional time investment on my part. However, I encountered fewer grantmaking opportunities that I couldn’t fund through Open Phil than I expected, partly because Open Phil’s funding in these areas grew more quickly than I expected. The main thing I’ve wanted to do at Open Phil, but haven’t been able to, is to increase the size of my EA and GCR (Global Catastrophic Risks) grants, which is a large part of the reason I made the set of grants that I did. I could in theory have spent additional time sourcing and evaluating other opportunities, but it has generally seemed that additional time would be better spent on grantmaking at Open Phil or other Open Phil work (especially given that other funders who have more time to devote to the effort and currently are working with smaller budgets - such as EA Grants and BERI - were becoming involved in the space).

Going forward, I believe it will be more impactful and more satisfying to our community to have people managing the EA Community and Long-term Future Funds who can’t make most of their grants through other means, and for whom it would make more sense to devote more substantial attention to the management of these funds. I believe this could also be beneficial because it would increase the diversity of perspectives of grantmakers and offer others an opportunity to gain experience and establish larger track records as grantmakers.

Nick and CEA agreed that he would step down from his role as manager of these two Funds after making his final set of grants in July 2018, and in late May 2018 a process for naming his successors was started. Nick has now recommended his final grants, and we are in late stages of the process to name the new managers for these Funds.

During this process we have been focusing on finding managers that are widely respected in the community, have some experience with funding relevant organisations at moderate scale and are able to commit the time required to making grants about twice a year or more. Given some of the considerations Nick outlines above, we’re also planning to experiment with a ‘committee’ approach, where several individuals may be involved in providing grant recommendations for a particular Fund. We look forward to announcing the new managers to the community in the coming weeks.

It’s worth noting that, notwithstanding the change in Fund manager, and the fact that we expect this to lead towards more regular regrants, we still believe that, as a general rule, grant managers should donate when they see good opportunities, not on the basis of a fixed schedule, as we think that good grants made later are likely to be better than rushed grants (at least for some bounded period of time, after which the effect of discounting would make the grants equally valuable).

Operational Updates

Over the last few months, we have been working on making improvements to the platform so that the day to day operations run smoothly, transparently and take a reasonable amount of staff time.

The first visible result of this is an update to the EA Funds website, where each of the four main Funds now has the current balance (as of the end of the previous calendar month) displayed above the list of regrants, taken directly from our accounting system. This will bypass the need for manual updates being posted, and allows everyone access to the latest numbers at all times. If you now view a Fund’s page (see e.g.  the Animal Welfare fund’s page ) you’ll see a ‘current balance’ section outlining the amount currently available to the Fund manager for regranting.

We hope to expand on this with more detailed reporting functionality in the medium term. Please do note that there will be some variability in the numbers month to month due to foreign exchange fluctuations (balances are held in both USD and GBP) and, at least initially, due to occasional accounting adjustments.

We note that we’ve had the intention to get a system for making Fund/Organization balances visible onto the website since early in the year. However, we faced the challenge of providing a definitive number for the amount of money in a fund. For a number of fairly mundane accounting reasons (e.g. the delay between a donor reporting a donation and us receiving the money via bank transfer or cheque, donors giving outside the EA Funds system, donors changing their donation allocation, complications with accounting for the appreciated value of cryptocurrency post-liquidation and decisions about whether to use numbers derived on a cash or accruals basis), it has been difficult for us to point to a single, canonical number that accurately represents the balance of a given Fund at a given point in time.

We’ve spent the better part of this quarter refining our processes so that we now feel confident that the number that we’re pulling from our accounting software is a good representation of the ‘current balance’ of the fund.

A note on differences between March balances and current balances

The balances as they appear on each Fund’s page are in some cases lower than what the last public update in March (less any grants) would suggest. This is due to the March numbers including accounts payable; money not yet held as cash but expected to come in. The largest difference is in the Animal Welfare Fund’s balance, where over $500,000 of such transactions was reported as part of the balance in March.

It appears that a number of large donations which were reported during the December ‘giving season’ turned out to be people testing/experimenting with the EA Funds platform; therefore, the March balances included amounts which were never meant as genuine donations. While it’s standard accounting practice to report accounts receivable as an asset (and then perform regular write-offs for money that doesn’t come in), we recognize this has caused confusion. As such, the account balances displayed on the site now exclude accounts receivable to prevent this confusion going forward.

We will also continue our work on the accounts in collaboration with our external accounting teams in the UK and US to reduce the need for core CEA staff to spend too much time on this, and we are developing new approaches to our regranting operations, which are currently a highly manual process. This means that we plan to process all regrants at the same time once per quarter.

We think the EA Funds is uniquely positioned to provide value to the EA community by saving the duplication of similar operations effort at other organisations, and we’re excited to keep improving the platform for the benefit of the community.

Messaging regarding effectiveness of donations to the managed Funds

We initially believed that the four managed Funds would significantly improve the effectiveness of the average EA donation. This was in part based on the assumption that our Fund Managers, drawn from GiveWell and the Open Philanthropy Project, would be able to easily find funding opportunities that they believed to be on par with their ‘institutional’ grants, but that their respective institutions would be more reticent to fund. We have since updated our thinking on this, because this has turned out not to be the case in all situations and also because we are now more worried about the effect that having Fund Managers exclusively from the Open Philanthropy Project and GiveWell has on the diversity of viewpoints represented in funding decisions within the community.

We are changing the messaging on the EA Funds site to better reflect this updated thinking. We continue to see the managed Funds as an excellent donation choice for donors who wish to save time and just want to donate whilst knowing their donation effectiveness will meet a very high bar. However, for donors more deeply engaged in the community it is completely plausible they are able to outperform the effectiveness of the managed Funds, especially in situations where they are in a unique position to fund an early stage project they are aware of due to their network, or other similar situations.

Further diversification of funding viewpoints: an extra tier of Funds

We have also been planning to further increase the diversity of views represented in the managed Funds model by creating a new ‘tier’ of Funds. Plans for this are still in early stages, and we’re not planning to increase the operational complexity of the platform until we have handled our administrative backlog discussed above. However we hope to be able to announce more detailed plans for this in the coming months.

The basic idea would be to allow less-tested grant makers, with particularly relevant personal networks or potentially less-mainstream views to manage a ‘Junior Fund’ (working title), where these would not be promoted to the same extent the main four Funds are, but would nevertheless be easily accessible to interested donors. We would then ‘let the market decide’ to a certain extent on which of these succeed, setting a minimum donation threshold to be reached within the first 12 months for these Funds to continue to be worthwhile to operate. It seems plausible that we could discover highly competent new grant-makers this way.

We will most likely run a public contest for these positions.

Conclusion

We would like to seek your feedback on the plans and updates laid out in this post, and I would like to encourage anyone who would like to discuss any of these matters in more detail to reach out to me on funds at effectivealtruism dot org. 

Comments (21)

Comment author: RandomEA 09 August 2018 04:55:09AM 1 point [-]

Would it be a good idea to create an EA Fund for U.S. criminal justice? It could potentially be run by the Open Phil program officer for U.S. criminal justice since it seems like a cause area where Open Phil is unlikely to fund everything the program officer thinks should be funded, which makes it more likely that extra funding can be spent effectively.

This could help attract more people into effective altruism. However, that could be bad if you think those people are less likely to fully embrace the ideas of effective altruism and thus would dilute the community.

Comment author: Carl_Shulman 09 August 2018 07:21:56AM *  3 points [-]

Would it be a good idea to create an EA Fund for U.S. criminal justice?

Open Phil's Chloe Cockburn has a fund for external donors. See Open Phil's recent blog post:

Chloe Cockburn leads our work in this area, and as such has led our outreach to other donors. To date, we estimate that her advice to other donors (i.e., other than Dustin and Cari) has resulted in donations moved (in the same sense as the metric GiveWell tracks) that amount to a reasonable fraction (>25%) of the giving she has managed for Open Philanthropy.

It appears that interest in her recommendations has been growing, and we have recently decided to support the creation of a separate vehicle - the Accountable Justice Action Fund - to make it easier for donors interested in criminal justice reform to make donations to a pool of funds overseen by Chloe. The Fund is organized as a 501(c)(4) organization; those interested in contributing to AJAF should contact us.

Comment author: RandomEA 09 August 2018 01:52:58PM *  1 point [-]

Do you know if it's just a fund for other large donors? It seems unusual to require small donors to send an email in order to donate.

If the fund is open to small donors, I hope CEA will consider mentioning it on the EA Funds website and the GWWC website.

Comment author: Carl_Shulman 09 August 2018 06:10:22PM 4 points [-]

I don't know, you could email and ask. If Chloe wanted to take only large donations one could use a donor lottery to turn a small donation into a chance of a large one.

Comment author: weeatquince  (EA Profile) 08 August 2018 11:17:18AM 4 points [-]

Marek, well done on all of your hard work on this.

Separate from the managed funds. I really like the work that CEA is doing to help money be moved around the world to other EA charities. I would love to see more organisations on the list of places that donations can be made through the EA Funds platform. Eg, REG or Animal Charity Evaluators or Rethink Charity. Is this in the works?

https://app.effectivealtruism.org/donations/new/organizations

Comment author: vollmer 09 August 2018 07:50:53AM 2 points [-]

(To support REG, you can select "Stiftung für Effektiven Altruismus" (= Effective Altruism Foundation) from the list. If you'd like to restrict your donation to EAF's philanthropic advice (which includes REG, but also a new crypto fundraising project and an "impact masterclass" for UHNWs), just shoot us an email.)

Comment author: MarekDuda 08 August 2018 02:34:55PM 3 points [-]

Thanks Sam!

Yes, it is. We are currently focussing on operational robustness, but after that we see no reason not to expand the offering to cover most of the organisations EAs give to.

Comment author: weeatquince  (EA Profile) 08 August 2018 11:04:27PM *  0 points [-]

YAY <3

Comment author: Eli_Nathan 07 August 2018 07:09:53PM 4 points [-]

Thanks Marek,

I remember some suggestions a while back to store the EA funds cash (not crypto) in an investment vehicle rather than in a low-interest bank account. One benefit to this would be donors feeling comfortable donating whenever they wish, rather than waiting for the last possible minute when funds are to be allocated (especially if the fund manager does not have a particular schedule). Just wondering whether there's been any thinking on this front?

Comment author: Ben_Todd 08 August 2018 06:04:37AM 4 points [-]

I agree that would be ideal, but it doesn't seem like a high priority feature. The risk-free 1yr interest rate is about 2% at the minute (in treasuries), so even if the money is delayed for a whole year, we're only talking about a gain of 2%, and probably more like 1% after transaction costs.

You could invest in the stock market instead, but the expected return is still probably only 1-5% per year (as I argue here: https://80000hours.org/2015/10/common-investing-mistakes-in-the-effective-altruism-community/). Plus, then you have a major risk of losing lots of the money, which will probably be pretty hard to explain to many of the users, the press etc.

I expect the staff time spent adding and managing this feature could yield much more than a couple of percent growth to the impact of the funds in many other ways (e.g. the features Marek lists above).

Comment author: vollmer 08 August 2018 03:27:57PM *  4 points [-]

The sources you quote seem to suggest more like 5% in real annual returns (or 7% nominal), and you wrote "2-7% nominal returns". If you're investing $2m, that would be $40k-$140k per year. I'd expect this to cost maybe one week of staff time per year, so it might easily be worth the cost. (Mission hedging and more diversification would push this up further; fees and risk aversion would push it down. Overall I don't expect these factors to be very strong though.)

To me it seems that the difficulty of explaining this to the users is the stronger reason against implementing this. (Unless the users themselves can choose but that would cost more staff time again.)

Comment author: Ben_Todd 09 August 2018 03:20:14AM 2 points [-]

Sorry, I was thinking 1-5% of real returns rather than nominal, though I agree for these purposes nominal might be more relevant.

There's a lot of room for different figures depending on how mean reverting you think valuations are. I think we should expect them to be somewhat mean reverting, so my best guess is more like 3% real.

I was also partly thinking that valuations are much higher again since I wrote that post, so I think that article is a bit optimistic today. http://www.multpl.com/shiller-pe/

With costs, I'm more thinking about the set-up cost. I wouldn't be surprised if this took several person-months, which could be used to add features that would add much larger proportional gains.

I also guess the on-going costs would be quite a bit more than one week per year, due to the reasons Rob lists.

And then there are also the costs of explaining this feature to users, which seem pretty significant - e.g. even if you write up a clear explanation of why you're doing this and speak to a bunch of people about it in-person, you'll still end up with lots of misunderstandings.

Comment author: kbog  (EA Profile) 08 August 2018 05:08:20PM *  1 point [-]

The S&P500 has returned an average of >9% per year for the last 90 years.

How can you say "past returns are no guarantee of future returns" in response to this fact, then turn around and estimate future returns based on a snapshot of what yield and buyback are "right now"? The latter is much less reliable.

It takes about a minute to put money into a fund and another minute to sell it. Seems pretty simple to me.

Comment author: Robert_Wiblin 08 August 2018 09:29:35PM 10 points [-]

The effort required to set up a non-profit trading account, go through KYC, make it secure, teach everyone in the org how it works, and do the necessary legal, budgetary and accounting compliance each year make this much more than a few minute job. Things that are easy for individuals are often less straightforward for organisations.

Comment author: RyanCarey 08 August 2018 06:30:30AM 3 points [-]

Agreed. You could get a higher effective ROI by mission-hedging -- investing AI-risk funds in things like Google. But even then, the returns seem like a pretty second-order issue.

Comment author: kbog  (EA Profile) 08 August 2018 05:06:31PM *  0 points [-]

The benefits from mission hedging are far lower. If your stocks go up 10%, you get 10% more money. If Google stock goes up 10%, then money spent on AI risk goes up in importance by, I don't know, 0.5%, so you now have 10.5% more impact.

Comment author: Milan_Griffes 07 August 2018 08:02:54PM 3 points [-]

+1 to holding funds in an ETF or something – seems pretty easy to set up & starts generating a return immediately.

Comment author: SamDeere 07 August 2018 07:39:10PM 3 points [-]

Hey Eli – there has definitely been thinking on this, and we've done a shallow investigation of some options. At the moment we're trying to avoid making large structural changes to the way EA Funds is set up that have the potential to increase accounting complexity (and possibly audit compliance complexity too), but this is in the pipeline as something we'd eventually like to make happen, especially as the total holdings get larger.

Comment author: MvdSteeg 08 August 2018 09:55:50AM 0 points [-]

Thanks for the update!

I'm glad to see the fund balance is up to date now. One piece of the puzzle that does still seem to be missing to me is an indication of how much money the fund managers expect to be able to allocate effectively, as without this the fund balance can only be interpreted relative to past balances and past grants.

I was also curious if there are any plans to register as a non-profit in other countries than the US and the UK. Looking at the registration process for the Netherlands this doesn't seem like a lot of work, and would increase effectiveness of Dutch donations by up to 80% (if I got the math right). Are there some obstacles here that I fail to see which prevent CEA/EA Funds from applying in other countries? Even if you had only a handful of donors per country it seems worthwhile.

Comment author: MarekDuda 08 August 2018 02:31:11PM 1 point [-]

By your first question do you mean whether there is a $ amount ceiling after which the fund managers expect the marginal effectiveness of further grants to drop off? I think if this was the case, the managers would have a good reason to wait to grant until more effective options appear again.

On your second point, we would want to have the platform a little more stable operationally before adding further jurisdictions, as this would increase the complexity non trivially. Nevertheless we do hope to be able to register elsewhere in due time and the Netherlands would likely be our first target.

Comment author: MvdSteeg 08 August 2018 04:17:33PM *  0 points [-]

Regarding the first point: if there's much more incoming donations than they can effectively allocate, I can imagine they automatically lower their standards by some margin so less funds would remain unallocated. If there's no expectation of more effective options showing up in the future that wouldn't get covered by new incoming donations, then this could be seen as a good thing, money spent on a fairly effective charity is better than not spending it at all. However, in addition to the responsibility of allocating the resources they have as well as they can, I think the fund managers have a responsibility to also communicate when they feel like their fund is hitting diminishing returns. This is difficult to quantify, but e.g. last dollar vs first dollar impact ballpark might suffice.

It is very much possible that the funds are -nowhere- near the point where this should cause any worry, and I'm not at all trying to say that any fund manager has neglected to give such a signal, because there may not have been any cause for it. However, I'm not sure if, as a (potential) donor, there's a way for me to tell this right now. Maybe I could deduce it from notes on past grants.

I would expect a red light to be given if at any point a fund manager feels like their fund will not be able to allocate their resources to the standard they hold for themselves. It would be helpful if there was also an explicit green light when this is -not- the case. I'm not sure this can be derived from just the current fund balance, because it does not say anything about future opportunities coming up. We might see a fund is holding 1 million after a series of grants, conclude that this means it cannot use our donations right now, while in fact the next batch of grants could have easily handled another 500k or more.

Glad to hear the Netherlands is high on your priority list in terms of expanding registrations! I don't suppose you're willing to risk any kind of ETA, right? ;)