How to improve EA Funds

[Note: I’ve writ­ten about EA Funds be­fore, but I be­lieve there’s enough new here to merit a sep­a­rate post – and it comes as lots of Brits are de­cid­ing where to donate be­fore the UK tax year ends.]

I think EA Funds is an awe­some idea – del­e­gate the hard work of de­cid­ing where to give money to the ex­perts. I’ve donated there in the past, and en­couraged oth­ers to do so, and I’m glad CEA is putting time into the pro­gramme.

How­ever, there are some is­sues with the way the funds are run. My con­cerns sum­marised:

  • The funds hand out money very in­fre­quently, and hold onto money for long pe­ri­ods of time. This erodes the value of the fund through time dis­count­ing. EA orgs have stated that they value dona­tions in a year’s time at a 12% dis­count to re­ceiv­ing them now[1], so this rep­re­sents a sub­stan­tial cost.

  • The funds hold their money as cash, for­go­ing any po­ten­tial in­ter­est the money could earn.

  • There is no sched­ule as to when the money will be handed out. This lack of trans­parency is trou­bling, and pre­vents donors mak­ing in­formed choices (e.g. to give di­rectly to char­i­ties in­stead of wait­ing).

  • (a weaker ob­jec­tion) As the funds hold onto dona­tions for so long, the chances of the fund man­ager’s and donors’ in­ten­tions drift­ing apart is high.

My sug­gested im­prove­ments:

  • Dis­burse funds reg­u­larly (e.g. once ev­ery three months).

  • If they’re not be­ing dis­bursed, hold funds in low-risk in­vest­ments (e.g. gov­ern­ment bonds), not cash.

  • Be trans­par­ent about fu­ture plans and re­al­is­tic about fund man­agers’ ca­pac­ity to ad­minister the fund.

Wait­ing to hand out money erodes its value – so dis­burse at a reg­u­lar cadence

EA Funds cur­rently holds $1.1mil­lion in cash. From the EA Funds web­site, here is the amount of money un­al­lo­cated in each fund, fol­lowed by the per­centage of the to­tal dona­tions that fund has re­ceived which re­main un­al­lo­cated (thanks to Peter Hur­ford for the idea):

  • Global Devel­op­ment: $497,957 [49% of all money it has ever re­ceived]

  • Long-Term Fu­ture: $348,167 [95%]

  • EA Com­mu­nity: $206,271 [71%]

  • An­i­mal Welfare: $75,109 [18%] – though it’s about to make a disbursement

EA or­gani­sa­tions sur­veyed at the end of 2017 said their av­er­age dis­count rate on dona­tions was 12% per year[1] – so the funds hold­ing onto dona­tions in­stead of hand­ing them out has a con­sid­er­able cost. Whether this dis­count rate is ac­cu­rate is an­other ques­tion – given the rel­a­tive abun­dance of cash available to EA orgs (through OpenPhil and Good Ven­tures), a rate as high as this is sur­pris­ing.

Funds are held as cash – in­vest them if they’re not be­ing disbursed

I’ve con­firmed with CEA that the money in the funds is held as cash, and not in­vested. If it was held in e.g. US Trea­sury bonds, it could be earn­ing 1~2% per year at no risk[2].

Keep­ing the funds liquid does have benefits – namely, al­low­ing fund man­agers to make dis­burse­ments at short no­tice – but I don’t fore­see that op­tion be­ing ex­er­cised of­ten. I also ad­mit that it isn’t “free” to in­vest the money in bond, in that there’s op­er­a­tional over­head in­volved, but with such a large amount of money held it seems worth­while.

A lack of trans­parency is bad – be clear about fund man­ager bandwidth

It seems like the ir­reg­u­lar dis­burse­ments from the funds are down largely to fund man­agers not hav­ing suffi­cient time on their hands – and hav­ing much more press­ing com­mit­ments (e.g. man­ag­ing big chunks of Open Phil’s bud­get!).

The funds could take a similar ap­proach to Giv­ing What We Can – al­lo­cate funds to the top char­i­ties in their cause area, and donate to those char­i­ties on a reg­u­lar ba­sis un­til the fund man­ager comes along and up­dates the al­lo­ca­tion. This would pre­vent funds from sit­ting in limbo for long pe­ri­ods of time, and ideally would mir­ror what donors would oth­er­wise be do­ing with the money if EA Funds didn’t ex­ist (i.e. in the ab­sence of any other in­for­ma­tion, give to the top char­i­ties).

(Of course, this is moot if fund man­agers have a good rea­son to defer donat­ing – rather than sim­ply not hav­ing the time to man­age the fund.)

Ad­di­tion­ally, the longer the fund holds on to dona­tions, the more likely it is for the donor’s in­ten­tion and the fund’s di­rec­tion to di­verge (I wrote more about this here). You could ar­gue that this is already hap­pen­ing: when I donate to a fund, I do so in the ex­pec­ta­tion that the money will be handed out rea­son­ably rapidly.

---

[1] – “The size-weighted av­er­age dis­count rate [of EA orgs sur­veyed] for dona­tions was 12%“. https://​​80000hours.org/​​2017/​​11/​​tal­ent-gaps-sur­vey-2017/​​
[2] – See US Trea­sury bond yields https://​​www.bloomberg.com/​​mar­kets/​​rates-bonds/​​gov­ern­ment-bonds/​​us. At the time of writ­ing, 3 month US Trea­sury bonds cur­rently offer a 1.7% an­nu­al­ised yield.