Founder of the AI productivity app https://ââdonethat.ai
Writing about effective habits for a good life at https://ââeuzoia.org including an app that works like beeminder but donates to effective charities at https://ââapp.euzoia.org
GWWC profile about me: https://ââwww.givingwhatwecan.org/ââblog/ââmember-profile-christoph-hartmann
The main point Iâm trying to make is not that we should pay all people market-rate salaries, but that, at least, we should make this assumption in cost effectiveness calculations. Then actually paying those salaries would be the logical next step but maybe a step too far for marketing /â market dynamics reasons.
Let me answer to two points you made:
In my view, there are several reasons for this. One important aspect is probably that many employees attribute the impact, at least partly, to themselves personally.
This is the core of what Iâm trying to get at. I am trying to argue that the imaginary pay cut vs a comparable for-profit job the employees are taking is the relative impact they are having. Therefore when we do cost effectiveness calculations, we should factor that in, so we donât overpromise cost-effectivness to donors. Because employees are also donors, just with their time instead of their money.
Another aspect is certainly that working at a charity is often simply more enjoyable because you are surrounded by inspiring people, typically have more autonomy, and so on.
This I fully agree with. So when comparing wiht for-profit salaries we should probably compare not with cut-throat banks but with companies that have a similar vibe/âculture. I know thatâs hard but some kind of estimate is better than nothing I think.
hahaha yes totally missed this, thank you!!
Hahaha yes I love that compensating paradox. I think the issue with that one is that itâs very narrow-minded and short-term focused. Sure in the short term for this specific intervention compensation might work, but as a movement we lose if we think like this.
Also agree that building movement and activating volunteers is value that probably isnât always accounted for.
Curious: Are you aware of these been used in the field /â cost-effectiveness estimations etc.? Iâm aware of the concept but have never seen it actually being used. But that Iâm also not that deep in that space.
Great opt-in vs opt-out framing, that snarky comment of mine about tricking people was written too fast!
About that example: I think you meant to write âat charity Bâ instead of âat charity Aâ?
* Charity A: 1 employees at market rate, one unit of impact > 1 impact for donor
* Charity B: 1 employees at market rate, one unit of impact + 1 volunteers with 1 impact > 2 impact for donor
I think that the donor should not have â2 counterfactual impactâ in this case. But maybe I am misunderstanding things.
Letâs play with some more examples
* Charity C: 1 employee and 10000 volunteers, 10001 impact total. Should the donor really get 10001 impact? To me that feels really off
* Charity D: The employee also becomes a volunteer but does the same job. In a way the employee is âdonatingâ their salary and becomes the donor. Should the employee now get â2 impactsâ? That again feels off? Are they somehow better than the other volunteer?
* Charity E: 100 donors, all donating $1. Itâs a blind kickstarter campaign and the charity can only do the intervention if they raise 100, otherwise money back, but the donors donât know how many others already donated, so order doesnât matter. In a way each donor should get the full âcounterfactualâ impact because without each individual donor the fundraise wouldâve failed. But again attributing the same full impact 100 times feels wrong.
Yes, I agree about the inspiring vs non-inspiring feelings and how this has impact.
I still feel there must be a way to flip this somehow and find a way to assign and recognise impact and contributions that feels motivating and uplifiting to everybody, but maybe thatâs for another post haha
Thanks for taking the time to write these comments, really appreciate it!
A disadvantage of starting with market rate is that the person who wouldâve accepted the NGO rate might on average pay themselves more because they would more consciously have to give up money.
This is implying we are right now tricking people into giving up more than they consciously would. Iâm not sure thatâs what we would want as a movement.
Most salaries at EA orgs are in the top couple percent of the world, so I donât think this is a big issue.
I think this might vary heavily depending on the focus area. In AI for sure, not sure about the others.
There are many EAs would would like to get a job at an EA org, and lower salaries means that more jobs can be offered. So I think EA âunemploymentâ is a bigger problem for the movement than the people who do have EAs jobs not making enough money.
Iâm aware of the EA unemployment but Iâm not sure if we should optimize for EAs working in EA orgs. Shouldnât the best talent or fit win? Especially for hard-to-fill roles like biorisk researchers, AI alignment people, a GiveDirectly country operation manager role, etc. Of course, value alignment shoudl play a part in assessing job match, but Iâm not sure willingness to take a low salary is the right selector.
Either way, all these points are downstream of the main point Iâm trying to make: We should take market rate equivalents in cost effectivness calculations so we donât overpromise impact to donors. We can do this and then still decide to only accept candidates that take a 50% pay cut.
Yes 100%. Sorry I might have written a bit convoluted but this is one of the points I wanted to make.
I agree that from a personal perspecive, donations and volunteering are very different. I also love volunteering for the community, the energy Iâm getting from it, and, letâs be honest, Iâm getting zero in-the-moment good vibes from my donations, thatâs a different category for me.
However as a donor, when evaluating charities, I do care how they use volunteers. Because in this framing the volunteers should get âpart of the saved fishâ. That leaves less âsaved fishâ for me.
In the thought experiment I gave I meant to assume that both orgs are identical except for that volunteering efficiency. Both are great to volunteer at, feel the same, everything the same, just that, for some reason, one is less efficient at using volunteer time.
Not sure if weâd have to make all donation public. But in the example of Marcus, he did that, he publicly stated that he is not taking salary. Your original claim was you prefer to donate to charities where you know that people take a low salary. The only way to know this is for this to be public.
For this to work weâd only need to know total market-equivalent salaries and total âdonatedâ fraction of this. Doesnât have to be per-staff. This is not very different from the overhead reporting right now and should be possible with privacy laws.
Re marketing: I agree that this differnt view of looking at impact is a bit harder to get across initially but it could eventually be much cleaner: It allows to give a âmathematicalâ quantifiable view on what âsplitâ of the impact I am getting as a donor. At least in this community, thatâs something people want I think.
Looking back I think my example wasnât the best one. But this âthe fish only get saved onceâ is exactly the point Iâm trying to get to.
Take this example: another charity âAtlantic Shepherdâ is doing exactly what you guys do, except they need ten times as many volunteers to raise the same funds. If we go by the model âeverybody has impactâ this would 10x the felt impact. Would you prefer to Atlantic Shepherd or to Sea Shepherd? Your reasoning would imply you would want to donate to Atlantic Shepherd but if we would take the Atlantic Shepherd volunteers and have them work on more efficient causes (like joining Sea Shepherd), we would increase impact. Modeling volunteer time explicitly as donations would help quantify this.
Thanks for your thoughts!
I think what I am proposing here is perfectly compatible with your requirement as a donor: We can make it the norm to pay market salaries but then you as a donor can still say you only donate to charities where the founder or staff donate a significant fraction of that salary back to the charity with salary sacrifices. Thatâs a bit like the common for-profit equivalent for investors to want founders to have âskin in the gameâ by having equity.
Letâs take Marcusâ example (please correct me if I am making wrong assumptions here!): Marcusâ fund will turn donations into impact by regranting. As a donor to the fund, you only get a âcutâ of that impact because the fund did some work (research, payment processing, âŚ). As a donor to the fund I am happy to give them a cut because it saves me time. How big is the cut? I think the foregone salary by Marcus (and potentially other staff) are a good estimate for this. Thatâs why we should model his foregone salary has an implicit donor to the fund and divide the fund impact by all the donations, including his foregone salary. I hope this is making sense!
Thank you for this write-up, I was thinking exactly the same when listening to that latest podcast on mythos.
I created a detailed write-up here: https://ââdonethat.ai/ââdata
tl;dr those cc numbers will never be stored because I donât store the screenshots and the LLMs are prompted to ignore any sensitive info
So you mean when you work on two systems side by side? I thought i had built something to keep it sane but could be that thatâs not working perfectly, have the use case seldomly. Could you share more context with the feedback button in-app?
Thanks! Tried to optimize the charities for âbrand recognitionâ but totally agree, we should at least have one of the top GW charities on there!
Collecting feedback right now and will then work on a next version. On the list so far
One-off goals (potentially auto-resolving with real-world evidence)
E-mail reminders (on top of the app notifications we have right now)
Other charities
Interesting I never thought of that. So a âhedgeâ for direct impact careers?
Iâm feeling some hesitation hereâpeople who take direct impact careers often already implicitly donate by taking a big pay cut, and Iâd see it as a portfolio playâas a movement we will have plenty of impact, even if not all direct âbetsâ work out.
But maybe I am not fully understanding this. For sure it feels like there is something there⌠can you maybe give a more specific example for when/âhow youâd use it?
Agree on this 100%. I originally wanted to bake in that line of reasoning in this article but I felt it was already controversial enough as it is hahaha. This is another example of everybody claiming marginal impact adding up to 10x of the imapct we actually had.
Happy to write a post about this together at some point!