This is fantastic. I speak just from Ugandan experience here again.
I’m sure there are grassroots orgs that are more effective at the margins than GiveWell’s orgs. Even just from a probability perspective with thousands of small orgs, the chances of a handful being super-effective in their context seems pretty high. I know we are obsessed with scale, but if a small org is effective and it would only take 100k a year to max out what they could do in a local context, why not give them that? As an example there is a great education org here that runs on a shoestring (maybe 60k a year) called Read4Life that has good evidence over a number of years of operation that their interventions have greatly sped up learning of reading and writing for thousands of kids. It’s wild how hard it is for orgs like this to get funding.
I don’t know enough about Givewell to say this with any confidence, butit might not be shard for Givewell to loosely assess say 1000 small orgs and select a handful that might be effective. Obviously with a less rigorous assessment, Givewell couldn’t be confident that the orgs were effective but they could at least offer some smaller orgs as options for donation. I would love like you say for this to be done in-country, but in places like here in Uganda I’m not sure the expertise is really around to do solid Givewell style assessments—but I could be wrong maybe there are people out there who could do it!
There are a some orgs already which can help serve as fiscal sponsors for grassroots organisations, but it’s not straightforward to get connected to them. I’m also not sure just being tax deductable to help people donate though would make a huge difference unless the money coming in was huge—Big Local NGOs here have better ways of getting large amounts of money.
Big local NGOs here in Uganda get the vast majority of their money through grant applications not individuals as well. The biggest 5 or so local NGOs here in Gulu have enormous budgets funded by DANIDA (danish aid), USAID, GIZ etc. This is a bad situation as these donors give close to zero weight to actual effectiveness—instead what mostly matters is the relationships that they build with the donor organisations and their grant writing skills. Also these NGOs just tailor their programs to whichever wind the donor funds are blowing in. Today it’s “Gender based violence”, tomorrow it’s “climate friendly agriculture”. And the programs they roll out when they get the funding I think often do close to zero good or even harm.
My point is that the way to get a lot of money jere as a local NGO isn’t to connect with a personal donor base in another country, but through getting big grants through big donor organisation, either philanthropies or National Aid organisatns.. Maybe EA incentives could help turn this tide, but it would take a LOT of EA money to push local NGOs away from this model of growth and towards more effective programs.
Despite that limitation I LOVE your idea of setting up an EA based fiscal sponsor for potential high impact small local organisations. I don’t particularly like the word “grass roots” though, because I’m never quite sure what it means.
Yeah, I’m not holding out much hope right now for the governmental funders and the large NGOs that rely upon them. The vision was much more modest—finding ways to provide “small” (e.g., high-4 to low-6 figures) funding for smaller nonprofits who have been effectively doing what they have been doing (rather than pivoting to what the politicals want this year).
I was thinking more about the private dollars (e.g., people who give to World Vision, which IIRC you do not rate highly). In particular, I speculate that there is a donor “market” for supporting efforts that come authentically out of the areas whose populations they serve, rather than being dreamt up by and run by Westerners. That is a market the big EA charities cannot compete in. (For my part, I don’t care whether a charity is being run by Westerners or not, but some donors do.) EA is not going to convince everyone of the EA way of thinking, but funneling non-EA dollars into better charities is still a win.
It may be challenging to make cost-effectiveness work out for having paid Western evaluators evaluating microsized charities in Africa. Let’s say we were looking at 1000 charities, 20 of which would clear whatever bar we set, and this could optimistically be done in 1500 hours (because most organizations could be non-selected quickly). Let’s say the 20 have on average $100K in room for funding. At an organization like GiveWell’s cost/overhead rates, that is probably something like a $150K or so project to select the 20 programs.
That level of analysis is not likely to persuade us that it is more likely than not that the bucket of microcharities is more effective than the big charities GiveWell would otherwise recommend. So we’d want to find funders for these 20 that wouldn’t displace (any, or much) funding that would have otherwise gone to a standard GiveWell charity. That sounds plausibly cost-effective by EA standards to me, but it’s a whole lot easier to justify if the bulk of the evaluation work could be done in the lower-income country/region with a final check of the proposed approvals by high-income country sources. The cost for that might be more on the $15K-$25K level? Some of that could potentially be justified by the informational value of having done the survey alone (e.g., we might find an idea that could scale or at least would learn about what local solutions are and aren’t working in a semi-effective manner).
I totally hear that that capacity may not exist in many lower-income countries right now. So I guess one initial pitch would be to find some recent college grads sympahetic to EA ideas, hire them at private-sector local salaries (is that on the order of a few thousand USD in Uganda a year?), and teach them how to do shallow EA-style investigations. Frankly, I think a few programs of that nature would also be an inexpensive way to lay the groundwork for mitigating the reputational issue of EA being viewed as largely a movement of people from certain countries, races, and educational backgrounds.
The less ambitious approach would be to solicit nominations for microcharities to evaluate from people with both EA sympathies and on-the-ground experience, like yourself. That would keep the evaluation time at a much more manageable level, but it would miss a lot of organizations. I’d be willing to consider making donations to small groups on the recommendation of a trusted source vouching for their cost-effectiveness, but I may not be representative of others.
I have assisted, in a non-lawyer capacity, with setting up a US fiscal sponsor for a foreign non-profit entity before . . . in Uganda, curiously! It is fairly easy (and a few hundred dollars) if you’re not expecting to bring in more than $50K in donations routed through the US non-profit in the first three years. Multiple non-profits can share the same fiscal sponsor...although the tax filing gets more complex at 50K and considerably more complex at 100K. (It seems that Read4Life has fiscal-sponsor relationships with an organization having like 200 projects. They have US, UK, and AUS tax deductibility.)
“So I guess one initial pitch would be to find some recent college grads sympahetic to EA ideas, hire them at private-sector local salaries (is that on the order of a few thousand USD in Uganda a year?), and teach them how to do shallow EA-style investigations. Frankly, I think a few programs of that nature would also be an inexpensive way to lay the groundwork for mitigating the reputational issue of EA being viewed as largely a movement of people from certain countries, races, and educational backgrounds.”
My only addition is that you would probably need someone more versed in EA evaluations out here supervising this. Also you’d need really good grads, because this would be a very new concept and would need a lot of learning.
I don’t love the soliciting idea, because like you say it would still favour bigger organisations and would miss a lot of potentially great really small ones.
This is fantastic. I speak just from Ugandan experience here again.
I’m sure there are grassroots orgs that are more effective at the margins than GiveWell’s orgs. Even just from a probability perspective with thousands of small orgs, the chances of a handful being super-effective in their context seems pretty high. I know we are obsessed with scale, but if a small org is effective and it would only take 100k a year to max out what they could do in a local context, why not give them that? As an example there is a great education org here that runs on a shoestring (maybe 60k a year) called Read4Life that has good evidence over a number of years of operation that their interventions have greatly sped up learning of reading and writing for thousands of kids. It’s wild how hard it is for orgs like this to get funding.
I don’t know enough about Givewell to say this with any confidence, butit might not be shard for Givewell to loosely assess say 1000 small orgs and select a handful that might be effective. Obviously with a less rigorous assessment, Givewell couldn’t be confident that the orgs were effective but they could at least offer some smaller orgs as options for donation. I would love like you say for this to be done in-country, but in places like here in Uganda I’m not sure the expertise is really around to do solid Givewell style assessments—but I could be wrong maybe there are people out there who could do it!
There are a some orgs already which can help serve as fiscal sponsors for grassroots organisations, but it’s not straightforward to get connected to them. I’m also not sure just being tax deductable to help people donate though would make a huge difference unless the money coming in was huge—Big Local NGOs here have better ways of getting large amounts of money.
Big local NGOs here in Uganda get the vast majority of their money through grant applications not individuals as well. The biggest 5 or so local NGOs here in Gulu have enormous budgets funded by DANIDA (danish aid), USAID, GIZ etc. This is a bad situation as these donors give close to zero weight to actual effectiveness—instead what mostly matters is the relationships that they build with the donor organisations and their grant writing skills. Also these NGOs just tailor their programs to whichever wind the donor funds are blowing in. Today it’s “Gender based violence”, tomorrow it’s “climate friendly agriculture”. And the programs they roll out when they get the funding I think often do close to zero good or even harm.
My point is that the way to get a lot of money jere as a local NGO isn’t to connect with a personal donor base in another country, but through getting big grants through big donor organisation, either philanthropies or National Aid organisatns.. Maybe EA incentives could help turn this tide, but it would take a LOT of EA money to push local NGOs away from this model of growth and towards more effective programs.
Despite that limitation I LOVE your idea of setting up an EA based fiscal sponsor for potential high impact small local organisations. I don’t particularly like the word “grass roots” though, because I’m never quite sure what it means.
Yeah, I’m not holding out much hope right now for the governmental funders and the large NGOs that rely upon them. The vision was much more modest—finding ways to provide “small” (e.g., high-4 to low-6 figures) funding for smaller nonprofits who have been effectively doing what they have been doing (rather than pivoting to what the politicals want this year).
I was thinking more about the private dollars (e.g., people who give to World Vision, which IIRC you do not rate highly). In particular, I speculate that there is a donor “market” for supporting efforts that come authentically out of the areas whose populations they serve, rather than being dreamt up by and run by Westerners. That is a market the big EA charities cannot compete in. (For my part, I don’t care whether a charity is being run by Westerners or not, but some donors do.) EA is not going to convince everyone of the EA way of thinking, but funneling non-EA dollars into better charities is still a win.
It may be challenging to make cost-effectiveness work out for having paid Western evaluators evaluating microsized charities in Africa. Let’s say we were looking at 1000 charities, 20 of which would clear whatever bar we set, and this could optimistically be done in 1500 hours (because most organizations could be non-selected quickly). Let’s say the 20 have on average $100K in room for funding. At an organization like GiveWell’s cost/overhead rates, that is probably something like a $150K or so project to select the 20 programs.
That level of analysis is not likely to persuade us that it is more likely than not that the bucket of microcharities is more effective than the big charities GiveWell would otherwise recommend. So we’d want to find funders for these 20 that wouldn’t displace (any, or much) funding that would have otherwise gone to a standard GiveWell charity. That sounds plausibly cost-effective by EA standards to me, but it’s a whole lot easier to justify if the bulk of the evaluation work could be done in the lower-income country/region with a final check of the proposed approvals by high-income country sources. The cost for that might be more on the $15K-$25K level? Some of that could potentially be justified by the informational value of having done the survey alone (e.g., we might find an idea that could scale or at least would learn about what local solutions are and aren’t working in a semi-effective manner).
I totally hear that that capacity may not exist in many lower-income countries right now. So I guess one initial pitch would be to find some recent college grads sympahetic to EA ideas, hire them at private-sector local salaries (is that on the order of a few thousand USD in Uganda a year?), and teach them how to do shallow EA-style investigations. Frankly, I think a few programs of that nature would also be an inexpensive way to lay the groundwork for mitigating the reputational issue of EA being viewed as largely a movement of people from certain countries, races, and educational backgrounds.
The less ambitious approach would be to solicit nominations for microcharities to evaluate from people with both EA sympathies and on-the-ground experience, like yourself. That would keep the evaluation time at a much more manageable level, but it would miss a lot of organizations. I’d be willing to consider making donations to small groups on the recommendation of a trusted source vouching for their cost-effectiveness, but I may not be representative of others.
I have assisted, in a non-lawyer capacity, with setting up a US fiscal sponsor for a foreign non-profit entity before . . . in Uganda, curiously! It is fairly easy (and a few hundred dollars) if you’re not expecting to bring in more than $50K in donations routed through the US non-profit in the first three years. Multiple non-profits can share the same fiscal sponsor...although the tax filing gets more complex at 50K and considerably more complex at 100K. (It seems that Read4Life has fiscal-sponsor relationships with an organization having like 200 projects. They have US, UK, and AUS tax deductibility.)
I love this this idea here a lot
“So I guess one initial pitch would be to find some recent college grads sympahetic to EA ideas, hire them at private-sector local salaries (is that on the order of a few thousand USD in Uganda a year?), and teach them how to do shallow EA-style investigations. Frankly, I think a few programs of that nature would also be an inexpensive way to lay the groundwork for mitigating the reputational issue of EA being viewed as largely a movement of people from certain countries, races, and educational backgrounds.”
My only addition is that you would probably need someone more versed in EA evaluations out here supervising this. Also you’d need really good grads, because this would be a very new concept and would need a lot of learning.
I don’t love the soliciting idea, because like you say it would still favour bigger organisations and would miss a lot of potentially great really small ones.