GiveDirectly receives some zakat funds and some unrestricted funds
Zakat funds are used to pay for payments to Muslims
Unrestricted funds are used to pay for overhead and payments to non Muslims.
How do you decide who is Muslim? Probably easiest is a population average, e.g., in Yemen 99.99% are Muslim (according to GD); or you could conduct surveys of targeted populations. Worst case you ask recipients after the fact.
Incidentally this is also the approach used by IRUSA.
Another similar example is when USAID funded GiveDirectly. They are prohibited from spending on alcohol, so GD estimated how much went to alcohol and other donors paid for that portion.
EDIT- see my response to Kaleem for a clearer version of what I say here
One issue with this is it puts GD in a bind with its unrestricted funds if GD wants to be net nondiscriminatory in its payments.
There would be a quantity, X, of Muslim-earmarked funds, such that the entirety of the unrestricted funds going to non-Muslims would have the net effect of the entire pool (earmarked and unrestricted) being disbursed indiscriminately based on need.
However, if the earmarked funds are lower than X, then, if GD wants to avoid be net discriminatory against Muslims, then some of the unrestricted funds would benefit Muslims. But this creates a problem for Zakat funders, because their funding the earmarked pool could indirectly benefit non-Muslims if this frees up more of the unrestricted fund to non-Muslims.
Hi Brad. Thanks for engaging with this quick take. I’ve read your comment multiple times and am struggling to understand what it means. I would appreciate if you could try and re-explain the second and third paragraphs of your comment for me.
My response regarded Ian’s proposal that GiveDirectly solve the problem by using Zakat funds to solely benefit Muslims and then using unrestricted funds to benefit non-Muslims (and operating expenses). The problem from the Zakat-funder’s perspective is whether or not GiveDirectly would use those earmarked funds to “fung” with its unrestricted funds to benefit non-Muslims.
Let’s assume GiveDirectly has a goal of maximizing welfare with its money transfers. So, without a separate fund earmarked for Muslims, let’s say that there would be funding for a million ideal recipients (determined strictly by need, feasibility of conveying funds, and other strictly utilitarian factors), two hundred thousand of those happening to be Muslims and eight hundred thousand of those happening to be non-Muslims.
Let’s modify the hypothetical to say that some of this funding for one hundred thousand of those million ideal recipients is earmarked for Muslims. GiveDirectly could dutifully deploy this funding for these hundred thousand Muslims. With its unrestricted funds for the remaining nine hundred thousand, GiveDirectly can achieve the same disbursement result by transferring to the 100,000 remaining Muslims in its ideal set and transferring to the 800,000 remaining non-Muslims. From this hypothetical, if I donate an earmarked amount to benefit one thousand Muslims, GiveDirectly can shift its unrestricted fund to benefit 1,000 less Muslims and then it can benefit the next 1,000 people that would most benefit, regardless of religion.
Because money is fungible, GiveDirectly can use the earmarked funds to benefit those Muslims it would have helped with unrestricted funds (because some of the ideal recipients on utilitarian grounds will happen to be Muslims) and this will free up other funds to do whatever GiveDirectly finds to be most marginally beneficial (including benefiting non-Muslims). Thus, even giving to a fund earmarked for Muslims operated by GiveDirectly could have the practical effect of benefiting (even primarily benefiting) non-Muslims. This may or may not be acceptable to the Zakat-funder, depending on whether a technical or functional interpretation of requirements is made.
On the other hand, if GiveDirectly did not update based on the pool earmarked for Muslims, it would be failing to act to maximize welfare with non-Zakat funds, thus potentially upsetting its other donors who want cash deployed where it can do the most good.
If there were an organization totally separate from GiveDirectly that distributed funds, this would address some concerns of “funging”, although one cannot wholly prevent other actors from adjusting their behaviors based on such an organization. But with a separate organization whose books is not totally open to GiveDirectly, it would be more likely that the functional effect of donating would have the functional effect of mostly directly benefiting Muslims.
If a religious donor needs influence over the allocation of other donors’ monies in order to make their donation consistent with their religious obligations, that donor really needs to find a religion-specific charity.[1] (From Kaleem’s comment in response to this one, it doesn’t sound like most zakat donors would be in that camp.)
I’d personally characterize a charity that didn’t update based on the existence of zakat-restricted funds as using my (non-zakat) donations in a discriminatory manner, contrary to my own religious and ethical beliefs. So for instance, if 10% of the funds are restricted, and the charity would have distributed to 200,000 Muslims and 800,000 non-Muslims absent the existence of restricted funds, the 200K/800K split needs to be maintained for me to continue giving to that charity.[2] If the balance shifted to maintain the counterfactual impact of the zakat donors, this means that the identity of some recipients funded with non-zakat funds has changed (and we already established in the hypo that giving to the 800,000th non-Muslim would do more good than giving to the 200,001th Muslim on the list). Thus, I don’t think it would be possible for the same organization to satisfy both donors who wanted the counterfactual benefit from their restricted donations to accrue to their religious group and donors like me. I also suspect that my beliefs are fairly common among would-be donors, but can’t cite anything for my hunch.
Accommodating religious beliefs can be a tricky subject to be sure.
If 30% of the funds were restricted, a 300K/700K split would be acceptable. In that universe, it is clear that all the non-restricted funds would be going to the beneficiaries for whom it would do the most good without respect to religion.
I agree with your perspective expressed in the second perspective and further agree that a non-updating charity would be discriminatory and contrary to my values as well.
Not sure I agree with your characterization in the first paragraph. If the spirit of the rule regarding Zakat is that Muslims predominantly benefit, it seems reasonable to question whether an action whose value does not predominantly benefit Muslims (due to the reactions of other actors) is in line with that spirit. If the counterfactual of the world in which you have donated is one in which there are 80% less funds to non-Muslims and 20% less funds to Muslims, I can see why one might say your donation might not be Zakat.
I didn’t intend for the first paragraph to state a personal opinion on zakat or Islamic law (which I am totally unqualified to offer). It’s meant to be more of an up-front conclusion for the rest of the post.
Because there’s no way to satisfy people who want the counterfactual benefit to flow predominately to their co-religionists and to satisfy people like me, people in the former camp should give to organizations that are openly designed and advertised to meet their religious needs. If someone like me gives money to the (hypothetical) Society for Effective, Zakat-Compliant Direct Cash Transfers to Muslims in Poverty, I can’t much complain of surprise that my money was used in a way inconsistent with my non-discrimination values.
Yep, I see that you’re saying it’s unreasonable for Zakat donors to expect their donations not to influence other funders such that their donations counterfactually predominantly benefit Muslims.
I suppose I am just a bit surprised (and, if Kaleem is correct, gladdened) that such donations that may not have the the counterfactual effect of predominantly benefiting Muslims would still qualify as Zakat.
thanks for this—I think I get it now. I think the points relating to the effects on zakat-donors and non-zakat donors are good ones, especially since I hadn’t considered the effect on non-zakat donors a huge amount up until now.
With regards to Zakat donors: I don’t think the majority of muslim donors would find this argument a reason not to donate. The thing they care most about is whether or not the entire amount of zakat they donate is reaching the hands of zakat-eligible recipients. There is a large amount of scholarship around the philosophy of zakat, and group/societal upliftment is the primary non-spiritual goal. So I don’t think the idea that there are spillover effects which benefit non-muslims would be an issue for most donors, since there is a general expectation that people who are not eligible recipients (e.g. Muslims who aren’t poor) will experience positive effects too.
With regards to maximization-oriented non-zakat donors: I’m not sure about this. I think in the scenario where GD somehow ignore the (hopefully massive) new restricted pool of funds, then yeah maybe this means that donating to GD stops being an extremely cost-effective thing to do. But I think the group of people who care very much about this either 1) don’t donate to GD already, since we seem to have many much more cost-effective options available and 2) would be fine with that because it’d be a result of an influx of donations which are contingent on the new program and are counterfactually significant when thinking about “all the money given to effective causes”.
But … It seems unlikely that GD would react that way to this type of influx in restricted funding? Given that the realistic way which this would happen would be that GD set up a new muslim country-specific program (e.g in Bangladesh or Afghanistan), I’d expect unrestricted funds to be used in the same way they’re currently being used with respects to the various programs they already run? Maybe I’m still missing something you’re pointing out here.
I think your point about new org vs GD benefits are right, but maybe also overlook many of the reasons why GD would be an appealing option to muslim donors (low admin cost and transaction costs, high transparency and accountability, track record) whereas setting up a new org means that these things aren’t there immediately and might never be there to the same degree
I think that Givedirectly, where it has free hands, will try to direct cash to where it can do the most good. If many of the world’s poorest are being served by the Zakat program, this will probably affect choices to some extent at a macro or micro level.
For instance, perhaps counterfactually to the Zakat-funded Bangladeshi program, such a program would have been funded with unrestricted funds (such unrestricted funds then being able to go elsewhere).
But I have no special insight into Givedirectly, just the general observation that if you earmark funds for anything that would otherwise be covered by unrestricted funds, that simply frees up those funds for the org’s marginal priorities.
Re Zakat-donors: if they have no issue with their donations functionally benefiting non-Muslims too, that’s great. I too would rather it all go under Givedirectly given its strengths.
I agree with the three bullet points—using unrestricted or dedicated non-zakat donations to cover operating costs is likely the best way to do this. Additionally:
Determining who is Muslim is a non-trivial and probably impossible thing to do, which is why I would probably just punt that to whichever external orgs we approach to get the program zakat- certified. They’re likely just going to do what they did when they zakat certified GD’s Yemen program, which is to look at the national or regional demographic data, as you suggested. This also prevents GD from having to ask recipients if they’re muslim or not, which I think would be a bad look for them and probably not be good for their overall trustworthiness amongst donors and non-muslim potential receipts in the future.
You can’t do the after-the-fact or USAID alcohol approach because zakat has to be paid directly to muslim recipients—it’s not got anything to do with whether or not the money is used to pay for halaal/haraam things.
Seems to me that the obvious solution here is:
GiveDirectly receives some zakat funds and some unrestricted funds
Zakat funds are used to pay for payments to Muslims
Unrestricted funds are used to pay for overhead and payments to non Muslims.
How do you decide who is Muslim? Probably easiest is a population average, e.g., in Yemen 99.99% are Muslim (according to GD); or you could conduct surveys of targeted populations. Worst case you ask recipients after the fact.
Incidentally this is also the approach used by IRUSA.
Another similar example is when USAID funded GiveDirectly. They are prohibited from spending on alcohol, so GD estimated how much went to alcohol and other donors paid for that portion.
EDIT- see my response to Kaleem for a clearer version of what I say here
One issue with this is it puts GD in a bind with its unrestricted funds if GD wants to be net nondiscriminatory in its payments.
There would be a quantity, X, of Muslim-earmarked funds, such that the entirety of the unrestricted funds going to non-Muslims would have the net effect of the entire pool (earmarked and unrestricted) being disbursed indiscriminately based on need.
However, if the earmarked funds are lower than X, then, if GD wants to avoid be net discriminatory against Muslims, then some of the unrestricted funds would benefit Muslims. But this creates a problem for Zakat funders, because their funding the earmarked pool could indirectly benefit non-Muslims if this frees up more of the unrestricted fund to non-Muslims.
Hi Brad. Thanks for engaging with this quick take. I’ve read your comment multiple times and am struggling to understand what it means. I would appreciate if you could try and re-explain the second and third paragraphs of your comment for me.
Hi Kaleem,
Sorry, I wrote my previous response quickly.
My response regarded Ian’s proposal that GiveDirectly solve the problem by using Zakat funds to solely benefit Muslims and then using unrestricted funds to benefit non-Muslims (and operating expenses). The problem from the Zakat-funder’s perspective is whether or not GiveDirectly would use those earmarked funds to “fung” with its unrestricted funds to benefit non-Muslims.
Let’s assume GiveDirectly has a goal of maximizing welfare with its money transfers. So, without a separate fund earmarked for Muslims, let’s say that there would be funding for a million ideal recipients (determined strictly by need, feasibility of conveying funds, and other strictly utilitarian factors), two hundred thousand of those happening to be Muslims and eight hundred thousand of those happening to be non-Muslims.
Let’s modify the hypothetical to say that some of this funding for one hundred thousand of those million ideal recipients is earmarked for Muslims. GiveDirectly could dutifully deploy this funding for these hundred thousand Muslims. With its unrestricted funds for the remaining nine hundred thousand, GiveDirectly can achieve the same disbursement result by transferring to the 100,000 remaining Muslims in its ideal set and transferring to the 800,000 remaining non-Muslims. From this hypothetical, if I donate an earmarked amount to benefit one thousand Muslims, GiveDirectly can shift its unrestricted fund to benefit 1,000 less Muslims and then it can benefit the next 1,000 people that would most benefit, regardless of religion.
Because money is fungible, GiveDirectly can use the earmarked funds to benefit those Muslims it would have helped with unrestricted funds (because some of the ideal recipients on utilitarian grounds will happen to be Muslims) and this will free up other funds to do whatever GiveDirectly finds to be most marginally beneficial (including benefiting non-Muslims). Thus, even giving to a fund earmarked for Muslims operated by GiveDirectly could have the practical effect of benefiting (even primarily benefiting) non-Muslims. This may or may not be acceptable to the Zakat-funder, depending on whether a technical or functional interpretation of requirements is made.
On the other hand, if GiveDirectly did not update based on the pool earmarked for Muslims, it would be failing to act to maximize welfare with non-Zakat funds, thus potentially upsetting its other donors who want cash deployed where it can do the most good.
If there were an organization totally separate from GiveDirectly that distributed funds, this would address some concerns of “funging”, although one cannot wholly prevent other actors from adjusting their behaviors based on such an organization. But with a separate organization whose books is not totally open to GiveDirectly, it would be more likely that the functional effect of donating would have the functional effect of mostly directly benefiting Muslims.
If a religious donor needs influence over the allocation of other donors’ monies in order to make their donation consistent with their religious obligations, that donor really needs to find a religion-specific charity.[1] (From Kaleem’s comment in response to this one, it doesn’t sound like most zakat donors would be in that camp.)
I’d personally characterize a charity that didn’t update based on the existence of zakat-restricted funds as using my (non-zakat) donations in a discriminatory manner, contrary to my own religious and ethical beliefs. So for instance, if 10% of the funds are restricted, and the charity would have distributed to 200,000 Muslims and 800,000 non-Muslims absent the existence of restricted funds, the 200K/800K split needs to be maintained for me to continue giving to that charity.[2] If the balance shifted to maintain the counterfactual impact of the zakat donors, this means that the identity of some recipients funded with non-zakat funds has changed (and we already established in the hypo that giving to the 800,000th non-Muslim would do more good than giving to the 200,001th Muslim on the list). Thus, I don’t think it would be possible for the same organization to satisfy both donors who wanted the counterfactual benefit from their restricted donations to accrue to their religious group and donors like me. I also suspect that my beliefs are fairly common among would-be donors, but can’t cite anything for my hunch.
Accommodating religious beliefs can be a tricky subject to be sure.
There’s an exception in at least some cases: if the other donors specifically and knowingly consent to that influence.
If 30% of the funds were restricted, a 300K/700K split would be acceptable. In that universe, it is clear that all the non-restricted funds would be going to the beneficiaries for whom it would do the most good without respect to religion.
I agree with your perspective expressed in the second perspective and further agree that a non-updating charity would be discriminatory and contrary to my values as well.
Not sure I agree with your characterization in the first paragraph. If the spirit of the rule regarding Zakat is that Muslims predominantly benefit, it seems reasonable to question whether an action whose value does not predominantly benefit Muslims (due to the reactions of other actors) is in line with that spirit. If the counterfactual of the world in which you have donated is one in which there are 80% less funds to non-Muslims and 20% less funds to Muslims, I can see why one might say your donation might not be Zakat.
*Note I know very little about Islamic law, etc.
I didn’t intend for the first paragraph to state a personal opinion on zakat or Islamic law (which I am totally unqualified to offer). It’s meant to be more of an up-front conclusion for the rest of the post.
Because there’s no way to satisfy people who want the counterfactual benefit to flow predominately to their co-religionists and to satisfy people like me, people in the former camp should give to organizations that are openly designed and advertised to meet their religious needs. If someone like me gives money to the (hypothetical) Society for Effective, Zakat-Compliant Direct Cash Transfers to Muslims in Poverty, I can’t much complain of surprise that my money was used in a way inconsistent with my non-discrimination values.
Yep, I see that you’re saying it’s unreasonable for Zakat donors to expect their donations not to influence other funders such that their donations counterfactually predominantly benefit Muslims.
I suppose I am just a bit surprised (and, if Kaleem is correct, gladdened) that such donations that may not have the the counterfactual effect of predominantly benefiting Muslims would still qualify as Zakat.
thanks for this—I think I get it now. I think the points relating to the effects on zakat-donors and non-zakat donors are good ones, especially since I hadn’t considered the effect on non-zakat donors a huge amount up until now.
With regards to Zakat donors: I don’t think the majority of muslim donors would find this argument a reason not to donate. The thing they care most about is whether or not the entire amount of zakat they donate is reaching the hands of zakat-eligible recipients. There is a large amount of scholarship around the philosophy of zakat, and group/societal upliftment is the primary non-spiritual goal. So I don’t think the idea that there are spillover effects which benefit non-muslims would be an issue for most donors, since there is a general expectation that people who are not eligible recipients (e.g. Muslims who aren’t poor) will experience positive effects too.
With regards to maximization-oriented non-zakat donors: I’m not sure about this. I think in the scenario where GD somehow ignore the (hopefully massive) new restricted pool of funds, then yeah maybe this means that donating to GD stops being an extremely cost-effective thing to do. But I think the group of people who care very much about this either 1) don’t donate to GD already, since we seem to have many much more cost-effective options available and 2) would be fine with that because it’d be a result of an influx of donations which are contingent on the new program and are counterfactually significant when thinking about “all the money given to effective causes”.
But … It seems unlikely that GD would react that way to this type of influx in restricted funding? Given that the realistic way which this would happen would be that GD set up a new muslim country-specific program (e.g in Bangladesh or Afghanistan), I’d expect unrestricted funds to be used in the same way they’re currently being used with respects to the various programs they already run? Maybe I’m still missing something you’re pointing out here.
I think your point about new org vs GD benefits are right, but maybe also overlook many of the reasons why GD would be an appealing option to muslim donors (low admin cost and transaction costs, high transparency and accountability, track record) whereas setting up a new org means that these things aren’t there immediately and might never be there to the same degree
I think that Givedirectly, where it has free hands, will try to direct cash to where it can do the most good. If many of the world’s poorest are being served by the Zakat program, this will probably affect choices to some extent at a macro or micro level.
For instance, perhaps counterfactually to the Zakat-funded Bangladeshi program, such a program would have been funded with unrestricted funds (such unrestricted funds then being able to go elsewhere).
But I have no special insight into Givedirectly, just the general observation that if you earmark funds for anything that would otherwise be covered by unrestricted funds, that simply frees up those funds for the org’s marginal priorities.
Re Zakat-donors: if they have no issue with their donations functionally benefiting non-Muslims too, that’s great. I too would rather it all go under Givedirectly given its strengths.
Thanks Ian.
I agree with the three bullet points—using unrestricted or dedicated non-zakat donations to cover operating costs is likely the best way to do this. Additionally:
Determining who is Muslim is a non-trivial and probably impossible thing to do, which is why I would probably just punt that to whichever external orgs we approach to get the program zakat- certified. They’re likely just going to do what they did when they zakat certified GD’s Yemen program, which is to look at the national or regional demographic data, as you suggested. This also prevents GD from having to ask recipients if they’re muslim or not, which I think would be a bad look for them and probably not be good for their overall trustworthiness amongst donors and non-muslim potential receipts in the future.
You can’t do the after-the-fact or USAID alcohol approach because zakat has to be paid directly to muslim recipients—it’s not got anything to do with whether or not the money is used to pay for halaal/haraam things.