Reading the comments below the linked video… there were responses written 2 years ago that author didn’t have time to reply… How specifically does donating anti-malaria nets “keep populations dependent, economically weak, and slaves to the whims of international donors”?
I would understand the part about whims: yeah, tomorrow some influential organization might decide that the nets are bad and you should stop supporting them, and there would be no more nets. Still, would that outcome be worse compared to a parallel reality, where there never was a movement to support the anti-malaria nets? The people would still have gained a few years of health.
Are the donated nets ruining a previously existing huge local anti-malaria-net industry?
there are systemic issues that make things tend to become scams, and charity evaluators aren’t in a better position with respect to this problem than charities themselves, one should expect charity evaluators to become scams as well.
This part feels true. Similarly how in medicine people started reading meta-reviews, and soon the homeopaths in addition to their studies also started producing their own meta-reviews supporting their own conclusions… as soon as charity evaluation becomes a generally known thing, some of the current ineffective charities will produce their own charity evaluators which will support whatever needs to be supported.
It’s just, instead of “GiveWell will become a scam”, to me the more likely scenario seems “in a few years there will be so many charity evaluator scams that when you google for ‘charity evaluator’, you won’t find GiveWell on the first three pages of results”.
Looks like some of comments were replied to, and most of the comments were either missing the point (e.g. thinking this is about overhead; the relevant section of the video was using overhead as an example of something that could explain efficiency differences) or pretty vague (“these arguments have been debunked by these people” but not being specific about where; I checked my copy of Doing Good Better and I couldn’t find anything about dependency). Is there a specific objection you think was not addressed? Not sure if you have watched the full “Dependency” section of the video, from 35:56 to 42:28; this is the part of the video I would want to see a good response to.
One of the replies does address the issue of local net makers:
“If you want hard evidence, I see it every day, in people that refuse to improve their own station in life because for their whole life charities have come do it for them. I see it in the fact that you cannot buy quality mosquito nets here. If you actually care about your family’s health you cannot do anything about it, because AMF and other such organizations have put the local net makers out of business. They increase apathy.
Additionally, the good they do is tentative at best, I know many families that only bring out their bed net when they know that the AMF inspector are coming around. Since they only come every six months, this is quite easy. Other families hang the nets in an area that they do not actually sleep (most families don’t have beds here). AMF does not invest in training the actual community health workers that are there every day and could not only work to convince the families that this was something important, but could do data collection for the organization for many years to come. But if they did this, they would work themselves out of a job and actually have data that their nets break down after four or five years.”
It might seem strange to take YouTube comments at face value, but my sense is that this guy has a lot of epistemic privilege by virtue of (a) working on international development in one of the locations that net distributions happen in and (b) being able to think well enough (and be honest enough) to make lots of pretty good philosophy videos. Additionally, everything he says seems quite likely according to Econ 101 models.
It is a bit costly for me to review a 48 minute video without a summary, but here it goes:
[8:00] If you’re donating a dollar, according to Singer you would not want 99 cents to go to overhead costs and only one cent to go to the actual process of saving lives. Where should you donate your 5 dollars, if you could choose between a program that would take $4.95 to pay their employees and cover other costs, and only take the five cents to pay for bed nets; or one that spends the $4.95 on the bed nets, and only five cents on the overheads? You probably want to pick the latter, at least according to Singer. And this criterion is going to naturally privilege some causes over others because we’re focusing on that specific impact that physical tangible impact that you’re having.
This seems like missing the point (surprisingly, right after having described how consequentialism means that only the consequences matter), or perhaps it is a motte-and-bailey statement.
The overhead is not the problem. I believe this is actually one of the central messages of the effective altruism: what matters is what the money as a whole causes to happen. Overhead is a part of the equation. Yes, if we—ceteris paribus—just randomly increase the overhead for no good reason, it obviously makes us less effective. But if spending 20% on overhead instead of 10% would double the effect of the remaining 80% (for example by hiring more qualified people, or double-checking things to prevent theft), then increasing the overhead that way would be the right thing to do. I strongly believe Singer would approve this.
So, the motte of the statement is that if we have two processes that convert money to anti-malaria nets in exactly the same ratio, only one of them also has a 5% administrative overhead and the other has a 95% overhead, it is better to choose the former. The bailey of the statement would be concluding that...
Of the eight charities listed on the [GiveWell] web page, seven focused solely on directly providing health care: providing bed nets, providing cures for various diseases, and so on. These programs had the lowest costs and according to Singer the highest return in lives saved and pain averted.
No, it’s not about “the least overhead is when we provide health care, therefore health care is what we should do”. It is about the ratio between the donated dollars and generated outcome (where “many lives saved” is considered to be a quite impressive outcome). The overhead is a red herring.
[11:50] I will argue in fact that does more harm than good. I will claim that charities which minimize overhead actually are less effective than those that use their funds to address other concerns.
Sigh. Go on, fight the straw-Singer!
Then the video explains how donating X to a country can ruin the local X-providing industry. Duh. That’s why I’m asking: Where is the local African anti-malaria-net industry that is being so thoughtlessly ruined by GiveWell?
Because I understand how donating food ruins the local food producers, or how donating t-shirts ruins the local t-shirt producers, so I would also understand how donating anti-malaria nets would ruin the local anti-malaria-net producers… the question is, do these “local anti-malaria-net producers” exist in real life, or only as a part of the thought experiment? What fraction of African population is employed in this industry? (My uninformed prejudices say “probably several orders of magnitude less than in local food production”, but I may be wrong. Please give me data, not thought experiments designed to prove your point.) Because I believe there is a number X such that “X people saved from malaria” outweighs “1 person losing a job”, even in the extreme case where the person losing the job would literally starve to death. (By the way, what about those people who now don’t die from malaria? What if they also take someone’s job?)
[16:15] There’s a mosquito-net maker in Africa. He manufactures around 500 nets a week, employs ten people who as with many African countries each have to support upwards of 15 relatives. As someone that lives in West Africa, I can corroborate that however hard they work they can’t make enough nets to combat the malaria-carrying mosquitoes.
Okay, so here is the local industry. I feel uncertain about the “×15” multiplier for the supported relatives, as an argument used in weighing the benefit of “more people saved from malaria” vs “local net producers not going out of business”. Some of those people dying from malaria also have relatives they support, don’t they? On the other hand, some of them are the supported relatives. (If I overthink it, some of them might even be the supported relatives of the local net producers.)
Now the argument goes: GiveWell sends the nets, puts the local producers (and their ×15 families) to poverty, and “in a maximum of five years the majority of the imported Nets will be torn, damaged, and of no further use”. Now we have 165 more people depending on foreign aid. (Hey, what about those whose lives were saved from malaria? Some of them will depend on foreign aid, too!) Also, no one will restart the local net industry, because now it is seen as a risky business.
The problem I have with this line of thought is that, hypothetically speaking, if I had a magic wand I could just wave and make malaria disappear forever… that would also be an evil thing to do. There would still be the 165 people depending on the foreign aid, right?
Then goes the general argument that by giving people specific aid, we are depriving them of the freedom to choose which aid they would prefer to get. In general, this is a good point, and there is a charity called Give Directly which addresses it… oh crap, donating cash to people makes them dependent, too! :( It seems like even keeping someone alive makes them dependent, because in the future, such person will require food, anti-malaria nets, etc.
Seems like the recommended solution is… to let Africa solve it’s own problems, without any kind of aid. Because that way, the solution will be sustainable. This will also prevent “brain drain”, because the smartest people will be motivated to keep living in the shitty environment, if they believe they are the only ones who can save it. (Win/win! Now even the jobs of the Westerners are safe.) Then those smart people will invest their savings in their countries of origin, and everything will become exponentially better.
[43:45] Singer and Give Well underestimate the amount of good that is done and pleasure created by bringing someone out of dependence.
Okay, here is a thought experiment: Your family is sick, you go to the doctor, and the doctors tells you: “I could give you a medicine, but imagine how much better it would feel to invent it for yourself! Yeah, it may cost a lot, and your family may die while you are researching, but I am giving you the long-term perspective here, for your own good.”
What I am trying to say is that there is the trade-off between the pleasures of being independent and the pleasures of having your relatives alive. Speaking for myself… well, I actually don’t think my country is economically independent, and I definitely wouldn’t trade my kids’ lives to make it so.
But perhaps the next time I will see a person suffering, I will remember that it is a superior option to just let them be, and not take away their motivation to become a well-paid software developer.
Additionally, everything he says seems quite likely according to Econ 101 models.
To me it seems like the Just-World Hypothesis. Specifically the part about how even donating cash to someone makes their life ultimately worse, feels like a status quo worship.
By this logic, there is no way to help another person, ever, without inflicting on them a horrible curse. You give your kids an Xmas present, and you just ruined their motivation to become financially independent. You help an old lady to cross the street, and you ruined her motivation to maintain her vision or to keep good relationships with her relatives. You invent the virus that kills mosquitoes worldwide, and you deprived the Africans of their motivation to study medicine. Any help is just a harm in disguise. (Or perhaps this only applies to helping Africans? Dunno.)
Sorry if you unnecessarily reviewed the whole video, I mostly wanted to point you at the 7-minute “Dependency” section.
The overhead is not the problem.
You’re missing the point in the same way the comment I called out was. He is giving an example of 2 charities and giving overhead as a possible reason one would be more effective than the other. In fact, if bednets have a constant price, than a charity that spends 99% of its money on bednets is more effective than one that spends 1% (as long as the bednets get distributed equally well etc).
No, it’s not about “the least overhead is when we provide health care, therefore health care is what we should do”.
He said costs not overhead.
Then the video explains how donating X to a country can ruin the local X-providing industry. Duh. That’s why I’m asking: Where is the local African anti-malaria-net industry that is being so thoughtlessly ruined by GiveWell?
I just quoted him talking about it, in the previous comment. Have you checked yourself whether there was an anti-malaria-net industry?
What fraction of African population is employed in this industry?
This is not the relevant number, the relevant number is more like “counterfactually on no aid, how many bednets would the local industry be producing.”
Seems like the recommended solution is… to let Africa solve it’s own problems, without any kind of aid.
Where in the text do you find this? Quoting 45:30 in the video:
“You can measure the amount of choice that someone has in a process. You can work on projects that decrease dependency and build capacity. The problem is that this measurement and these projects come with overhead costs. They don’t save the most lives for the dollar, but they do the work in sustainable ways. They inspire populations to try doing the work themselves. Even if that costs more now, it will be cheaper in the long run.”
What I am trying to say is that there is the trade-off between the pleasures of being independent and the pleasures of having your relatives alive.
The independence point is about long-term economic growth (causing more pleasure down the line), not being happy that you’re less dependent.
By this logic, there is no way to help another person, ever, without inflicting on them a horrible curse.
Again, not supported in the text. Why would this person work in international development if he thought this?
In general this comment fails to engage with the strongest part of the video, which is the 7-minute “Dependency” section I mentioned, in which he says the problems caused by aid are extremely bad in some of the countries that are targets of aid (like, they essentially destroy people’s motivation to solve their community’s problems).
[16:15] There’s a mosquito-net maker in Africa. He manufactures around 500 nets a week, employs ten people who as with many African countries each have to support upwards of 15 relatives. As someone that lives in West Africa, I can corroborate that however hard they work they can’t make enough nets to combat the malaria-carrying mosquitoes.
This is specifically the part that my understanding of Econ 101 fails to process.
There is this one guy and his 10 employees, and they can’t make enough nets for the whole Africa. Okay, that part is simple to imagine. But why doesn’t he employ more people and increase the production? Or why someone else doesn’t copy his business model?
If my memory serves me well, Econ 101 assumes that if there is an effective demand, sooner or later there will also be the supply to match it.
Should I assume that there is no effective demand to buy more nets? Like, perhaps people are not aware of the dangers of malaria-carrying mosquitoes, or they don’t believe the nets are helpful, or they simply do not have enough money to match the production costs of the nets: either because the nets are too expensive, or because they have to prioritize other necessities. But then, the problem is not that they “can’t make enough nets”, but rather that they “can’t sell enough nets”.
Another thing about Econ 101 is the principle of comparative advantage. According to this principle, trying to produce everything at home is worse than trading internationally. (Otherwise an embargo would be a blessing instead of a punishment.) But an international trade will inevitably put some of your local producers out of business. Is it possible that Africa has a comparative disadvantage at producing anti-malaria nets?
It seems to me that the author is pattern-matching food aid to nets aid. But that is a different situation. In a situation without foreign aid, you cannot have a long-term imbalance between local food production and local food needs… the starving people will die. But you can have a long-term imbalance between local anti-malaria net production and local anti-malaria net needs… people unprotected against malaria only get it with some probability, not certainty; some who get malaria will die but some will survive. In other words, the resulting balance can’t include people who don’t eat, but it can include people who are not protected against malaria but still some of them survive until they can reproduce. Foreign food aid tries to solve a temporary imbalance; and in the process perhaps introduces more harm than good. But foreign anti-malaria-net aid tries to change the long-term balance.
the problems caused by aid are extremely bad in some of the countries that are targets of aid (like, they essentially destroy people’s motivation to solve their community’s problems).
I understand how an intervention that puts half of your population out of business can have this effect. I find it less likely that an intervention that puts one person in a million out of business would have the same effect. That is why I asked how many people are employed in the anti-malaria-net industry, compared with agriculture.
This seems to me like a mistaken pattern-matching. Pretty much anything can make someone lose their job. But there is a difference between “save thousand people, destroy thousand jobs” (food aid) and “save thousand people, destroy one job” (anti-malaria nets).
Re the section you quoted: if you watched for 1 minute longer you would see that the issue is that the local net manufacturers can’t scale up because they would have to compete with free nets, so the local infrastructure atrophies. (Absent aid, they could scale up, it would just take longer; scaling up might require additional capital or training)
The issue isn’t number of jobs. The issue is (a) infrastructure to solve your own problems and (b) motivation to solve your own problems rather than waiting to have someone else solve them for you. This is all covered in the “Dependency” section of the video, which I am still not convinced you have watched.
EAs say they don’t care about overhead per se and just care about outcomes, but public-facing recommenders like GiveWell and CEA keep recommending charities with low overhead relative to programmatic spending, such as AMF and GiveDirectly, rather than charities that do the sort of hard-to-account-for institution-building which low-overhead charities depend on.
EAs are good at explaining why you shouldn’t do what they’re (we’re?) doing. That’s different than actually doing the right thing.
EAs are good at explaining why you shouldn’t do what they’re (we’re?) doing. That’s different than actually doing the right thing.
I agree (with the second part, at least).
The author of the video said “according to Singer” repeatedly, so I assumed he also disagrees with what EAs are saying. If the real objection was “Singer says to do the right thing, but then actually does exactly the thing he said was wrong”, I didn’t get that message from the video. (Maybe the problem is on my side.)
Singer also recommends charities like the GW top charities, specifically mentions GiveWell approvingly, does cost-benefit analyses that ignore social & physical infrastructure, etc. These are also things Singer is saying.
Reading the comments below the linked video… there were responses written 2 years ago that author didn’t have time to reply… How specifically does donating anti-malaria nets “keep populations dependent, economically weak, and slaves to the whims of international donors”?
I would understand the part about whims: yeah, tomorrow some influential organization might decide that the nets are bad and you should stop supporting them, and there would be no more nets. Still, would that outcome be worse compared to a parallel reality, where there never was a movement to support the anti-malaria nets? The people would still have gained a few years of health.
Are the donated nets ruining a previously existing huge local anti-malaria-net industry?
This part feels true. Similarly how in medicine people started reading meta-reviews, and soon the homeopaths in addition to their studies also started producing their own meta-reviews supporting their own conclusions… as soon as charity evaluation becomes a generally known thing, some of the current ineffective charities will produce their own charity evaluators which will support whatever needs to be supported.
It’s just, instead of “GiveWell will become a scam”, to me the more likely scenario seems “in a few years there will be so many charity evaluator scams that when you google for ‘charity evaluator’, you won’t find GiveWell on the first three pages of results”.
Looks like some of comments were replied to, and most of the comments were either missing the point (e.g. thinking this is about overhead; the relevant section of the video was using overhead as an example of something that could explain efficiency differences) or pretty vague (“these arguments have been debunked by these people” but not being specific about where; I checked my copy of Doing Good Better and I couldn’t find anything about dependency). Is there a specific objection you think was not addressed? Not sure if you have watched the full “Dependency” section of the video, from 35:56 to 42:28; this is the part of the video I would want to see a good response to.
One of the replies does address the issue of local net makers:
“If you want hard evidence, I see it every day, in people that refuse to improve their own station in life because for their whole life charities have come do it for them. I see it in the fact that you cannot buy quality mosquito nets here. If you actually care about your family’s health you cannot do anything about it, because AMF and other such organizations have put the local net makers out of business. They increase apathy.
Additionally, the good they do is tentative at best, I know many families that only bring out their bed net when they know that the AMF inspector are coming around. Since they only come every six months, this is quite easy. Other families hang the nets in an area that they do not actually sleep (most families don’t have beds here). AMF does not invest in training the actual community health workers that are there every day and could not only work to convince the families that this was something important, but could do data collection for the organization for many years to come. But if they did this, they would work themselves out of a job and actually have data that their nets break down after four or five years.”
It might seem strange to take YouTube comments at face value, but my sense is that this guy has a lot of epistemic privilege by virtue of (a) working on international development in one of the locations that net distributions happen in and (b) being able to think well enough (and be honest enough) to make lots of pretty good philosophy videos. Additionally, everything he says seems quite likely according to Econ 101 models.
It is a bit costly for me to review a 48 minute video without a summary, but here it goes:
This seems like missing the point (surprisingly, right after having described how consequentialism means that only the consequences matter), or perhaps it is a motte-and-bailey statement.
The overhead is not the problem. I believe this is actually one of the central messages of the effective altruism: what matters is what the money as a whole causes to happen. Overhead is a part of the equation. Yes, if we—ceteris paribus—just randomly increase the overhead for no good reason, it obviously makes us less effective. But if spending 20% on overhead instead of 10% would double the effect of the remaining 80% (for example by hiring more qualified people, or double-checking things to prevent theft), then increasing the overhead that way would be the right thing to do. I strongly believe Singer would approve this.
So, the motte of the statement is that if we have two processes that convert money to anti-malaria nets in exactly the same ratio, only one of them also has a 5% administrative overhead and the other has a 95% overhead, it is better to choose the former. The bailey of the statement would be concluding that...
No, it’s not about “the least overhead is when we provide health care, therefore health care is what we should do”. It is about the ratio between the donated dollars and generated outcome (where “many lives saved” is considered to be a quite impressive outcome). The overhead is a red herring.
Sigh. Go on, fight the straw-Singer!
Then the video explains how donating X to a country can ruin the local X-providing industry. Duh. That’s why I’m asking: Where is the local African anti-malaria-net industry that is being so thoughtlessly ruined by GiveWell?
Because I understand how donating food ruins the local food producers, or how donating t-shirts ruins the local t-shirt producers, so I would also understand how donating anti-malaria nets would ruin the local anti-malaria-net producers… the question is, do these “local anti-malaria-net producers” exist in real life, or only as a part of the thought experiment? What fraction of African population is employed in this industry? (My uninformed prejudices say “probably several orders of magnitude less than in local food production”, but I may be wrong. Please give me data, not thought experiments designed to prove your point.) Because I believe there is a number X such that “X people saved from malaria” outweighs “1 person losing a job”, even in the extreme case where the person losing the job would literally starve to death. (By the way, what about those people who now don’t die from malaria? What if they also take someone’s job?)
Okay, so here is the local industry. I feel uncertain about the “×15” multiplier for the supported relatives, as an argument used in weighing the benefit of “more people saved from malaria” vs “local net producers not going out of business”. Some of those people dying from malaria also have relatives they support, don’t they? On the other hand, some of them are the supported relatives. (If I overthink it, some of them might even be the supported relatives of the local net producers.)
Now the argument goes: GiveWell sends the nets, puts the local producers (and their ×15 families) to poverty, and “in a maximum of five years the majority of the imported Nets will be torn, damaged, and
of no further use”. Now we have 165 more people depending on foreign aid. (Hey, what about those whose lives were saved from malaria? Some of them will depend on foreign aid, too!) Also, no one will restart the local net industry, because now it is seen as a risky business.
The problem I have with this line of thought is that, hypothetically speaking, if I had a magic wand I could just wave and make malaria disappear forever… that would also be an evil thing to do. There would still be the 165 people depending on the foreign aid, right?
Then goes the general argument that by giving people specific aid, we are depriving them of the freedom to choose which aid they would prefer to get. In general, this is a good point, and there is a charity called Give Directly which addresses it… oh crap, donating cash to people makes them dependent, too! :( It seems like even keeping someone alive makes them dependent, because in the future, such person will require food, anti-malaria nets, etc.
Seems like the recommended solution is… to let Africa solve it’s own problems, without any kind of aid. Because that way, the solution will be sustainable. This will also prevent “brain drain”, because the smartest people will be motivated to keep living in the shitty environment, if they believe they are the only ones who can save it. (Win/win! Now even the jobs of the Westerners are safe.) Then those smart people will invest their savings in their countries of origin, and everything will become exponentially better.
Okay, here is a thought experiment: Your family is sick, you go to the doctor, and the doctors tells you: “I could give you a medicine, but imagine how much better it would feel to invent it for yourself! Yeah, it may cost a lot, and your family may die while you are researching, but I am giving you the long-term perspective here, for your own good.”
What I am trying to say is that there is the trade-off between the pleasures of being independent and the pleasures of having your relatives alive. Speaking for myself… well, I actually don’t think my country is economically independent, and I definitely wouldn’t trade my kids’ lives to make it so.
But perhaps the next time I will see a person suffering, I will remember that it is a superior option to just let them be, and not take away their motivation to become a well-paid software developer.
To me it seems like the Just-World Hypothesis. Specifically the part about how even donating cash to someone makes their life ultimately worse, feels like a status quo worship.
By this logic, there is no way to help another person, ever, without inflicting on them a horrible curse. You give your kids an Xmas present, and you just ruined their motivation to become financially independent. You help an old lady to cross the street, and you ruined her motivation to maintain her vision or to keep good relationships with her relatives. You invent the virus that kills mosquitoes worldwide, and you deprived the Africans of their motivation to study medicine. Any help is just a harm in disguise. (Or perhaps this only applies to helping Africans? Dunno.)
Sorry if you unnecessarily reviewed the whole video, I mostly wanted to point you at the 7-minute “Dependency” section.
You’re missing the point in the same way the comment I called out was. He is giving an example of 2 charities and giving overhead as a possible reason one would be more effective than the other. In fact, if bednets have a constant price, than a charity that spends 99% of its money on bednets is more effective than one that spends 1% (as long as the bednets get distributed equally well etc).
He said costs not overhead.
I just quoted him talking about it, in the previous comment. Have you checked yourself whether there was an anti-malaria-net industry?
This is not the relevant number, the relevant number is more like “counterfactually on no aid, how many bednets would the local industry be producing.”
Where in the text do you find this? Quoting 45:30 in the video:
“You can measure the amount of choice that someone has in a process. You can work on projects that decrease dependency and build capacity. The problem is that this measurement and these projects come with overhead costs. They don’t save the most lives for the dollar, but they do the work in sustainable ways. They inspire populations to try doing the work themselves. Even if that costs more now, it will be cheaper in the long run.”
The independence point is about long-term economic growth (causing more pleasure down the line), not being happy that you’re less dependent.
Again, not supported in the text. Why would this person work in international development if he thought this?
In general this comment fails to engage with the strongest part of the video, which is the 7-minute “Dependency” section I mentioned, in which he says the problems caused by aid are extremely bad in some of the countries that are targets of aid (like, they essentially destroy people’s motivation to solve their community’s problems).
This is specifically the part that my understanding of Econ 101 fails to process.
There is this one guy and his 10 employees, and they can’t make enough nets for the whole Africa. Okay, that part is simple to imagine. But why doesn’t he employ more people and increase the production? Or why someone else doesn’t copy his business model?
If my memory serves me well, Econ 101 assumes that if there is an effective demand, sooner or later there will also be the supply to match it.
Should I assume that there is no effective demand to buy more nets? Like, perhaps people are not aware of the dangers of malaria-carrying mosquitoes, or they don’t believe the nets are helpful, or they simply do not have enough money to match the production costs of the nets: either because the nets are too expensive, or because they have to prioritize other necessities. But then, the problem is not that they “can’t make enough nets”, but rather that they “can’t sell enough nets”.
Another thing about Econ 101 is the principle of comparative advantage. According to this principle, trying to produce everything at home is worse than trading internationally. (Otherwise an embargo would be a blessing instead of a punishment.) But an international trade will inevitably put some of your local producers out of business. Is it possible that Africa has a comparative disadvantage at producing anti-malaria nets?
It seems to me that the author is pattern-matching food aid to nets aid. But that is a different situation. In a situation without foreign aid, you cannot have a long-term imbalance between local food production and local food needs… the starving people will die. But you can have a long-term imbalance between local anti-malaria net production and local anti-malaria net needs… people unprotected against malaria only get it with some probability, not certainty; some who get malaria will die but some will survive. In other words, the resulting balance can’t include people who don’t eat, but it can include people who are not protected against malaria but still some of them survive until they can reproduce. Foreign food aid tries to solve a temporary imbalance; and in the process perhaps introduces more harm than good. But foreign anti-malaria-net aid tries to change the long-term balance.
I understand how an intervention that puts half of your population out of business can have this effect. I find it less likely that an intervention that puts one person in a million out of business would have the same effect. That is why I asked how many people are employed in the anti-malaria-net industry, compared with agriculture.
This seems to me like a mistaken pattern-matching. Pretty much anything can make someone lose their job. But there is a difference between “save thousand people, destroy thousand jobs” (food aid) and “save thousand people, destroy one job” (anti-malaria nets).
Re the section you quoted: if you watched for 1 minute longer you would see that the issue is that the local net manufacturers can’t scale up because they would have to compete with free nets, so the local infrastructure atrophies. (Absent aid, they could scale up, it would just take longer; scaling up might require additional capital or training)
The issue isn’t number of jobs. The issue is (a) infrastructure to solve your own problems and (b) motivation to solve your own problems rather than waiting to have someone else solve them for you. This is all covered in the “Dependency” section of the video, which I am still not convinced you have watched.
EAs say they don’t care about overhead per se and just care about outcomes, but public-facing recommenders like GiveWell and CEA keep recommending charities with low overhead relative to programmatic spending, such as AMF and GiveDirectly, rather than charities that do the sort of hard-to-account-for institution-building which low-overhead charities depend on.
EAs are good at explaining why you shouldn’t do what they’re (we’re?) doing. That’s different than actually doing the right thing.
I agree (with the second part, at least).
The author of the video said “according to Singer” repeatedly, so I assumed he also disagrees with what EAs are saying. If the real objection was “Singer says to do the right thing, but then actually does exactly the thing he said was wrong”, I didn’t get that message from the video. (Maybe the problem is on my side.)
Singer also recommends charities like the GW top charities, specifically mentions GiveWell approvingly, does cost-benefit analyses that ignore social & physical infrastructure, etc. These are also things Singer is saying.
This is minor, but GiveWell already says “Our best guess is that [nets] last, on average, between 2 and 2.5 years.” (https://www.givewell.org/charities/amf)