What do you think the unintended consequences of these efforts to stop malaria could be? Nassim Taleb argues that the Gates Foundation is repeating the errors of Mao Zedong. It’s also possible that donating malaria nets could cause local net manufacturers to go out of business, which could increase African dependence on foreign aid in the long run.
I’ve seen you link to the Mao video multiple times. Whenever you’re linking to a long resource in a way that isn’t self-explanatory, it really helps to share a summary of what you mean.
Mao Zedong made (to be charitable) many errors, so that summary is much less informative than “cause local net manufacturers to go out of business”.
But since you’ve already seen the video, you could probably write a brief summary in much less time than it will take, say, five interested readers to watch enough of the video to see what you mean. And you’ll be able to use that summary in other threads where you want to raise the same question, so it pays dividends.
The Gates Foundation is financing a campaign to genetically engineer the mosquito population in order to control malaria. He compares it to Mao Zedong’s Four Pests Campaign, and how Mao’s attempts to wipe out the sparrow population resulted in the Great Chinese Famine. Taleb argues that there may be similar unintended consequences, and something similar could happen with genetically modifying mosquitoes. He also talks about processes that are too fast for nature, and he draws a graph comparing the speed at which the ecosystem changes and the corresponding risk of harm, and how harm scales non-linearly in proportion to speed.
Thanks for sharing a summary! It doesn’t seem like it applies to AMF’s work, but it does describe other malaria control efforts. My impression is that the scientists who work on these things all day often pay more attention to risks and safety than other people realize, but I hope that the initial tests being run on this technology include appropriate follow-up to understand any unintended consequences.
On the question about AMF’s impact on local manufacturers, here’s Rob Mather, head of AMF, on exactly those concerns. The response (copied below) is ten years old, so the information may be out of date.
It sounds like a difficult trade-off, and I’d be happy to see data on manufacturing conditions or other economic conditions in areas where AMF has worked, or on longer-term malaria rates that might reflect the impact of nets becoming less available locally. But I’ll note that I haven’t really seen a “go out of business” argument that reflects these points:
Lower malaria rates obviously increase productivity in a vacuum. I’d expect that losing a child, or having to care for a sick child, also has a negative impact on productivity. If one local manufacturer goes out of business, but thousands of additional cases of malaria are prevented, what’s the net economic effect?
If a net manufacturer goes out of business, and AMF’s nets only last a few years, how often can that manufacturer (or another one) get back into business? Consider that:
Any local business must have been a startup at some point, grown from nothing.
Someone who used to run such a business would have useful contacts and experience for starting it again — presumably, that’s easier than starting up the first time!
If lots more people are now accustomed to sleeping under nets, local demand for nets may be higher post-AMF, another good sign for local manufacturers.
Rob’s answer seems to imply that many areas can’t actually support local manufacturers (I don’t know how common this is).
How likely is it that someone who gets a free AMF net would otherwise have purchased a net locally? In other words, is AMF actually taking much business from local manufacturers?
The story where AMF has a major negative impact seems to indicate some combination of:
Lots of people taking free nets who would otherwise have bought them locally, to the extent that the local manufacturer goes out of business, and
Demand for nets not shooting back up once the AMF nets wear out and there are no more free net options, or
There being no one who has the ability + desire to start a net business, even though economic conditions prior to AMF allowed at least one such business to succeed and a huge number of people now need new nets
...and finally, the health and productivity benefits of much higher net coverage for a few years not outweighing the losses that come from slightly lower net coverage for some amount of time after that.
I say “slightly lower” because, based on the coverage data above, it seems like few people in the areas AMF covers actually had nets in the first place — if an area jumps from 20% coverage to 80% coverage, then drops to 0% coverage, the 0% condition has to persist for quite some time to negate the 80% condition.
I can imagine this story happening, perhaps because people see AMF crush a local business and decide not to start a new one themselves. But I’d be somewhat surprised if this set of conditions were common, mostly because net coverage was so low originally.
Of course, you could argue that given this dynamic, it would be more effective to find a way to support and scale local manufacturers by donating or investing, or that AMF would do more total good by purchasing from local producers even when it isn’t (as Rob says) “near-economic”. That seems more likely to me than AMF actually being net-negative.
If you’re interested in evaluating this, next steps might include:
Trying to convince GiveWell that it’s worth investigating (they often respond to emails, and they have regular open threads on their blog).
Doing some research and presenting it on the Forum (it might be feasible to find data on long-term coverage and malaria rates in places AMF or other net distributors have worked, and there are other ways to build evidence for the pro-”out of business” case).
Asking Rob what he now thinks about this issue, since the below response is ten years old. Or you could talk to other charities that work in net distribution and may have done their own internal research or built their own models.
Anyway, Rob’s response:
Hi Rob,
Thanks very much for your comprehensive answer. My main concern is (as I have read) that floods of foreign-sourced charitable nets have actually had the unfortunate negative effect of putting African mosquito net manufacturers out of business. As you probably know, Africa’s health, environmental, social, political and economic woes are largely tied together. This concern is mentioned in Dambisa Moyo’s book entitled “Dead Aid”. I’m not sure if you’re familiar with this argument but I felt compelled to contact you regarding this issue, since it can potentially make a difference for the lives of Africans. I would suggest that as whole it might be of greater benefit to Africans to make sure they manufacture the nets they use—that way the charity $$$ would pay dividends of creating jobs which would help lift them out of the poverty which makes them so vulnerable. What do you think?
-Alan
Dear Alan-
Yes, I think this is an important issue.
It must be a better situation if long-lasting insecticide treated nets (LLINs) are manufactured in the countries in which they are needed. That would bring two advantages. First, reduced transport costs. Second, local employment. There is a manufacturing facility in Tanzania, a Sumitomo-AtoZ textiles joint venture, so ‘local’ production, and the employment this brings, is possible.
We are alive to this issue and, where we can, act in a way to support local enterprise. We have bought tens of thousands of nets that way when it has been ‘near-economic’ to do so.
However, there are challenges to the speed of local capacity development and the number of facilities that could be developed.
First, economies of scale mean only a small number of large factories are required to produce world demand. Micro-factories, located in each net-consuming country would not be economic. The number of countries that could benefit from locally located facilities therefore would be small. Some countries benefiting would be better than none of course.
Second, domestic markets are often not enough to sustain a production facility: nets are required to be exported. This leads to a problem, or inefficiency, in that shipping and transport from some African countries to others can be more difficult and more expensive than shipping from Asia to many African countries given established shipping routes. This raises overall prices for net buyers and reduces the number of nets that can bought for a given level of funds.
Further, technology transfer is an issue with challenges around training a workforce and guarding against technology intellectual property loss, the latter being a reasonable concern of the primary manufacturer. Ensuring raw material, spare parts supply and quality control are also issues to overcome.
The first and second issues are the structural ones and present the greatest challenge. The other issues can be overcome as the Tanzania joint venture has indicated. With major capital investment required and the need to consider the long term viability of a new facility, these developments take year/s not months.
Another method of developing local capacity has been via shipping large rolls of netting from an African based manufacturing facility to another African country where the cutting and stitching of nets then takes place. This has some shipping cost savings and provides local employment. We have bought nets in this way also.
Note, we are only talking about LLINs here as that is the only sensible net to distribute. If part of the background to your comments is concern over local insecticide treated nets (ITNs, but not long-lasting, so an entirely different net), or untreated net production being threatened by the import of LLINs, the higher issue is going to be the need to protect people with LLINs rather than ITNs or untreated nets. This is because LLINs are much more effective than these other nets at protecting people from malaria. For information on different types of nets, see: http://www.againstmalaria.com/FAQ_Bednets.aspx
Our approach, therefore, is with our priority being to buy the most nets possible for the funds available. We keep a close eye on local-sourcing options and where it is ‘near-economic’ to do so, we do. Economics will drive manufacturers to locally locate and we can do our bit by applying this ‘near-economic’ approach.
What do you think the unintended consequences of these efforts to stop malaria could be? Nassim Taleb argues that the Gates Foundation is repeating the errors of Mao Zedong. It’s also possible that donating malaria nets could cause local net manufacturers to go out of business, which could increase African dependence on foreign aid in the long run.
I’ve seen you link to the Mao video multiple times. Whenever you’re linking to a long resource in a way that isn’t self-explanatory, it really helps to share a summary of what you mean.
Mao Zedong made (to be charitable) many errors, so that summary is much less informative than “cause local net manufacturers to go out of business”.
But since you’ve already seen the video, you could probably write a brief summary in much less time than it will take, say, five interested readers to watch enough of the video to see what you mean. And you’ll be able to use that summary in other threads where you want to raise the same question, so it pays dividends.
The Gates Foundation is financing a campaign to genetically engineer the mosquito population in order to control malaria. He compares it to Mao Zedong’s Four Pests Campaign, and how Mao’s attempts to wipe out the sparrow population resulted in the Great Chinese Famine. Taleb argues that there may be similar unintended consequences, and something similar could happen with genetically modifying mosquitoes. He also talks about processes that are too fast for nature, and he draws a graph comparing the speed at which the ecosystem changes and the corresponding risk of harm, and how harm scales non-linearly in proportion to speed.
Thanks for sharing a summary! It doesn’t seem like it applies to AMF’s work, but it does describe other malaria control efforts. My impression is that the scientists who work on these things all day often pay more attention to risks and safety than other people realize, but I hope that the initial tests being run on this technology include appropriate follow-up to understand any unintended consequences.
On the question about AMF’s impact on local manufacturers, here’s Rob Mather, head of AMF, on exactly those concerns. The response (copied below) is ten years old, so the information may be out of date.
It sounds like a difficult trade-off, and I’d be happy to see data on manufacturing conditions or other economic conditions in areas where AMF has worked, or on longer-term malaria rates that might reflect the impact of nets becoming less available locally. But I’ll note that I haven’t really seen a “go out of business” argument that reflects these points:
Lower malaria rates obviously increase productivity in a vacuum. I’d expect that losing a child, or having to care for a sick child, also has a negative impact on productivity. If one local manufacturer goes out of business, but thousands of additional cases of malaria are prevented, what’s the net economic effect?
If a net manufacturer goes out of business, and AMF’s nets only last a few years, how often can that manufacturer (or another one) get back into business? Consider that:
Any local business must have been a startup at some point, grown from nothing.
Someone who used to run such a business would have useful contacts and experience for starting it again — presumably, that’s easier than starting up the first time!
If lots more people are now accustomed to sleeping under nets, local demand for nets may be higher post-AMF, another good sign for local manufacturers.
Given that AMF targets areas with very high numbers of people not sleeping under nets, how often are they actually competing with local manufacturers?
Rob’s answer seems to imply that many areas can’t actually support local manufacturers (I don’t know how common this is).
How likely is it that someone who gets a free AMF net would otherwise have purchased a net locally? In other words, is AMF actually taking much business from local manufacturers?
In AMF’s post-distribution monitoring, they find that nearly every net in households they cover is an AMF net, implying relatively low rates of net usage before that time.
The story where AMF has a major negative impact seems to indicate some combination of:
Lots of people taking free nets who would otherwise have bought them locally, to the extent that the local manufacturer goes out of business, and
Demand for nets not shooting back up once the AMF nets wear out and there are no more free net options, or
There being no one who has the ability + desire to start a net business, even though economic conditions prior to AMF allowed at least one such business to succeed and a huge number of people now need new nets
...and finally, the health and productivity benefits of much higher net coverage for a few years not outweighing the losses that come from slightly lower net coverage for some amount of time after that.
I say “slightly lower” because, based on the coverage data above, it seems like few people in the areas AMF covers actually had nets in the first place — if an area jumps from 20% coverage to 80% coverage, then drops to 0% coverage, the 0% condition has to persist for quite some time to negate the 80% condition.
I can imagine this story happening, perhaps because people see AMF crush a local business and decide not to start a new one themselves. But I’d be somewhat surprised if this set of conditions were common, mostly because net coverage was so low originally.
Of course, you could argue that given this dynamic, it would be more effective to find a way to support and scale local manufacturers by donating or investing, or that AMF would do more total good by purchasing from local producers even when it isn’t (as Rob says) “near-economic”. That seems more likely to me than AMF actually being net-negative.
If you’re interested in evaluating this, next steps might include:
Trying to convince GiveWell that it’s worth investigating (they often respond to emails, and they have regular open threads on their blog).
Doing some research and presenting it on the Forum (it might be feasible to find data on long-term coverage and malaria rates in places AMF or other net distributors have worked, and there are other ways to build evidence for the pro-”out of business” case).
Asking Rob what he now thinks about this issue, since the below response is ten years old. Or you could talk to other charities that work in net distribution and may have done their own internal research or built their own models.
Anyway, Rob’s response: