Insecticide-treated nets significantly harm mosquitoes, but one can easily offset this?

By Vasco Grilo🔸 @ 2025-02-03T18:03 (+31)

The views expressed here are my own, not those of the people who provided feedback on the draft.

Summary

Context

I strongly endorse expectational total hedonistic utilitarianism (increasing happiness, and decreasing suffering), and think one can reasonably make comparisons across species based on Rethink Priorities’ (RP’s) median welfare ranges. This post explores potential implications of having these 2 views. I think the takeaways are basically the same under desire theories, as beings want to be happy, and not suffer. However, they may differ significantly if you strongly reject impartiality, or consider RP’s median welfare ranges dramatically overestimate animals’ capacity for welfare.

Harm caused to mosquitoes

Here are my calculations. I describe them below.

According to Open Philanthropy (OP), “GiveWell uses moral weights for child deaths that would be consistent with assuming 51 years of foregone life in the DALY framework [this one] (though that is not how they reach the conclusion)”. I guess 1 mosquito-year of fully healthy life is 1.3 % as good as 1 human-year of fully healthy life, which is RP’s median welfare range of black soldier flies[1]. So I think 3.92 k mosquito-years of fully healthy life are as good as the additional human welfare from saving 1 life under GW’s moral weights.

I guess disabling pain is 10 times as intense as fully healthy life, in which case 2.4 hours (= 24/10) of disabling pain neutralise 1 day of fully healthy life. Consequently, I infer that 392 mosquito-years of disabling pain neutralise the additional human welfare from saving 1 life under GW’s moral weights.

GW’s last grant to AMF, as available on their database on 15 January 2025, was of 41 M$, and targeted DRC, for which GW calculates a cost per distributed net and life saved of 6.78 $ and 5.10 k$. These imply AMF has to distribute 752 nets to save a life in DRC. As a result, 0.522 mosquito-years of disabling pain neutralise the benefits to humans of distributing a net in DRC.

GW calculates that nets in DRC effectively last 1.2 years. As a consequence, 0.435 mosquito-years of disabling pain neutralise the benefits to humans of 1 net-year in DRC.

I guess nets kill 1 mosquito every hour[2], or 0.0167 mosquitos per net-minute. Accordingly, 26.1 mosquito-minutes of disabling pain per mosquito killed by nets neutralise the benefits to humans of GW’s last grant to AMF.

I estimate mosquitoes’ welfare loss per mosquito killed by nets is equivalent to 19.9 k mosquito-minutes of disabling pain. I got this based on:

I conclude GW’s last grant to AMF of 41 M$ caused 763 times as much harm to mosquitoes via ITNs as it benefited humans. Here are a few ways of the harm caused to mosquitoes via ITNs to be as large as the benefits to humans:

Discussion

It would be great if there were ITNs which painlessly kill mosquitoes, but it looks like there are not any. According to Claude:

I neglected the effects of ITNs on the number of wild animals because it is super unclear whether they have positive or negative lives. Yet, there is still lots of uncertainty even just in the effects I considered. RP’s 5th and 95th percentile welfare ranges of black soldier flies are 0 and 15.1 (= 0.196/0.013) times their median. This suggests that, even ignoring effects on the number of wild animals, and just accounting for uncertainty in mosquitoes’ capacity for welfare, the 5th and 95th percentile harm to mosquitoes caused by ITNs are 0 and 11.5 k (= 15.1*763) times their benefits to humans. So it is unclear to me whether ITNs increase or decrease welfare.

I believe the large uncertainty about the effects of human welfare interventions on wild (and farmed) animals should push one towards prioritising:

No one from the animal welfare organisations I mentioned above reviewed the draft of this post. As always unless stated otherwise, I am speaking for myself.

I strongly endorse maximising expected welfare. Nevertheless, I think donating to the above organisations is even better if one intrinsically cares about minimising the probability of causing harm.

I estimate people donating to AMF can offset the harm ITNs cause to mosquitoes pretty cheaply, donating 1.19 % (= 763/(64.3*10^3)) as much to SWP as to AMF. This fraction is more robust than it may seem because the harm ITNs cause to mosquitoes is decently proportional to the cost-effectiveness of SWP[5]. Under my views:

I would move any marginal donations from helping humans to helping invertebrates, so I would not donate to AMF. Nonetheless, I think directing a small fraction (10 %?) of one’s donations to helping invertebrates would be a good compromise to offset potential negative effects. I encourage people donating to animal welfare to do this too. Decreasing the consumption of animal-based foods, especially beef, has major effects on wild animals, and improving animals’ conditions can indirectly change their consumption too, although arguably much less. On the impact of human diet on animal welfare, Michael St. Jules suggested Matheny (2005), this and these posts from Brian Tomasik, this post from Carl Shulman, and Fischer (2018). There is also the sequence Human impacts on animals created by Michael.

Acknowledgements

Thanks to CB for the comment which motivated me to make this post. Thanks to Abraham Rowe, CB, and Michael St. Jules for feedback on the draft[7].

  1. ^

     Mosquitoes belong to the order Diptera, and black soldier flies are the only species analysed by RP of that order. I asked Bob Fischer, who led RP’s moral weight project, about a best guess for the median welfare range of mosquitoes that RP would have obtained if they had analysed them. I privately disclaimed I would publish this post without Bob’s guess, but Bob did not share one.

  2. ^

     I did not easily find estimates.

  3. ^

     The quotes describing the 4 stages are from the chat where I got the 1st set of estimates.

  4. ^

     Aggregating with the continuous version of the geometric mean of odds lognormal distributions whose logarithms have the same standard deviation results in a distribution whose mean is equal to the geometric mean of the means of the lognormal distributions.

  5. ^

     If they were proportional, the fraction would be constant regardless of their uncertainty.

  6. ^

     The past cost-effectiveness of SWP becomes 10.0 % (= 64.1/639) as large if excruciating pain becomes 10 % as intense. Likewise, the harm to mosquitoes becomes 10.0 % (= 76.5/763) as large if excruciating pain becomes 10 % as intense.

  7. ^

     I ordered the names alphabetically.

David T @ 2025-02-03T20:52 (+20)

If I'm understanding your calculations correctly, the underlying assumption is that the pain you estimate a mosquito to experience for two minutes has the same weight as an entire afternoon of incomparably blissful human existence even taking into account the cognitive differences between a human and a mosquito? There doesn't appear to be an obviously correct way to weight the relative intensity of experience of a human and a mosquito, but this one seems like an outlier; typically arguments for considering insect suffering depend on them being more numerous rather than their individual suffering being orders of magnitude more intense than human enjoyment. In all seriousness, if you do attach such high weights to the possible suffering of individual insects, I highly recommend nontoxic spider repellent, especially around your light fittings as an extremely cost effective intervention.

Some of your more quantifiable estimates also seem selected to be particularly unfavourable to humans. For example, the robustly established fact that humans experience days of pain from malaria infections, (including the vast majority of malaria infections which are nonfatal) is disregarded. Medical literature evaluating anti-malaria interventions often focuses on mortality rather than morbidity too, but it's not weighing up human DALYs against a few minutes of mosquito morbidity! Likewise, the assumption that a typical ITN is killing an average of 24 mosquitos per day seems to depend on an inflated number mosquitos per dwelling, even before the mild repellent effect and low killing efficiency of fleeting contact with the nets is considered.

Vasco Grilo🔸 @ 2025-02-03T22:36 (+4)

Thanks, David.

If I'm understanding your calculations correctly, the underlying assumption is that the pain you estimate a mosquito to experience for two minutes has the same weight as an entire afternoon of incomparably blissful human existence even taking into account the cognitive differences between a human and a mosquito?

Which 2 min of pain in mosquitoes are you referring to? "I infer that 392 mosquito-years of disabling pain neutralise the additional human welfare from saving 1 life under GW’s moral weights".

In all seriousness, if you do attach such high weights to the possible suffering of individual insects, I highly recommend nontoxic spider repellent, especially around your light fittings as an extremely cost effective intervention.

I do not kill mosquitoes or other insects inside my house, but I guess quickly crushing insects causes them much less pain than ITNs. "I estimate mosquitoes’ welfare loss per mosquito killed by nets is equivalent to 19.9 k mosquito-minutes of disabling pain".

I kill insects by driving, but I think this is fine. Firstly, the deaths are hopefully pretty quick. Secondly, driving saves me time, and I believe most of my impact comes from my work and donations, not from my direct impact on animals via my diet or driving. I calculate neutralising the harms caused to poultry birds and farmed aquatic animals per person in 2022 only requires donating 0.0214 $ to SWP. I guess my potentially negative direct impact on wild animals can also be neutralised with very little donations.

Some of your more quantifiable estimates also seem selected to be particularly unfavourable to humans. For example, the robustly established fact that humans experience days of pain from malaria infections, (including the vast majority of malaria infections which are nonfatal) is disregarded. Medical literature evaluating anti-malaria interventions often focuses on mortality rather than morbidity too, but it's not weighing up human DALYs against a few minutes of mosquito morbidity!

I used GW's estimate of 5.10 k$ per life saved by AMF in DRC, which is supposed to account for effects on mortality, morbidity, and income. GW's estimate may well not be perfect, but I think the uncertainty is the cost to save a life is negligible in comparison with that of other parameters like the welfare range of mosquitoes.

Likewise, the assumption that a typical ITN is killing an average of 24 mosquitos per day seems to depend on an inflated number mosquitos per dwelling, even before the mild repellent effect and low killing efficiency of fleeting contact with the nets is considered.

I wish I had a better estimate, but I did not easily find one. My assumption was somewhat informed by a trip I did to Moshi (Tanzania) in early 2020. There were certainly more than 1 mosquito bitting me per hour during dust, and I was using repellent if I recall correctly. People most affected by malaria may sleep indoors, but have houses with very limited ability to prevent mosquitoes from entering. The roofs can be made of turf too, which would attract mosquitoes.

David T @ 2025-02-04T00:01 (+10)

Which 2 min of pain in mosquitoes are you referring to?

The 2 minutes corresponds to the estimated 119 seconds of estimated excruciating pain per mosquito death in the aggregate estimate in your spreadsheet, comprising nearly all the estimated utility loss.

I do not kill mosquitoes or other insects inside my house, but I guess quickly crushing insects causes them much less pain than ITNs.

It was less about your personal footprint and more about the spiders.  I once lived in a place by a river an enormous quantity of insects were attracted by any sort of light bulb, which was where the spiders liked to dine out (unless they were deterred with peppermint spray or their cobwebs repeatedly swept away). A web full of wriggling flies wasn't a particularly attractive sight, but I'm disinclined to believe that web was experiencing utility loss far more significant than anything going on in my life[1] But since you are arguing a few minutes of a single insect ingesting a neurotoxin may be of extremely high negative value, keeping spiders away from insects using cheap peppermint spray seems like an highly net positive form of harm reduction worth considering?

My assumption was somewhat informed by a trip I did to Moshi (Tanzania) in early 2020. There were certainly more than 1 mosquito bitting me per hour during dust, and I was using repellent if I recall correctly

Outdoors at dusk is peak mosquito time though, and 2-3 mosquitos are capable of a lot of bites. I would imagine you had access to some sort of treated nets, and didn't have have to clean 20 or 30 dead mosquitos off the floor every day? 

  1. ^

    I'd have a particularly hard time believing insects had evolved a complex and intense appreciation of neurological pain whilst far more useful traits like navigation were as simplistic and mechanistic as repeatedly flying into a light source... 

Vasco Grilo🔸 @ 2025-02-04T00:34 (+2)

The 2 minutes corresponds to the estimated 119 seconds of estimated excruciating pain per mosquito death in the aggregate estimate in your spreadsheet, comprising nearly all the estimated utility loss.

Thanks for clarifying. For my guess that excruciating pain is 100 k (= 10*10^3/0.1) times as intense as fully healthy life, the 119 mosquito-seconds of excruciating pain per mosquito killed by ITNs neutralise 138 mosquito-days (= 119/60^2/24*100*10^3) of fully healthy life, or 1.79 human-days (= 138*0.013) of fully healthy life based on RP's median welfare range of black soldier flies.

It was less about your personal footprint and more about the spiders.

Got it. I estimate paying farmers to use more humane pesticides to decrease the suffering of wild insects helps 5.74 M insects per $. So, valuing my time at 20 $/h, I would have to save 115 M insects per hour (= 5.74*10^6*20) to be similarly cost-effective. One can support research on more humane pesticides by donating to WAI. They intend “to use current and new funding” for, among other activities, “Conducting an analysis of agricultural pest control to better understand the best targets for welfare interventions — first identifying scientific gaps and then developing research plans to help fill them”.

Outdoors at dusk is peak mosquito time though, and 2-3 mosquitos are capable of a lot of bites.

There were certainly more than 1 different mosquito per hour too. I was indeed outdoors at dusk, which is peak mosquito time, so I used a value corresponding to much less mosquitoes than what I seem to recall. In any case, my takeaways would be the same if the number of mosquitoes killed per net was e.g. 10 % as high. This would correspond to only 2.4 mosquitoes (= 0.1*24) killed per net-day, but still result in AMF causing 76.3 (= 0.1*763) times as much harm to mosquitoes via ITNs as it benefits humans. With large uncertainty, such that it would still be "unclear to me whether ITNs increase or decrease welfare".

David T @ 2025-02-04T17:41 (+8)

For my guess that excruciating pain is 100 k (= 10*10^3/0.1) times as intense as fully healthy life, the 119 mosquito-seconds of excruciating pain per mosquito killed by ITNs neutralise 138 mosquito-days (= 119/60^2/24*100*10^3) of fully healthy life, or 1.79 human-days (= 138*0.013) of fully healthy life based on RP's median welfare range of black soldier flies.

Thanks for that clarification. So essentially your claims rest on the utility value of over a day and a half of human life being lower than that of two minutes of a dying insect.

Two comments here: 

  1. this, like some of your other estimates relies rather heavily on an unconventional and extremely skewed pain scale, whereby a certain degree of pain is worth many times more than maximal pleasure[1], as well as confidently attributing that maximal degree of pain that vastly exceeds the pleasure experienced by more complicated creatures to a particular scenario
  2. I 'm not sure this is actually how RP intend their welfare ranges to be used. My understanding (and I welcome clarification/correction from RP on this point) is that when their researchers estimate that $creature's welfare range is 1.3% that of humans, they intend that to be interpreted as "$creature's pain sensations are at most 1.3% as intense as human experience", not "to establish how intensely $creature feels pain, multiply 1.3% by a pain scale which may contain an arbitrarily large number of digits, to reach the conclusion that this creature's pain is potentially thousands of times as intense as human pleasure."

I'd also point out that even with those pain scales and welfare ranges, the calculation looks completely different if one also factors in potentially intense human pain from [nonfatal] malaria infections multiple times per year and experienced over several days, with [rare] neurological systems which may persist for the rest of a natural human life. Again, I'm not sure exactly what a pain scale for celebral malaria should look like but I'm unconvinced there are reasons for regarding it as so much less intense than mosquito pain it can be disregarded when comparing between species.

  1. ^

    I recognise that extremely wide-ranging and asymmetric pain scales are convenient to pure hedonic utilitarians who might otherwise be troubled by philosophical problems like utility monsters or trading off a single torture for a speck of dust in my eye: I just think they're unusual positions not well supported by evidence. 

Vasco Grilo🔸 @ 2025-02-04T18:38 (+2)

this, like some of your other estimates relies rather heavily on an unconventional and extremely skewed pain scale, whereby a certain degree of pain is worth many times more than maximal pleasure

My guess that excruciating pain is 100 k (= 10*10^3/0.1) times as intense as fully healthy life ( 100 k times as intense as the maximally happy instant) implies that 0.864 s (= 24*60^2/(100*10^3)) of excruciating pain in humans neutralises 1 day of fully healthy life in humans. Do you think this is "unconventional and extremely skewed"? Examples of excruciating pain include "scalding and severe burning events [in large parts of the body]", or "dismemberment, or extreme torture". I have not experienced excruciating pain myself, so I would not be surprised if it was 10 % as intense as I assumed, but this would not change my takeaways. I estimate the harms caused to mosquitoes are proportional to the intensity of excruciating pain, so AMF would cause 76.3 (= 0.1*763) times as much harm to mosquitoes as it benefits humans according to the updated mainline numbers.

I am not assuming excruciating pain is 100 k times as intense as the maximally happy instant (as you implied above). This is much more intense than fully healthy life, so I guess excruciating pain is much less than 100 k times as intense as the maximally happy instant.

I 'm not sure this is actually how RP intend their welfare ranges to be used. My understanding (and I welcome clarification/correction from RP on this point) is that when their researchers estimate that $creature's welfare range is 1.3% that of humans, they intend that to be interpreted as "$creature's pain sensations are at most 1.3% as intense as human experience", not "to establish how intensely $creature feels pain, multiply 1.3% by a pain scale which may contain an arbitrarily large number of digits, to reach the conclusion that this creature's pain is potentially thousands of times as intense as human pleasure."

I assumed the welfare per living time of fully healthy life in a given species is proportional to the welfare range of that species. This does not directly follow from RP's welfare ranges, but does not obviously favour helping animals or humans. My understanding it that RP's median welfare ranges refer to the ratio ("welfare per time of the practically maximally happy animal-year (positive)" - "welfare per time of the practically maximally sad animal-year (negative)")/("welfare per time of the practically maximally happy human-year" - "welfare per time of the practically maximally sad human-year").

David T @ 2025-02-04T20:40 (+10)

0.864 s (= 24*60^2/(100*10^3)) of excruciating pain in humans neutralises 1 day of fully healthy life in humans. Do you think this is "unconventional and extremely skewed

Yes. I can't think of any pain in which I would prefer to die than suffer for 0.864 seconds per day, particularly not if the remaining aspects of my life were "practically maximally happy".[1]

I find it even harder to imagine that an insect can distinguish between painful sensations to the degree that a pain scale with at least 100k points on it would be appropriate to approximate their welfare on,[2] still less that an appropriate use of such a scale is to multiply a human/insect welfare ratio to conclude that the complete cessation of function of that simple insect nervous system is a few orders of magnitude more intense conscious experience than the "practically maximally happy" (or even average) utility of a human.

  1. ^

    If I did think [potential] sub second pain was as significant as an entire day's welfare, I would probably not endorse electrical stunning...

  2. ^

    I mean, they've only got 200k neurons to divide between all their functions. (This isn't an argument for neuron count being a good proxy for moral weights overall, merely an observation of how extreme the pain scale looks in the context of how simple the insect's system for parsing stimuli appears to be)

bruce @ 2025-02-03T21:06 (+17)

TL;DR
I think you are probably at least a few OOMs off with these figures, even granting most of your assumptions, as this implies (iiuc) ~8 million mosquito deaths per human death averted.

At 763x GiveWell, a tradeoff of 14.3 days of mosquito excruciating pain (MEP) for 1 happy human life, 2 minutes of MEP per mosquito, a $41 million dollar grant in DRC, and $5100 per life saved, this implies 7.9 million mosquitos killed per human life saved, and that the grant will kill ~63 billion mosquitos.[1]


EDIT: my initial estimates were initially based on 11 mosquito minutes of excruciating pain neutralising 1 day of human life as stated in the text. This was incorrect because I misinterpreted the text. The true value that this post endorses is approximately a factor of 10 in the direction of the mosquito side of the tradeoff (i.e. the equivalent of ~68 mosquito seconds of excruciating pain neutralising 1 day of human life, and ~8million mosquito deaths per human death averted by bednets). I have edited the topline claim accordingly.

============
[text below is no longer accurate / worth reading; see above]
A quick sense check using your assumptions and numbers (I typed this up quickly so might have screwed up the maths somewhere!)

When you say:
"1 day of [a practically maximally happy life] would be neutralised with 23.9s of excruciating pain.
and
"As a result, 11.0 min of excruciating pain would neutralise 1 day of a practically maximally happy life"

I'm assuming you mean "23.9 mosquito seconds of excruciating pain" and "11.0 mosquito minutes of excruciating pain" trading off against 1 human day of a practically maximlly happy life (please correct me if I'm misunderstanding you!)

At 763 times as much harm to mosquitos to humans, ~50 DALYs per life saved, and 11min (or 23.9 seconds) of MEP, this implies you are suggesting bednets are causing something like 333 million ~ 9 billion seconds of MEP per human death averted.[2]

Using your figures of 2 minutes of excruciating pain per mosquito killed this gives a range of 3million~77 million mosquito deaths per human death averted in order for your 763x claim to be correct.[3]

Using your stated figures of $41 million and $5100 per life for the GW grant, this implies you think the grant will lead to somewhere between 22~616 billion mosquito deaths in DRC alone.[4]

For context, this source estimates global mosquito population as between 110 trillion and 'in the quadrillions'.

  1. ^

    763*(14.3*24*60)/2 = 7,855,848
    41 million / 5100 * 763*(14.3*24*60)/2 = 63,154,856,470.6

  2. ^

    365.25*50*11*60*763 = 9,196,629,750
    365.25*50*23.9*763 = 333,029,471.25

  3. ^

    333029471 / 120 = 2,775,245.59
    9196629750 / 120 = 76,638,581.25

  4. ^

    41 million / 5100 * 2,775,245.59 = 22,310,797,880.4
    41 million / 5100 * 76,638,581.25 = 616,114,084,559

Vasco Grilo🔸 @ 2025-02-03T22:01 (+5)

Thanks for the comment, bruce.

I think you are probably at least a few OOMs off with these figures, even granting most of your assumptions, as this implies (iiuc) ~8 million mosquito deaths per human death averted.

I very much disagree. I explain why below.

I'm assuming you mean "23.9 mosquito seconds of excruciating pain" and "11.0 mosquito minutes of excruciating pain" trading off against 1 human day of a practically maximlly happy life (please correct me if I'm misunderstanding you!)

You are misunderstanding me. I mean that, if excruciating pain was 3.62 % (= 0.00131^(1/2)) as intense as I assumed, I would guess 23.9 s of excruciating pain in a given species (humans, of mosquitoes, or other) would neutralise 1 day of a practically maximally happy life in that same species.

Using your stated figures of $41 million and $5100 per life for the GW grant, this implies you think the grant will lead to somewhere between 1.3 trillion and 37 trillion mosquito deaths in DRC alone.[3]

For context, this source estimates global mosquito population as between 110 trillion and 'in the quadrillions'.

I estimate GW's last grant to AMF will kill 0.0183 % as many mosquitoes as the ones currently alive globally over the lifetime of the bednets it funds. This would correspond to killing 1.19 % (= 1.83*10^-4*195/3) of the mosquitoes in DRC per year assuming mosquitoes were uniformly distributed across the existing 195 countries, and that the nets funded by the grant are distributed over 3 years. In reality, it would be less because DRC should have more mosquitoes than a random country. AMF being responsible for killing less than 1.19 % of the mosquitoes in DRC does not sound implausible.

bruce @ 2025-02-03T22:14 (+6)

Gotcha RE: 23.9secs / 11mins, thanks for the clarification!

Looking at this figure you are trading off 7910000 * 2 minutes of MEP for a human death averted, which is 15820000 minutes, which is ~30 mosquito years[1] of excruciating pain trading off for 50 human years of a practically maximally happy life.

Is this a correct representation of your views?

(Btw just flagging that I think I edited my comment as you were responding to it RE: 1.3~37 trillion figures, I realised I divided by 2 instead of by 120 (minutes instead of seconds).)

  1. ^

    7910000 * 2 / 60 / 24 / 365.25 = 30.08

Vasco Grilo🔸 @ 2025-02-03T23:42 (+3)

Looking at this figure you are trading off 7910000 * 2 minutes of MEP for a human death averted, which is 15820000 minutes, which is ~30 mosquito years[1] of excruciating pain trading off for 50 human years of a practically maximally happy life.

Is this a correct representation of your views?

"I think 3.92 k mosquito-years of fully healthy life are as good as the additional human welfare from saving 1 life under GW’s moral weights", and I guess excruciating pain is 100 k times as intense as fully healthy life, so I estimate 14.3 mosquito-days (= 3.92*10^3/(100*10^3)*365.25) of excruciating pain neutralise the benefits of the additional human welfare from saving 1 life under GW’s moral weights. I understand that may seem very little time, but I do not think it can be dismissed just on the basis of seeming surprising. I would say one should focus on checking whether the results mechanistically follow from the inputs, and criticing these:

  • "I guess 1 mosquito-year of fully healthy life is 1.3 % as good as 1 human-year of fully healthy life, which is RP’s median welfare range of black soldier flies".
  • My guess that excruciating pain is 100 k (= 10*10^3/0.1) times as intense as fully healthy life implies that 0.864 s (= 24*60^2/(100*10^3)) of excruciating pain in humans neutralises 1 day of fully healthy life in humans. Examples of excruciating pain include "scalding and severe burning events [in large parts of the body]", or "dismemberment, or extreme torture".
bruce @ 2025-02-04T01:54 (+6)

I estimate 14.3 mosquito-days of excruciating pain neutralise the benefits of the additional human welfare from saving 1 life under GW’s moral weights.

Makes sense - just to clarify:

My previous (mis)interpretation of you suggesting 11minutes of MEP trading off 1 day of fully healthy human life would indicate a tradeoff of 11 / (24*60) = 0.0076.

Your clarification is that 14.3 mosquito-days trades off against 1 life:
assuming 1 life as 50 DALYs this is 14.3 / (50*365.25) = 0.00078

So it seems like my misinterpretation was ~10x overvaluing the human side compared to your true view?

I understand that may seem very little time, but I do not think it can be dismissed just on the basis of seeming surprising. I would say one should focus on checking whether the results mechanistically follow from the inputs, and criticing these:

My view is probably something like:
"I think on the margin most people should be more willing to entertain radical seeming ideas rather than intuitions given unknown unknowns about moral catastrophes we might be contributing to, but I also think the implicit claim[1] I'm happy to back here is that if your BOTEC spits out a result of "14.3 mosquito days of excruciating pain trades off with 50 human years of fully healthy life" then I do expect on priors that some combination of your inputs / assumptions / interpretation of the evidence etc have lead to a result that is likely many factors (if not OOMs) off the true value (if we magically found out what it was (and I think such a surprising result should also prompt similar kinds of thoughts on your end!)). I'll admit I don't have a strong sense of how to draw a hard line here, but I can imagine for this specific case that I might expect the tradeoff for humans is closer to 3.5 hours of excruciating pain vs a life, and that I value / expect the human capacity for welfare to be >100x that of a mosquito. If you believe both of those to be true then you'd reject your conclusion.

Another thing to consider might be something like "the way you count/value excruciating pain in humans vs in animals is inconsistent in a way that systematically gives results in favour of animals"

I don't have too much to offer here in terms of this - I just wanted to know what the implied trade-off actually was and have it spelled out.

  1. ^

    Referring only to this specific example, not necessarily other posts of yours I've commented on

Vasco Grilo🔸 @ 2025-02-04T09:52 (+6)

So it seems like my misinterpretation was ~10x overvaluing the human side compared to your true view?

Yes. I should disclaim I have not reviewed your other calculations in detail, but I am confident I calculated correctly the numbers I shared in this thread.

if your BOTEC spits out a result of "14.3 mosquito days of excruciating pain trades off with 50 human years of fully healthy life" then I do expect on priors that some combination of your inputs / assumptions / interpretation of the evidence etc have lead to a result that is likely many factors (if not OOMs) off the true value (if we magically found out what it was (and I think such a surprising result should also prompt similar kinds of thoughts on your end!)).

This makes it sound like that result I got came from many calculations, but it only depends on very few inputs. Besides 51 DALYs averted per life saved under GW's moral weights:

  • "I guess 1 mosquito-year of fully healthy life is 1.3 % as good as 1 human-year of fully healthy life, which is RP’s median welfare range of black soldier flies".
  • My guess that excruciating pain is 100 k (= 10*10^3/0.1) times as intense as fully healthy life implies that 0.864 s (= 24*60^2/(100*10^3)) of excruciating pain in humans neutralises 1 day of fully healthy life in humans. Examples of excruciating pain include "scalding and severe burning events [in large parts of the body]", or "dismemberment, or extreme torture".

Another thing to consider might be something like "the way you count/value excruciating pain in humans vs in animals is inconsistent in a way that systematically gives results in favour of animals"

Why would it be inconsistent? I assume that the ratio between the intensities of 2 categories of pain does not vary across species, as do AIM and RP.

I might expect the tradeoff for humans is closer to 3.5 hours of excruciating pain vs a life, and that I value / expect the human capacity for welfare to be >100x that of a mosquito.

You are implying excruciating pain is 128 k (= 51*365.25*24/3.5) times as intense as fully healthy life, which is 1.28 (= 128*10^3/(100*10^3)) times as intense as I assumed. If by ">100x" you mean something like 300, and you are referring to "capacity for welfare" per unit time, my ratio between the intensity of fully healthy life in mosquitoes and humans would be 3.90 (= 300*0.0130) times as high as yours. So you would value averting excruciating pain in mosquitoes 32.8 % (= 1.28/3.90) as much as I do. The harm to mosquitoes would continue to be roughly proportional to the value of averting excruciating in mosquitoes, given your intensity of excruciating pain is higher than mine. So AMF would cause 250 (= 763*0.328) times as much harm to mosquitoes as it benefits humans under your own assumptions. I guess you would want to consider other adjustments, but this illustrates you can get surprising results starting from reasonable assumptions.

bruce @ 2025-02-05T03:08 (+6)

The values I provide are not my personal best guesses for point estimates, but conservative estimates that are sufficient to meaningfully weaken your topline conclusions. In practice, even the assumptions I just listed would be unintuitive to most if used as the bar!

I agree "what fits intuition" is often a bad way of evaluating claims, but this is in context of me saying "I don't know where exactly to draw the line here, but 14.3 mosquito days of excruciating suffering for one happy human life seems clearly beyond it." 
It seems entirely plausible that a human might take a tradeoff of 100x less duration (3.5 hours * 100 is ~14.5 days), and also value human:mosquito tradeoff at >100x. It wouldn't be difficult to suggest another OOM in both directions for the same conclusion.

The main thing I'm gesturing at is that for a conclusion as unintuitive as "2 mosquito weeks of excruciating suffering cancels out 1 happy human life", I think it's reasonable to consider that there might be other explanations, including e.g. underlying methodological flaws (and in retrospect perhaps inconsistent isn't the right word, maybe 'inaccurate' is better). 

For example, by your preferred working definition of excruciating pain, it definitionally can't exist for more than a few minutes at a time before neurological shutdown. I think this isn't necessarily unreasonable, but there might be failure modes in your approach when basically all of your BOTECs come down to "which organisms have more aggregate seconds of species-adjusted excruciating pain".
 

Vasco Grilo🔸 @ 2025-02-05T11:28 (+2)

For example, by your preferred working definition of excruciating pain, it definitionally can't exist for more than a few minutes at a time before neurological shutdown. I think this isn't necessarily unreasonable, but there might be failure modes in your approach when basically all of your BOTECs come down to "which organisms have more aggregate seconds of species-adjusted excruciating pain".

I estimate each mosquito affected by ITNs experiences 1.98 min (= 119/60) of excruciating pain, which is in line with the definition.

bruce @ 2025-02-05T12:09 (+6)

The claim isn't that your answers don't fit your definitions/methdologies, but that given highly unintuitive conclusions, one should more strongly consider questioning the methodology / definitions you use.

For example, the worst death imaginable for a human is, to a first approximation, capped at a couple of minutes of excruciating pain (or a couple of factors of this), since you value excruciating pain at 10,000 times as bad as the next category, and say that by definition excruciating pain can't exist for more than a few minutes. But this methodology will be unlikely to accurately capture a lot of extremely bad states of suffering that humans can have. On the other hand, it is much easier to scale even short periods of excruciating suffering with high numbers of animals, especially when you're happy to consider ~8 million mosquitos killed per human life saved by a bednet - I don't have empirical evidence to the contrary, but this seems rather high.

Here's another sense check to illustrate this (please check if I've got the maths right here!):
-GiveWell estimate "5.53 deaths averted per 1000 children protected per year" or 0.00553 lives saved per year of protection for a child, or 1 life saved per 180.8 children protected per year.
-They model 1.8 children under each bednet, on average.

This means it requires approximately 100 bednets over the course of 1 year to save 1 life/~50 DALYs.

At your preferred rate of 1 mosquito death per hour per net[1] this comes to approximately 880,000 mosquito deaths per life saved,[2] which is 3 OOMs 1 OOM lower than the ~8 million you would reach if you do the "excruciating pain" calculation, assuming your 763x claim is correct[3]

(I may not continue engaging on this thread due to capacity constraints, but appreciate the responses!)

  1. ^

    Here I make no claims about the reasonableness of 1 mosquito per hour killed by the net as I don't have any empirical data on this / I'm more uncertain than Nick is but also note that he has more relevant experience than I do here.

  2. ^

    180.8/1.8 * 24* 365 = 879,893

  3. ^

    Assuming 763x GiveWell is correct, a tradeoff of 14.3 days of mosquito excruciating pain (MEP) for 1 happy human life, 2 minutes of MEP per mosquito, this requires a tradeoff of 7.9 million mosquitos killed for one human life saved.

    763*(14.3*24*60)/2 = 7,855,848

Vasco Grilo🔸 @ 2025-02-05T12:36 (+3)

On the other hand, it is much easier to scale even short periods of excruciating suffering with high numbers of animals, especially when you're happy to consider ~8 million mosquitos killed per human life saved by a bednet - I don't have empirical evidence to the contrary, but this seems wildly high.

Great point! It makes sense that harm is proportional to the number of individuals holding the pain caused to each individual constant. So my numbers would only be off if there was a reason to think mosquitoes cannot experience excruciating pain for as much time as humans.

This means it requires approximately 100 bednets over the course of 1 year to save 1 life/~50 DALYs.

"GW calculates a cost per distributed net and life saved of 6.78 $ and 5.10 k$ [in DRC]. These imply AMF has to distribute 752 nets [= 5.10*10^3/6.78] to save a life in DRC".

At your preferred rate of 1 mosquito death per hour per net[1] this comes to approximately 880,000 mosquito deaths per life saved,[2] which is 3 OOMs lower than the ~8 million you would reach if you do the "excruciating pain" calculation, assuming your 763x claim is correct[3]

There are 2 issues here:

  • 880 k is 1 order of magniture lower than 8 M, not 3.
  • I got 7.91 M (= 1*24*365.25*1.20*752) with the 1st approach, which practically matches the 8 M you got with the 2nd approach, multiplying:
    • 1 mosquito killed per net-hour.
    • 24 net-hours per net-day.
    • 365.25 net-days per net-year.
    • 1.20 net-years per net in DRC, as estimated by GW.
    • 752 nets per life saved in DRC, as estimated by GW (see above).

So I think it would be better if you edited the claim below you make at the start of your 1st comment in this thread.

I think you are probably at least a few OOMs off with these figures, even granting most of your assumptions, as this implies (iiuc) ~8 million mosquito deaths per human death averted.

I do not think you uncovered any errors in my calculations.

NickLaing @ 2025-02-05T07:26 (+15)

I think this is an interesting post and warrants discussion, but I ended up strongly downvoting because I don't want this to be the kind of discussion which is front and center on our public EA forum for external facing reasons and PR risk. This might well be the wrong response and many people might disagree with me on this which is fair enough.

I'll also flag that find the idea that saving poor kids' lives might be bad on the account of mosquito suffering makes me emotionally and almost physically squeamish, even if intellectually I think its an important discussion to have.

After flagging my biases, it still seems to me Vasco's analysis swings unnecessarily mosquito friendly, to the point where it almost feels motivated. I'm not saying it is motivated but there is a lack of "conservatism" or sanity checks in this estimate.

Vasco Grilo🔸 @ 2025-02-05T11:21 (+18)

Thanks, Nick.

I don't want this to be the kind of discussion which is front and center on our public EA forum for external facing reasons and PR risk. This might well be the wrong response and many people might disagree with me on this which is fair enough.

Bryan Walsh published a newsletter for Vox, Human progress has come at the expense of animals. It doesn’t have to., where there is a discussion of the EA Forum post Net global welfare may be negative and declining by Kyle Fish. In addition, "A version of this [Bryan's] newsletter originally appeared in the Future Perfect newsletter".

If AMF and the nets weren't doing their job, I would guess the alternatives might be at least as bad for mosquito welfare.

I do not think this is the relevant counterfactual. I assume ITNs (or malaria chemoprevention, or vaccines) save human lives more cost-effectively than the alternatives you mentioned. Otherwise, I would expect GiveWell to recommend them. So, to the extent people want to save lives as cost-effectively as possible, they will use the available resources to fight malaria to distribute ITNs. As a result, I think less donations to AMF simply result in less ITNs around. 

  • As many have mentioned, @Vasco Grilo🔸's blunt tool of estimating "excruciating pain" seems almost absurd. The idea that a mosquito experiences that degree of suffering as it dies of poison is possible but so unlikely that I think that best guesses for that number should be revised down by orders of magnitude. See other comments for in-depth arguments, I appreciated this comment from @bruce 

    "I don't know where exactly to draw the line here, but 14.3 mosquito days of excruciating suffering for one happy human life seems clearly beyond it."

I estimate 14.3 mosquito-days (= 51*365.25/0.013*10^-5) of excruciating pain neutralise the additional human welfare from saving a life under GW's moral weights multiplying:

  • 51 DALYs averted per life saved under GW's moral weights.
  • 365.25 days per year.
  • The ratio between the intensity of fully healthy human and mosquito life, which I set to the reciprocal of RP’s median welfare range of black soldier flies of 0.013.
  • The ratio between the intensity of fully healthy life and excruciating pain of 10^-5 (= 0.1/(10*10^3)). This implies 1 day of fully healthy life is neutralised by 0.864 s (= 24*60^2*10^-5) of "scalding and severe burning events [in large parts of the body]", or "dismemberment, or extreme torture".

Do the 3rd and 4th inputs seem "almost absurd"?

  • There's just no way mosquito nets kill anything like 24 mosquitos a day on average. I would guess this at under 5. Basing this estimate on Vasco getting a lot of bites while sitting outdoors at night (Mosquito nets are indoors) doesn't make much sense.

I was seeing much more than 1 mosquito per hour during dust outdoors in Moshi (Tanzania). People most affected by malaria may sleep indoors, but have houses with very limited ability to prevent mosquitoes from entering. The roofs can be made of turf too, which would attract mosquitoes.

I did not easily find empirical estimates for the number of mosquitoes killed by ITNs. I would be happy to update the number if you find them, although my takeaways would be the same even if ITNs only killed e.g. 10 % as many mosquitoes as I assumed.

  • Anchoring on RP's moral weights which I argue might be too animal friendly will often mean that any human helping intervention which hurts animals at all might look "bad". Especially in the case of smaller animals which are extremely different from humans, their use of binary behavioural proxies is almost guaranteed to give high moral weights.

As Bob Fischer said, "I don’t think you’ve said anything that should cause someone to question that headline result", "the ultimate question is whether our [RP's] decisions were wrong, not whether they can be construed as animal-friendly".

NickLaing @ 2025-02-05T13:19 (+5)
  1. I think this first example isn’t comparable, and a bit of a strawman. The Vox article is about how bad factory farming is, and how we don’t need to do that to help humans to flourish. This discussion is about potentially withdrawing life-saving interventions because they might be detrimental to animals. This directly connects the saving lives to harming animals – the Vox article doesn’t.
  2. I think you're basically right abobut the ITN counterfactual that if marginal funding goes away from ITNs then it might not go to alternatives. Less donations to AMF might simply result in less ITNS. If this argument though was taken up on a larger scale and AMF reduced net distribution, many people would want to save lives from and kill mosquito malaria regardless. The public health world isn’t likely tolerate malaria rapidly worsening because EAs decided Nets were bad for mosquitos. I think you are mainly right on this point though.
     
  3. On your imputs, its the 4th that seems most important to me but I'll comment on the 3rd too.

    I don’t think the 3rd input is absurd, but I do think it is probably too animal friendly. Like I said in my critique, their moral range calculation basically ends up as probability of sentience x behaviour score, and because all animals exhibit some of the behaviours they measure I think especially for smaller animals this is likely to overestimate their welfare. RPs behaviour score only really allows for a narrow range of something like between like 20 and 100 so its almost impossible for any small animal to come out of RP’s moral weights caculation with a negligibly small moral weight. In the article I was trying to be very measured and not make huge calls, but if you’re asking me what I might have done differently to weight animal behaviours, then I’d say I would have allowed for a much bigger behaviour score range, perhaps through having more than a binary yes/no system on some of the behaviours – giving less complex pain response behaviors a fraction of more complex ones.

    I think the 4th input seems absurd and I won’t rehash this much as many others have made arguments against your reasoning on this thread. You’re translating a figure which is on the upper bound of judging severe human pain (which like Bruce said, by definition can’t last long) directly onto what you think might be happening in mosquitos – a wildly different organism. On what grounds really would mosquitos dying of poisoning likely cause that severe pain at a best guess? I think its possible but very unlikely so I think it would be reasonable for the sake of conservatism to reduce this number by orders of magnitude.
     

4. On the number of mosquitos front for a start I don’t like comments like “my takeways would probably be the same even if….” Multipliers can add up, and we’re trying to move towards accuracy so I think it can be an unhelpful copout to question how much any element of an analysis matters – Rethink Priorities said things like this a number of times during their moral weights project which was a small red flag for me.

I agree there’s no empirical research on the mosquito number front, but from my perspective having travelled around Africa and living in a grass thatched hut and sleeping under a mosquito net for the last 10 years, 24 mosquitos killed a day on average per net seems extremely unlikely. That would be something like 240 million mosquitos killed by nets alone every day in Uganda – which seems to me perhaps plausible but unlikely. From a distance I think you could have been more conservative with your “best guess”

I’ve already discussed the RP thing above thanks! 

Vasco Grilo🔸 @ 2025-02-05T14:26 (+11)

I think this first example isn’t comparable, and a bit of a strawman. The Vox article is about how bad factory farming is, and how we don’t need to do that to help humans to flourish. This discussion is about potentially withdrawing life-saving interventions because they might be detrimental to animals. This directly connects the saving lives to harming animals – the Vox article doesn’t.

I understand they are not directly comparable, but I guess the newsletters from Vox also reach a much wider audience who is less acceptable of controversial takes compared to readers of the EA Forum.

I’d say I would have allowed for a much bigger behaviour score range

Do you mean you would account for other behavioural proxies for welfare capacity besides the 90 RP considered? Which ones? RP seemed to be super comprehensive.

perhaps through having more than a binary yes/no system on some of the behaviours

I do not understand. RP did not have a binary system to determine the probability of sentience.

The program runs 10,000 simulations where the presence or absence of each proxy in the “Simple_Scoring” spreadsheet was a random variable. For a given organism, the steps taken in a single simulation to generate the proxies possessed by that organism were:

  1. Start with a dictionary containing each proxy and an empty list to add scores to.
  2. For each proxy in the Simple Scoring sheet:
    1. First, randomly select the probability that the organism possesses the proxy from a uniform distribution whose range maps onto the judgment determined by our contractors. The probability map is as follows:
      1. No: [0.00, 0.00] (Used for the “motile” proxy)
      2. Likely no: [0, 0.25)
      3. Lean no: [0.25, 0.5)
      4. Lean yes: [0.5, 0.75)
      5. Likely yes: [0.75, 1.0]
      6. Yes: [1.00, 1.00] (Used for the “motile” proxy)

For example, if pigs scored a “Likely yes” on taste aversion behavior, then the probability that pigs exhibit taste aversion behavior is sampled uniformly over the interval [0.75, 1.0]. If a proxy was judged “Unknown”, then we defaulted to giving a zero probability of it being present; however, this default can be changed at the start of running the program.

  1. Second, generate a Bernoulli random variable using this probability of the species possessing the trait, where 1 indicates that the trait is present and 0 indicates that it is absent.
  2. Add the score (0 or 1) to the list corresponding to its particular proxy in the dictionary.

For a given organism, this process was repeated for 10,000 simulations, where each proxy’s score in a given simulation was appended to its respective list. Then, we repeated this procedure for all eighteen organism types studied and saved the simulated proxy data.

RP did not use a binary system to determine the welfare range conditional on sentience, and actually underestimated this by giving zero weight to proxies for which there was no information (see what I bolded below).

To generate the distributions of welfare ranges across species and models, the user must answer the same three questions about whether to give non-zero probability to “Unknown,” “Lean no,” and “Likely no” judgments and what weight should be given to proxies we’re highly confident matter for welfare ranges as were asked for the probabilities of sentience. As before, users can change the probabilities given to “Unknowns” for one or more species of their choosing. 

In our final simulations: 

  1. We chose not to assign any weight to proxies with “unknown” judgments for any species. This likely leads to underestimating the welfare ranges for several animal types.
  2. We chose to assign positive probabilities to proxies whose judgments are “likely no” and “lean no.”
  3. We weighted the proxies that we are highly confident matter for welfare ranges as being five times as important as all other proxies.

giving less complex pain response behaviors a fraction of more complex ones

I do not understand what you are referring to. Which specific proxies do you think should be weighted more heavily?

I think the 4th input seems absurd and I won’t rehash this much as many others have made arguments against your reasoning on this thread. You’re translating a figure which is on the upper bound of judging severe human pain (which like Bruce said, by definition can’t last long) directly onto what you think might be happening in mosquitos – a wildly different organism.

For all the analyses relying on pain intensities I am aware of, from AIM and RP, the ratio between the intensity of a pain of a given category and that of another is the same across species. I have now asked Cynthia Schuck-Paim, who is the research director of WFP (the organisation defining the pain intensities).

I agree excruciating pain "can’t last long", but this is consistent with my estimate that ITNs cause 119 s of excruciating pain per mosquito they kill.

On what grounds really would mosquitos dying of poisoning likely cause that severe pain at a best guess? I think its possible but very unlikely so I think it would be reasonable for the sake of conservatism to reduce this number by orders of magnitude.

As I say in the post, my estimates for the time in pain come from aggregating 3 sets of estimates provided by WFP's GPT Pain-Track. My estimate of 119 s of excruciating pain may well not be accurate, but which evidence do you have for it being "possible but very unlikely" to be that long?

4. On the number of mosquitos front for a start I don’t like comments like “my takeways would probably be the same even if….” Multipliers can add up, and we’re trying to move towards accuracy so I think it can be an unhelpful copout to question how much any element of an analysis matters – Rethink Priorities said things like this a number of times during their moral weights project which was a small red flag for me.

I agree. At the same time, I think it is worth having in mind how far one is from reversing the conclusions.

I agree there’s no empirical research on the mosquito number front, but from my perspective having travelled around Africa and living in a grass thatched hut and sleeping under a mosquito net for the last 10 years, 24 mosquitos killed a day on average per net seems extremely unlikely. That would be something like 240 million mosquitos killed by nets alone every day in Uganda – which seems to me perhaps plausible but unlikely. From a distance I think you could have been more conservative with your “best guess”

As I replied to Bruce:

I estimate GW's last grant to AMF will kill 0.0183 % as many mosquitoes as the ones currently alive globally over the lifetime of the bednets it funds. This would correspond to killing 1.19 % (= 1.83*10^-4*195/3) of the mosquitoes in DRC per year assuming mosquitoes were uniformly distributed across the existing 195 countries, and that the nets funded by the grant are distributed over 3 years. In reality, it would be less because DRC should have more mosquitoes than a random country. AMF being responsible for killing less than 1.19 % of the mosquitoes in DRC does not sound implausible.

Laura Duffy @ 2025-02-06T17:50 (+11)

Hi Vasco, 

Thanks for this interesting post, and in general for the amount of time and consideration you’ve given to analyzing animal welfare issues here on the Forum. I want to reiterate the points others in this comment section, and urge you to consider much more explicitly the wide range of uncertainty involved in asking a question like this. In particular, the following model choices are in my opinion deserving of a more careful uncertainty treatment in your analysis:

Though you mention there is uncertainty in each of these variables, I think that it’s important to consider how they multiplicatively add up when combined and their aggregate effect on the range of plausible results. Otherwise, there’s a good risk of arriving at a directionally incorrect conclusion that can have big consequences if we act too quickly on it. This, in my view, is especially true if you’re bringing a set of controversial assumptions to bear on a sensitive and morally important topic. 

NickLaing @ 2025-02-06T19:02 (+2)

I really appreciate you weighing in here Laura. It's reassuring to have someone from RP speak into this with some recognition of these issues.

Omnizoid @ 2025-02-04T11:27 (+10)

I mean, the alternative is the mosquitos dying later.  Do we have reason to expect their later deaths to be any better?  Given that mosquitos are R-strategists, it might be good to kill them if we think they live negative lives (which I do--I know you're less certain).

Vasco Grilo🔸 @ 2025-02-04T13:42 (+12)

Hi Matthew,

I think it is super unclear whether wild animals (which are overwhelmingly r-strategists) have positive or negative lives. So I implicitly assumed mosquitoes not affected by ITNs have neutral lives, such that the welfare loss caused by ITNs corresponds to a net welfare loss. I agree ITNs may be beneficial if mosquitoes not affected by ITNs have negative lives, but they may also be more harmful if they have positive lives. There is also large uncertainty about how decreasing the number of mosquitoes impacts the welfare of other wild animals. So I conclude "it is unclear to me whether ITNs increase or decrease welfare".

I recommend pursuing actions which are robustly beneficial under uncertainty instead of acting as if our best guesses are correct. In particular:

I believe the large uncertainty about the effects of human welfare interventions on wild (and farmed) animals should push one towards prioritising:

  • Animal welfare interventions improving the conditions of animals instead of decreasing the number of animals with negative lives, or increasing the number of animals with positive lives. I recommend donating to the Shrimp Welfare Project (SWP), which I estimate has been 64.3 k times as cost-effective as GW’s top charities (neglecting their effects on animals).
  • Learning more about helping invertebrates, whose total capacity for welfare vastly exceeds that of vertebrates. I recommend donating to (I ordered the organisations alphabetically):
    • The Arthropoda Foundation. Their research priorities are humane slaughter protocols, stocking densities and substrate research, and automated welfare assessment.
    • The Wild Animal Initiative (WAI). For instance:
      • They intend â€śto use current and new funding” for, among other activities, “Conducting an analysis of agricultural pest control to better understand the best targets for welfare interventions — first identifying scientific gaps and then developing research plans to help fill them”.
      • I estimate paying farmers to use more humane pesticides to decrease the suffering of wild insects is 23.7 k times as cost-effective as GW’s top charities.
Omnizoid @ 2025-02-04T15:48 (+4)

I disagree with you about wild animal welfare--I think it's clearly negative.  I agree though that we should be cautious and give to the wild animal institute.  But even if they have positive lives, if they'll still die eventually, this just pushes them back to have another more painful death.  

Do you think paying for more human pesticides is more effective than SWP?  And is there a charity doing that? 

Vasco Grilo🔸 @ 2025-02-04T18:12 (+2)

But even if they have positive lives, if they'll still die eventually, this just pushes them back to have another more painful death.

I would much prefer ITNs which painlessly kill mosquitoes, but even these may be harmful. They would still decrease the population of mosquitoes, which would be harmful for mosquitoes if they had positive lives.

Do you think paying for more human pesticides is more effective than SWP?  And is there a charity doing that? 

"I estimate paying farmers to use more humane pesticides to decrease the suffering of wild insects is 23.7 k times as cost-effective as GW’s top charities", which is 36.9 % (= 23.7*10^3/(64.3*10^3)) of my estimate for the past cost-effectiveness of SWP. I do not know about organisations paying farmers to use more humane pesticides. However:

I guess research on and advocacy for more humane pesticides are way more cost-effective than paying farmers to use them more. For example, I estimate cage-free campaigns improve 408 (= 10.8/0.0265) times as many chicken-years per $ as buying organic instead of barn eggs, which I determine from the ratio between 10.8 chicken-years per $ improved by cage-free campaigns, and 0.0265 chicken-years per $ (= 1/37.8) improved by buying organic instead of barn eggs. So I guess research on and advocacy for more humane pesticides is more cost-effective than SWP has been.

One can support research on more humane pesticides by donating to Wild Animal Initiative (WAI). They intend “to use current and new funding” for, among other activities, “Conducting an analysis of agricultural pest control to better understand the best targets for welfare interventions — first identifying scientific gaps and then developing research plans to help fill them”. So I think WAI is much more cost-effective than cage-free campaigns if a decent fraction of their additional funds go towards that, or if their various activities do not differ dramatically in cost-effectiveness, which both seem plausible. Feel free to contact Simon Eckerström Liedholm to learn about WAI’s work on humane pesticides.

There is lots of uncertainty in many of the inputs to my calculations, and I do not know which fraction of WAI's additional funds are going towards research on more humane pesticies. So I do not have a strong view about whether WAI is more or less cost-effective than SWP. Arthropoda Foundation, SWP and WAI are my current top donation options.

wild animal institute

Nitpick. Wild Animal Initiative.

Angelina Li @ 2025-02-03T21:10 (+10)

(A nitpick responding to just one point and not the whole post)

I feel a bit wary of using SWP as your default example here because this comment from @Aaron Boddy🔸 makes me think that SWP doesn't have a ton of room for rapidly deploying more funding right now -- I'd expect further donations to have lower marginal ROI than directly buying more stunners which I /expect/ is what you're assuming in the counterfactual (which is not a dig at SWP, seems fine for them to use extra donations for lower marginal ROI things if their top priority tickets are comfortably funded!)

A broader point: I'd expect the marginal ROI of additional funding in the invertebrate welfare space to diminish much faster than the marginal ROI of AMF donations.

Vasco Grilo🔸 @ 2025-02-03T23:19 (+2)

Thanks, Angelina.

I feel a bit wary of using SWP as your default example here because this comment from @Aaron Boddy🔸 makes me think that SWP doesn't have a ton of room for rapidly deploying more funding right now

Aaron said SWP's room for more funding in 2025 was "Something in the ballpark of a few hundred thousand dollars". I guess my post will influence much less than this, but I would also be happy with supporting SWP in subsequent years. SWP's marginal cost-effectiveness would ideally be the same across time. If that was not the case, SWP should move funds from the worst to the best years until the marginal cost-effectiveness is the same across years. I think this is hard to achieve over long time horizons, but I assume their marginal cost-effectivenes is 2025 will not be too different from that in 2026.

I also see Arthropoda Foundation and WAI as great donation options.

A broader point: I'd expect the marginal ROI of additional funding in the invertebrate welfare space to diminish much faster than the marginal ROI of AMF donations.

I tend to agree, because there is much more funding going towards AMF. On the other hand, donating to organisations doing research on invertebrate welfare like Arthropoda and WAI can open up more funding opportunities in the space. In any case, donating to invertebrate welfare seems way more cost-effective as of now. I do not even know whether donating to AMF is beneficial or harmful.

Nithin Ravi @ 2025-02-03T19:50 (+6)

Thanks for writing this! I've wondered this and would be interested in seeing something similar for screwworms as well, if you ever get around to estimating that.

I'm also curious to know why you chose the same median welfare range as black soldier flies. Is this just the best guess you had, or is there a reason that mosquitoes would have similar experiences to them?

Vasco Grilo🔸 @ 2025-02-03T22:44 (+5)

Thanks for the comment too, Nithin.

I've wondered this and would be interested in seeing something similar for screwworms as well, if you ever get around to estimating that.

I did some quick calculations about this which suggest it is unclear whether eradicating screwworms is beneficial or harmful. If they have positive lives, I think the decrease in welfare caused by them not existing may well outweight the increase in welfare of the animals which would no longer be infected by them.

I'm also curious to know why you chose the same median welfare range as black soldier flies. Is this just the best guess you had, or is there a reason that mosquitoes would have similar experiences to them?

Thanks, @Nithin Ravi[1]. I have added the following to the post.

Mosquitoes belong to the order Diptera, and black soldier flies are the only species analysed by RP of that order.

I tried to get a better estimate, but without success.

I asked Bob Fischer, who led RP’s moral weight project, about a best guess for the median welfare range of mosquitoes that RP would have obtained if they had analysed them. I privately disclaimed I would publish this post without Bob’s guess, but Bob did not share one.

  1. ^

    I am tagging you because I am expanding the comment starting with this sentence.

Daniel Birnbaum @ 2025-02-03T23:03 (+5)

Perhaps I’m misunderstanding something, so please correct me if I’m wrong: 

If one accepts all these assumptions, why would the best course of action be to offset AMF donations rather than to avoid donating to AMF in the first place? 

If ITNs cause vastly more harm to mosquitoes than they help humans, wouldn’t this imply that AMF is not just a weak investment, but actually a net-negative intervention? It seems like these numbers, if taken seriously, suggest AMF should be deprioritized rather than merely balanced with shrimp welfare donations. 

I assume that this is mostly about hedging against uncertainty under diff moral theories, but it seems like making this tradeoff of offset compared to counterfactual more money to AMF implies a certain tradeoff that you're okay with such that you should never make the initial investment. 

I'm confused about what sorta epistemic/ moral uncertainty theory someone would need to be offsetting the way you propose. Tbh I've already confused myself with this comment, but I hope it's helpful(?)

Vasco Grilo🔸 @ 2025-02-04T00:04 (+3)

Thanks for the comment, Noah.

If one accepts all these assumptions, why would the best course of action be to offset AMF donations rather than to avoid donating to AMF in the first place?

Sorry for the confusion. By "I would move any marginal donations from helping humans to helping invertebrates", I meant I would not donate to AMF. I have now clarified this in the post. "Nonetheless, I think directing a small fraction (10 %?) of one’s donations to helping invertebrates would be a good compromise to offset potential negative effects", even if I think donating just to SWP, WAI or Arthropoda Foundation would be better.

If ITNs cause vastly more harm to mosquitoes than they help humans, wouldn’t this imply that AMF is not just a weak investment, but actually a net-negative intervention? It seems like these numbers, if taken seriously, suggest AMF should be deprioritized rather than merely balanced with shrimp welfare donations.

I do not know whether AMF is beneficial or harmful. My mainline numbers suggest it is harmful, but there is lots of uncertainty, and I have not covered all the relevant effects.

  • I neglected the effects of ITNs on the number of wild animals because it is super unclear whether they have positive or negative lives. Yet, there is still lots of uncertainty even just in the effects I considered. Just accounting for uncertainty in mosquitoes’ capacity for welfare, I estimate the 5th and 95th percentile harm to mosquitoes caused by ITNs are 0 and 11.5 k times their benefits to humans. So it is unclear to me whether ITNs increase or decrease welfare.

I assume that this is mostly about hedging against uncertainty under diff moral theories, but it seems like making this tradeoff of offset compared to counterfactual more money to AMF implies a certain tradeoff that you're okay with such that you should never make the initial investment. 

I'm confused about what sorta epistemic/ moral uncertainty theory someone would need to be offsetting the way you propose. Tbh I've already confused myself with this comment, but I hope it's helpful(?)

I have not thought about this. I just think some people are drawn to offseting, and that donating to animal and human welfare is better than just to human welfare (although I also believe that donating just to animal welfare would be even better).

Noah Birnbaum @ 2025-02-04T16:24 (+5)

Thanks for the response. 

The part that I'm still stuck on is that this last part about the implicit tradeoff in one's offset seems crucial. The degree of offsetting is entirely based on tradeoff (maybe with some risk aversion under diff moral theories), but if you put that much into offsetting than it seems like you either have a major moral or epistemic disagreement with those that are donating in the first place. If that is the case, one person has got to give (either they don't offset near this much or they don't donate to AMF). 

While I'm here, I also wanted to thank you for writing this post. Super interesting, thoughtful, and I've shared with a bunch of people already! 

Vasco Grilo🔸 @ 2025-02-04T17:51 (+2)

The degree of offsetting is entirely based on tradeoff (maybe with some risk aversion under diff moral theories), but if you put that much into offsetting than it seems like you either have a major moral or epistemic disagreement with those that are donating in the first place.

I think you are right. In my mind, one's marginal donations should always go towards their best donation option, and therefore small donors should just donate to their best donation option. This is in contrast with donating to one organisation for offsetting negative impacts on animals, and then simultaneously donate to other organisations. 

While I'm here, I also wanted to thank you for writing this post. Super interesting, thoughtful, and I've shared with a bunch of people already! 

Thanks for the kind words, and sharing the post around, Noah!

CB🔸 @ 2025-02-03T19:00 (+5)

Thanks for writing this. 

This is an important consideration that almost nobody has talked about (hence my comment that flagged the topic).

Despite the uncertainty, this might well change completely the expected value of bednets, if they are not accompanied with some actions such as donations to offset the negative effects.

Jamie E @ 2025-02-03T23:33 (+4)

I've heard people talk about it quite a lot, usually as a joke

CB🔸 @ 2025-02-04T06:22 (+2)

Really? I haven't seen that. 

I've seen rapid joke comments about the killing of mosquitoes, but it's pretty rare that people talk about the suffering caused by insecticides - especially in an attempt to quantify and compare seriously.

If anybody else has done a serious analysis, I'd be interested. But I don't expect much more depth than arguments similar to "these AI safety people believe in terminator sci-fi scenarios, how silly".