I calculated the welfare ranges per calorie consumption for a few species.
They vary a lot. The values for bees and pigs are 4.88 k and 0.473 times as high as that for humans.
They are higher for non-human animals:
- 5 of the 6 species I analysed have values higher than that of humans.
- The lower the calorie consumption, the higher the median welfare range per calorie consumption.

Methods

I calculated the welfare ranges per calorie consumption from the ratio between:

Rethink Priorities’ 5th percentile, median, and 95th percentile welfare ranges.
Calorie consumption per day for mature individuals. Note there is variation within the same species.

My calculations are here.

Results

The results are below by descending median welfare range per calorie consumption.

Welfare range and calorie consumption

Species	5th percentile welfare range	Median welfare range	95th percentile welfare range	Calorie consumption (kcal/d)
Bees^[1]	0	0.071	0.461	0.0327
Shrimp^[2]	0	0.031	1.149	0.831
Crayfish^[3]	0	0.038	0.491	4.88
Salmon^[4]	0	0.056	0.513	34.9
Chickens^[5]	0.002	0.332	0.869	300
Humans^[6]	1	1	1	2.25 k
Pigs^[7]	0.005	0.515	1.031	2.45 k

Welfare range per calorie consumption

Species	5th percentile welfare range per calorie consumption (d/kcal)	Median welfare range per calorie consumption (d/kcal)	95th percentile welfare range per calorie consumption (d/kcal)
Bees	0	2.17	14.1
Shrimp	0	0.0373	1.38
Crayfish	0	7.79 m	0.101
Salmon	0	1.61 m	0.0147
Chickens	6.67 μ	1.11 m	2.90 m
Humans	0.444 m	0.444 m	0.444 m
Pigs	2.04 μ	0.210 m	0.421 m

Species	5th percentile welfare range per calorie consumption as a fraction of that of humans	Median welfare range per calorie consumption as a fraction of that of humans	95th percentile welfare range per calorie consumption as a fraction of that of humans
Bees	0	4.88 k	31.7 k
Shrimp	0	83.9	3.11 k
Crayfish	0	17.5	226
Salmon	0	3.61	33.1
Chickens	1.50 %	2.49	6.52
Humans	1.00	1.00	1.00
Pigs	0.459 %	47.3 %	94.7 %

Discussion

The median welfare ranges per calorie consumption vary a lot. The values for bees and pigs are 4.88 k and 0.473 times as high as that for humans^[8].

In addition, the values are higher for non-human animals. 5 of the 6 species I analysed have values higher than that of humans. Moreover, the lower the calorie consumption, the higher the median welfare range per calorie consumption. These suggest welfare can be created more efficiently via small non-human animals. However, one should also account for:

The energy needed besides that in food.
The level of welfare as a fraction of the median welfare range.

I would still guess that it is possible to achieve higher total welfare with mostly small non-human animals, in agreement with animal welfare arguably being dominated by that of arthropods and nematodes. In the future, one might be able to produce welfare much more efficiently via non-biological beings.

^{^}
According to Mark Patterson, “in summer a full strength Honey Bee hive contains around 50-60,000 workers and needs to consume around 1800 calories per day”. So I used 0.0327 kcal/d (= 1.8/((50 + 60)/2)).
^{^}
A shrimp of 40 g is fed 0.84 g/d (= 0.021*40), and shrimp feed has 0.989 kcal/g (= 84/(3*28.3)). So I used 0.831 kcal/d (= 0.84*0.989).
^{^}
Adult crayfish weigh 24.8 g (= (23.29 + 26.40)/2), and therefore, “at a daily feeding rate of 5 % of body weight”, eat 1.24 g/d (= 24.8*0.05). The feed contains 3.93 kcal/g (= 16.43/4.184). So I used 4.88 kcal/d (= 1.24*3.93).

^{^}
“After two years at sea”, adult salmon have “8 to 12 pounds”, so it grows 6.21 g/d (= (8 + 12)/2*454/(2*365.25)). In addition, “Atlantic salmon needs 1.15 kg feed to gain 1 kg body weight”, which translates into 7.15 g/d (= 6.21*1.15) of feed, which contains 4.88 kcal/g (= (0.356 + 0.110)*4 + 0.335*9). So I used 34.9 kcal/d (= 7.15*4.88).
^{^}
According to Ken Macklin and Joe Hess, “a hen may consume up to 340 calories of metabolizable energy per day during winter to keep warm. But in summer, she may consume only 260 calories of metabolizable energy per day”. So I used 300 kcal/d (= (260 + 340)/2).
^{^}
According to the National Health Service, “the recommended daily calorie intake is 2,000 calories a day for women and 2,500 for men”. So I used 2.25 k kcal/d (= (2 k + 2.5 k)/2).
^{^}
Mean between the lower and upper bounds of 2.4 k and 2.5 k for breeder pigs.
^{^}
Fun fact, I was stung by a bee/wasp just today.

Show all footnotes

12 Reactions

Mentioned in

48Nuclear war tail risk has been exaggerated?

11Concepts of existential catastrophe

More posts like this

291

Why Neuron Counts Shouldn't Be Used as Proxies for Moral Weight

Adam Shriver

229

Meaningfully reducing meat consumption is an unsolved problem: meta-analysis

Seth Ariel Green 🔸+2 more

407

Can we help individual people cost-effectively? Our trial with three sick kids

NickLaing

Comments15

Sorted by

New & upvoted

Click to highlight new comments since: Today at 9:46 PM

Laura Duffy2y6

This is a really interesting project and way of approaching the topic!

One thing to note: welfare ranges don’t factor in the lifespans of animals, so we’d also need to factor in the typical time a farmed animal lives and then weight by welfare range to get a moral weight-adjusted sense of per calorie animal impacts.

But again, approaching this from a per calorie perspective is really interesting!

Vasco Grilo🔸2y4

I am glad you found it interesting, Laura!

To clarify, my post is about the calories consumed by the animals, not about the animal calories consumed by humans. However, I agree running a similar analysis for the calories provided per animal, and then factoring in their lifespan (and median welfare range, and ideally quality of life) would be quite interesting too! Just in case you are not aware, Brian Tomasik has a piece about "equivalent days of suffering caused per kg demanded". There is a website which weights days of suffering by various functions of the number of neurons, and also accounted for climate change, but I do not remember the name (Michael St. Jules should know). Maybe they could add an option to weight by Rethink's median welfare ranges.

I was thinking about welfare ranges per calories consumed by the animals with the goal of getting a sense of what type of beings would fill the world in order to increase welfare. For a given amount of energy, and welfare as a fraction of the welfare range, it looks like feeding bees produces much more welfare than feeding humans. Nevertheless, I do not think one should start tilling the universe with bees just yet! This is only one of many factors to consider.

Laura Duffy2y3

Oh I see! Thanks for the clarification!

Vasco Grilo🔸2y2

There is a website which weights days of suffering by various functions of the number of neurons, and also accounted for climate change, but I do not remember the name (Michael St. Jules should know).

I asked Michael, and the website is Food impacts, which is now using Rethink's median welfare range estimates instead of functions of the number of neurons. In the methodology page, other analyses of welfare impacts per calorie are mentioned:

Rankings based on welfare have been developed previously by various individuals and groups such as Peter Hurford, Brian Tomasik, Charity Entrepreneurship and Dominik Peters. This tool is a minor extension of the work of Dominik Peters that also considers emissions in addition to welfare. I want to thank Dominik for kindly providing the data and methodology that he used.

I think Julian Galed also did one many years ago, looking into days of animal living time per calorie.

FYI, I also have a short draft related to the badness of eating farmed animals as a fraction of the human goodness supported by their calories. Your comments are welcome, but no worries if you have other priorities now. I trust your decision-making! Update on July 22: published!

Chris Said4mo3

Hi Vasco, I'm not sure Food Impacts is calculating things correctly. They start off by calculating the number of animal hours lived to create 2,000 calories, which is reasonable. The next step should be to multiply that number by the average welfare of an animal, since that should tell you how many negative welfare units would be averted by not creating the demand for 2,000 calories.

But instead, they multiply by the welfare range of the animal.

This doesn't make sense to me. If a farmed cow's actual welfare is -0.1, why does it matter that its welfare range is -0.25 to 0.25? To figure out how much negative welfare I can avert, I care about the -0.1!

Am I thinking about this correctly?

And if so, is there a good resource for actual welfare values for farmed animals, rather than the theoretical ranges?

Vasco Grilo🔸4mo2

Thanks for the comment, Chris!

Am I thinking about this correctly?

Yes.

And if so, is there a good resource for actual welfare values for farmed animals, rather than the theoretical ranges?

I have estimated the welfare per living time of chickens in various conditions in animal quality-adjusted life years (AQALYs) per chicken-year. 1 AQALY corresponds to 1 year of a practically maximally happy life. As a rough approximation, you can get the welfare in QALYs mutiplying the welfare in AQALYs by Rethink Priorities' median welfare ranges^[1].

Animal	Broiler in a conventional scenario	Broiler in a reformed scenario	Hen in a conventional cage	Hen in a cage-free aviary
Welfare per living time (AQALY/year)	-2.27	-0.161	-1.69	-0.333

I have some estimates for shrimp too (this post has estimates for chickens, but these rely on underestimates of the time they spend in pain, whereas the ones above try to correct for this).

^{^}

This would only be 100 % correct if the welfare per time of the practically maximally happy life as a fraction of the welfare range is constant across species.

Chris Said4mo3

Thanks Vasco,

As a rough approximation, you can get the welfare an QALYs mutiplying the welfare in AQALYs by Rethink Priorities' median welfare ranges

I might be misunderstanding something, but I'm not sure that's right, even with your footnote. My understanding is that animal AQALYs per years and human QALYs per year both range from +1 at the top, to some species-specific negative value at the bottom. The same is true of the Rethink welfare units, but with a different scale. If so, shouldn't the formula be as described below?

Vasco Grilo🔸4mo2

This would only be 100 % correct if the welfare per time of the practically maximally happy life as a fraction of the welfare range is constant across species.

In this case, "maximum welfare of a chicken-year"/("maximum welfare of a chicken-year" - "minimum welfare of a chicken-year") = "maximum welfare of a human-year"/("maximum welfare of a human-year" - "minimum welfare of a human-year") <=> "maximum welfare of a chicken-year" = ("maximum welfare of a chicken-year" - "minimum welfare of a chicken-year")/("maximum welfare of a human-year" - "minimum welfare of a human-year")*"maximum welfare of a human-year". Since "welfare range of chickens" = ("maximum welfare of a chicken-year" - "minimum welfare of a chicken-year")/("maximum welfare of a human-year" - "minimum welfare of a human-year"), "maximum welfare of a chicken-year" = "1 AQALY in chickens", and "maximum welfare of a human-year" = "1 QALY", "1 AQALY in chickens" = "welfare range of chickens"*"1 QALY". So, given the condition I mentioned in the footnote, one can get the welfare in QALYs mutiplying the welfare in AQALYs by the welfare range.

Chris Said4mo3

Got it, thanks. For those following along at home, I misread your footnote and the graphs I made do not reflect the condition in the footnote.

Chris Said4mo1

If it makes things easier, you can copy the Google Slides source to tweak the illustration https://docs.google.com/presentation/d/1LuSpONztS9Tl0OSn-YeyWJG7B6UIYtff49p1WREPPgA/edit#slide=id.p

David M2y4

What is a welfare range?

Vasco Grilo🔸2y2

Hi David,

Thanks for asking! A welfare range is the difference between the best and worst possible states a being can experience.

Henry Howard🔸2y1

The error bars on the Rethink Priorities' welfare ranges are huge. They tell us very little, and making calculations based on them will tell you very little.

I think without some narrower error bars to back you up, making a post suggesting "welfare can be created more efficiently via small non-human animals" is probably net negative, because it has the negative impact of contributing to the EA community looking crazy without the positive impact of a well-supported argument.

Laura Duffy2y15

Hi Henry! While the 90% confidence intervals for the RP welfare ranges are indeed wide, this is because they’re coming from a mixture of several theories/models of welfare. The uncertainty within a given theory/model of welfare is much lower, and you might have more or less credence in any individual model.

Additionally, if we exclude the neuron count model, the welfare ranges from the mixture of all the other models have narrower distributions.

Here’s a document that explains the different theories/models used: https://docs.google.com/document/d/1xUvMKRkEOJQcc6V7VJqcLLGAJ2SsdZno0jTIUb61D8k/edit

And here’s a spreadsheet with all the confidence intervals from each theory/model individually (after adjusting for probability of sentience): https://docs.google.com/spreadsheets/d/1SpbrcfmBoC50PTxlizF5HzBIq4p-17m3JduYXZCH2Og/edit

Vasco Grilo🔸2y4

Hi Henry,

To be honest, that is a quite funny meme!

The error bars on the Rethink Priorities' welfare ranges are huge. They tell us very little, and making calculations based on them will tell you very little.

I have now added the 5th and 95th percentiles. Thanks for the nudge!

I think without some narrower error bars to back you up, making a post suggesting "welfare can be created more efficiently via small non-human animals" is probably net negative, because it has the negative impact of contributing to the EA community looking crazy without the positive impact of a well-supported argument.

I think the post is still beneficial, because I am not endorsing taking any specific actions to create welfare via small non-human animals. However, I think you have a good point, and I agree the post could plausibly be harmful (although my best guess is that it is beneficial!). I would only disagree with views strongly asserting that the post is harmful.

PS: I upvoted your comment.