what should we eat?
let’s think about what we eat. i’ve been thinking recently that i should stop eating certain animal products, less motivated out of concern for the well-being of the animal, but more for the well-being of humans. namely, there are easily identifiable, as well as subtle, reasons that eating meat and the prevalence of factory farming are bad for humanity. the obvious ones include: eating meat is a poor allocation of resources, and the meat industry contributes hugely to carbon emissions.
in defense of focusing on food. for starters, everyone is unethical.
one thing that bothers me when someone decides to drive electric, or go vegan, or cut down on panda burgers, is that their focus is too narrow. for example, say you go electric. maybe you’ve thought about this, or maybe you were just told it helps, regardless, it’s like you’ve picked one unethical thing out of hundreds of unethical things you help contribute to, and made an easy change. feeling superior isn’t useful and self-righteousness is not productive. you’ve gone electric, but you still eat tortured animals, own electronics full of conflict minerals, wear clothes made by oppressed third world laborers, etc etc etc. so it’s great that you thought about one bad thing you might contribute to and made a change! but there are still lots of other bad things and you should subject all aspects of your life to the same scrutiny and try and hold yourself to an even standard. with this in mind, we are going to focus on food. noting that we should hold ourselves to the same standard for everything we own/do/buy/etc. but one thing at a time. the one thing now is what we eat.
bad things about eating animals
ok, let’s now proceed by listing the potential harms associated to eating food, and how important we think they are. at this stage, this is subjective, but we will try and re-evaluate each point with data and make the reasons more quantitative.
i.) eating animals is a huge waste of resources. there are a lot of people in the world. there are also a lot of starving people in the world. feeding animals to then eat those animals is a pretty huge waste. it is estimated that we could eliminate most of humanity’s starvation with 40 million tons of food. but approximately 20 times this is fed to livestock every year.
ii.) the meat industry is a huge contributor to global carbon emissions. ruminates produce a lot of methane to break down grains in their quadra-stomachs. in the US, agriculture is more of a contributor to carbon emissions than the automobile industry. this is mostly because methane is a way worse greenhouse gas than CO2.
iii.) livestock take up land. this is essentially the same point as using resources, but here the wasted resource is land, not food. also, the food for our food takes up land.
iv.) meat is bad for you. some meats more than others. red meat had officially been declared carcinogenic and contributes to heart disease, high blood pressure. also, people getting heart attacks is a cost to society.
v.) torturing animals maybe isn’t a super awesome thing to do. this one is hard to quantify. i think i kind of agree that the mass subjugation of like 20 species to a system of torture and murder is a bad thing. but it’s hard to pinpoint what is bad. i think humans are more important than animals and morally am willing to expend animal lives to further human goals. billions of mice have died to cure disease and better our understanding of the brain, our bodies, answer fundamental questions in biology, etc. also, some animals are dumber than others, and some animals hardly experience a neurological response similar to what we might call pain. we’ll come back to this, but i think this will likely only play a role as an afterthought, as in we’ll get to pat our backs when we end up not eating the smarter animals.
vi.) antibiotics. we give factory farm animals lots of antibiotics. two facts: one, livestock farms are hotbeds for bacteria growth, this is what happens when you jam a bunch of animals into styes, we fix this with antibiotics. two, the antibiotic ‘revolution’ was way less of a revolution than you think. since the 1940’s, scientists continue to develop new antibiotics, and every time, resistant strains of bacteria develop, usually before or at the same time the antibiotic is approved for use. chart here. anyway, giving humans lots of antibiotics is bad, overuse of antibiotics in hospitals is bad. but pumping all the animals in factory farms full of antibiotics is the dumbest thing ever. it is such an unbelievably huge waste of a really really expensive, delicate, and finite resource.
ok, these are our bad things with which we want to construct a metric
related problems (but not as relevant here?)
farmers picking food. farmers are typically very overworked. your hippie kale can be as earth friendly and lovey hugs as you want (i actually love kale), but it was still picked by a human. that human likely worked long hours, for minimal pay, in pretty shitty working conditions. this is a somewhat separate point though. this is really an issue with workplace regulations and should obviously be addressed, but it is not a reason not to eat food. i will gladly pay more for the food if it means a human suffered less.
food waste. wasting food is dumb. a huge amount of food is wasted in this country. but again this is not a core problem with the product, but more about misguided aesthetic standards that must be met, bad regulations, restrictions on donating food, and people wasting things. if you can show me that some industries systematically waste food, in a way that is fundamentally related to that product and not some regulation, then i will add this to the above.
oooook, with this laid out, let’s rank in order of importance. honestly, i think i’m willing to rank i, ii, and vi as all equally the worst, and iii and iv as secondary bad things (since they are kind of the result of the first bad things). and v is bad, but the least bad, sorry animal lovers. i would love to hear moral arguments about why this might matter more.
let’s get some numbers and weigh out some facts:
industrial farms have been feeding livestock antibiotics since 1946. this became super wide spread super quickly when studies showed that antibiotics made the animal grow faster and bigger, hence more profit. antibiotics are fed to livestock, poultry, and fish, so basically everything. and an estimate 80% of antibiotics in the US each year are given to factory farmed animals, i.e. 20% go to people. well that’s dumb. an estimated 75% of all antibiotics given to livestock passes through them and enters the environment, hence speeds up resistance. also, the national academy of sciences estimates that the increased health care cost associated with antibiotic resistance (i.e. in using ones that don’t work, developing new ones, etc) is $4 billion a year.
free range or generally bigger farms don’t need to use antibiotics are conditions aren’t as bad. federal standards mandate that labeling the product as certified organic prohibits antibiotic use. note: ‘no antibiotic use’ means never giving animals antibiotics. ‘no routine use’ means only given to sick animals and then wait some time before it can enter food supply.
global estimates of emissions of CO2-equivalents (i.e. greenhouse gas’s effect measured wrt CO2, ie equivalent amount of CO2), per species per year
beef cattle (2.5 gigatons) + dairy cattle (2.1 gigatons) = 4.6 gigatons of CO2-eq (beef cattle=meat, dairy cattle = dairy + meat)
pigs = 0.67, chickens = 0.61, other ruminants = 0.47
as a percentage of sector emissions: beef=41%, cattle milk=20%, pigs = 9%, chickens = 8%, other ruminants (meat+milk) = 6%
average intensities: (kg eCO2/kg protein) beef = 300, cattle milk = 80, pork = 50, chicken meat = 40, chicken eggs= 30.
emissions include supply chains, i.e. feed production and livestock production, fertilizers, etc
tuna, salmon, nuts, veggies, lentils, all seem to have pretty minor carbon footprint, fruit gets a little higher than expected, but still minor
the resources that are used in the production of food include land, water, grain, and energy. given that the population density of the planet is not such that we can make a clear argument for why using land is bad, or could quantify the collective negative impact of using land for agriculture, at least in some monetary sense, we will ignore land for now. the impact of energy use is environmental, and we have already quantified this in the carbon footprint of food. so we are left with what we will call water footprint and grain footprint.
first a few numbers for scale: about 5.7 trillion liters of water are used residentially per month in the US, 196 billion liters per day. an average American uses 670 liters of water per day (average African family uses 18). 6 liters per toilet flush, 10 liters per minute in the shower. at 1 drip per second a leaky faucet can leak 11000 liters per year. apparently, it takes 148000 liters of water to manufacture a new car.
water footprint of food includes: water the animal drinks, water used to irrigate land that the animal grazes, water used to grow crops that are fed to the animal, and water used in processing the meat.
estimates are (in liters of water/kg)
beef = 15500, chicken = 3900, pork=6000, cheese/butter=5000, milk=1000, eggs=3200,
nuts=9000, average for fruits ~ 900, average for vegetables ~300, bread=1300
also interestingly: coffee=140 per cup, and chocolate=24000 (by far the highest i could find)
kgs of grain and forage the animal is fed per kg of meat (kg grain+kg forage)
lamb: 21+30=51, beef: 13+30=43, eggs: 11, pork: 6, chicken=4, dairy=2
some more numbers
kg grain/kg animal weight: beef: 8, pork: 4, chicken: 2.5, milk: 1
% edible weight: beef: 40%, pork: 55%, chicken: 55%, milk: 95%
kg grain/kg edible weight: beef: 20, pork: 7, chicken: 4.5, milk: 1
animal’s efficiency in protein conversion: beef: 4%, pork: 10%, chicken: 20%
even though we are not taking into account land use, here are the numbers anyway
about 780 million acres of the total 2264 million acres of US land was grazed land including roughly 200 million acres of grazed lands owned by the government. this is 34% of all land in the country.
note: the numbers for water use seemed to vary a little and it was hard to ascribe a confidence to these numbers. but most of the published articles seemed to agree pretty closely.
this is much harder to quantify and nutrition science is hard, so we won’t go into the details here, but instead will simply defer to the experts. the world health organization has declared red meat carcinogenic. further, red meat and processed meat have been linked to high blood pressure, heart disease, chronic obstructive pulmonary disease, bowel cancer, and stomach cancer.
while there are nutritional benefits to eating meat, namely nutrients like healthy fats, omega-3’s, B12, creatine, carnosine, and high quality proteins (there are 9 amino acids we cannot produce and exist at high densities in meat). also, meat seems to be good for bone density. while most of these things can be found in other plants, foods, they exist in high densities in meat.
so the take home message here is: red meat is bad. white meat is good for you.
this is definitely the hardest to quantify. and whatever we can quantify we won’t really use and this will really just be an afterthought. if we are willing to eat plants we are already conceding the morality of eating something that is alive. maybe eating something that is more intelligent and more capable to perceive itself is less ok. this get’s difficult because we are breeding the animals in the first place, so it wouldn’t have been alive anyway. intelligence and ability to perceive pain seem like reasonable things, but we wouldn’t eat dead/brain dead/less intelligent humans, nor would we eat animals we like or identify with (cats, dogs, monkeys, etc). so it seems like the metric is: intelligent animals that are self aware and able to perceive pain who we, as humans, do not identify with or feel emotionally attached to. again, this is hard.
so the shitty metric i will arbitrarily decide to use is number of cortical neurons: recent estimates put the number of neurons in the cerebral cortex for humans at about 21-26 billion, out of about 120 billion neurons in the human brain.
elephants have the most in the animal kingdom at 11 billion, whales have 10 billion. gorillas have 4 billion and chimpanzees have 6 billion. cats have 300 million, dogs have 160 million, and mice have 4 million.
of the things we eat: pigs have around 400 million, cows probably have nearer 1 billion, while chickens and fish seem to be down near 20 million or possibly even lower.
the numbers here were really hard to find and the estimates that do exist were pretty sketchy and based on some hand-wavey science.
the conclusion here is that cows are probably pretty smart, so are pigs, chickens and fish are probably pretty dumb. the bigger conclusion is that neuroscience is hard and not quite an exact science.
ok, what products are the worst offenders given the metric we’ve constructed and the data we’ve compiled.
there is one super obvious outlier: cows (and other ruminants).
in every single category ruminants were the worst, meaning cows, lamb, goat, etc.
beef contributes 4.6 gigatons of CO2 per year, this is 61% of all agriculture emissions
the intensity of emissions was 300 kg CO2/kg for beef and 80 for milk, this was higher than anything else.
the water footprint and grain footprint were the highest: 15500 liters of water/kg for beef and 51 kg grain+forage/kg, way higher than everything else
cows are also the least efficient in protein conversion: 4%
red meat is carcinogenic
cows are the smartest
DO NOT EAT: beef, milk, cheese, dairy, lamb, goat
no more beef. also, no more milk, yogurt, cheese, butter. it seems the footprint of dairy isn’t as bad, especially when taking into about the emissions/protein, but i seems very hard to dissociate this industry with the cattle industry as a whole.
Also when buying pork, chicken, eggs, milk, etc, make sure the product is labeled as organic, federal mandates forbid labeling a product as ‘certified organic’ if antibiotics were used.
more on morality (moral footprint)
we downplayed the morality of eating animals here. in fact we ranked it as the least bad bad-thing. by far the most prevalent justification for vegetarianism is the wrongness of killing animals. there is one obvious thing that many vegetarians overlook when they claim to not want to kill animals, if you buy milk or eggs or cheese you do kill animals. just cause you don’t eat the animal and didn’t want the animal to die, the animal that produces those products for you is still killed to sell as meat. this is simple profit-maximizing on the part of the factory farms. regardless of your personal ethics, you gave money to FINSIISH
and works out to be 4ish years for diary cows (out of a 25 year lifespan), and for chickens, hens are killed after one or two laying rounds. there are relatively few male chickens; hens are genetically modified to only lay female eggs. when male chickens are born there are usually killed right after they hatch. even if you are vegan, the agriculture industry is still responsible for the deaths of many animals. ideally…
there are a few interesting conclusions we reach if we say that we care about animals and that all animal lives are weighted equally. say we are not vegan but we to minimize the number of animals we eat conditioned on the
appendices and footnotes
units (an apology): the units i opted to use in each separate discussion were just the ones most common in the source material used for that particular issue, which means i jumped around between gallons, kg, etc. i apologize for this but as it was really only comparing numbers that concerned us, i don’t think this was very important. also, it saved me from having to do lots of unit conversions. also also, using the default units makes fact checking my numbers easier.
hmm subtleties in water percentages
see how US water use compares to other countries
why water scarcity is a problem and where it matters
water intensity of transportation
breakdown of indoor water use
more on beef production
water footprint data:
grain footprint data: