Carbon and food production, split from packaging

[dyqik]

Doing a top down analysis.

A quick google reveals that the total worldwide meat production was around 320 million tonnes. The total worldwide soybean production is around 350 millions tonnes. Of that soybean production about 85% of the protein is fed to animals as soybean meal.

Mature soybean has a roughly the same protein content (15%) vs beef (16%)

Other crops are also being fed to animals, eg: 70%-80% of maize's* one billions tonnes are being fed to animals. Maize is around 9-10% protein.

So to a first order approximation**, there is more than enough protein in existing maize and soybean crops being fed to animals to replace the protein from all global meat production, regardless of how those animals are raised. This could free up rangeland for other purposes.

*A big chunk is also turned into alcohol to put in cars, which is very very stupid.

**Yes, I'm not taking into account animals grown for dairy/eggs, yes, some essential amino acids would require other crops, and yes not everyone can eat soy protein plus a hundred other details.

The thing is, it would be a waste of already committed natural resources to make that conversion. Much of the hilly dairy/sheep farming type places would stop farming altogether - e.g. Vermont, Wales, the Alps, etc. - leaving those environments that are already adapted to hill farming to change to other uses, and displacing people from them to other places. You can't grow maize, soy etc. efficiently there, so you have to commit new agricultural land to doing so to replace the protein lost here, rather than using land that is already in agricultural use.

It's not an insurmountable problem, but it is an inefficiency.

[shpalman]

Well ok how about we keep growing animals which have enough to eat by grazing on land which you can't easily grow crops on, and then for the animals which don't have enough to eat and have to be fed soybean or whatever, we don't grow those animals, and we feed the soybean or whatever direct to humans? Isn't that more efficient?

[Gfamily]

There is an argument that allowing uplands to reforest (with associated carbon sequestration) would be better use environmentally and economically than the frankly marginal economics of supporting upland grazing.

[dyqik]

There is an argument that allowing uplands to reforest (with associated carbon sequestration) would be better use environmentally and economically than the frankly marginal economics of supporting upland grazing.

Yeah. But it really doesn't work as it means removing the major economic drivers and identities from large regions and even entire US states.

[bjn]

The thing is, it would be a waste of already committed natural resources to make that conversion. Much of the hilly dairy/sheep farming type places would stop farming altogether - e.g. Vermont, Wales, the Alps, etc. - leaving those environments that are already adapted to hill farming to change to other uses, and displacing people from them to other places. You can't grow maize, soy etc. efficiently there, so you have to commit new agricultural land to doing so to replace the protein lost here, rather than using land that is already in agricultural use.

It's not an insurmountable problem, but it is an inefficiency.

From my first order approximation, you don't have to add any extra land to replace meat taken out of production. All the protein could be grown with land now used to grow protein to feed to animals.

• 320Mt of meat * 16% = 52Mtonnes protein
• 350Mt of soybean * 85% * 15% = 44.5Mt protein currently being fed to animals
• 1000Mt of maize * 75% * 9% = 67.5Mt protein currently being fed to animals

From just two crops*, we feed over 110Mt of protein from crops to animals to get 52M tonnes of protein from meat. Maybe add another (?) 40Mt for dairy/eggs. If we stopped feeding that to animals and ate it instead, we'd have a protein surplus of about 20Mt. No extra land would be needed to be put into production for lack of meat protein from all sources, including grazed animals that aren't fed from crops.

I'm not saying this will happen, that it is desirable in all cases, that it wouldn't cause issues. I'm saying that the numbers indicate that a 100% vegan world would need less land in production to produce enough food for all of us. And by implication, excluding meat from your diet today will reduce the amount of land needed to feed you, going vegan even more so.

*These are the main two, but we also feed animals barley, rice, sorghum and peanuts to name a few more.

[Grumble]

There is an argument that allowing uplands to reforest (with associated carbon sequestration) would be better use environmentally and economically than the frankly marginal economics of supporting upland grazing.

Yeah. But it really doesn't work as it means removing the major economic drivers and identities from large regions and even entire US states.

It might make it politically unpopular in those regions, but that doesn’t mean it doesn’t work.

[dyqik]

There is an argument that allowing uplands to reforest (with associated carbon sequestration) would be better use environmentally and economically than the frankly marginal economics of supporting upland grazing.

Yeah. But it really doesn't work as it means removing the major economic drivers and identities from large regions and even entire US states.

It might make it politically unpopular in those regions, but that doesn’t mean it doesn’t work.

Er, yes, that does mean it doesn't work. That's how Brexit and Trump and sh.t like that happens, which makes any kind of coordinated political action impossible.

[Bird on a Fire]

There is in fact a lot of complexity involved in the idea that the world would go vegetarian. As a simple extrapolation, I would note that doing so would lead to an increase in the area required for crop production.

With a corresponding decrease in the area required for meat production, of course. All the detailed analyses I've seen show a reduction in total area of agricultural production with increasingly plant-based diets. Are you singling out cropland for a reason?

While the world technically is growing enough food to feed itself, if we didn’t feed all that grain to animals, the reality is not so (even including food waste in the discussion). The reason being that a lot of the food grown is the wrong sort of food. The world has become very good at growing calories, but much less good at growing plant protein.

Changing the world’s diet to plant only would mean a shift of protein sources from meat to plants. The Green Revolution resulted in more than doubling the yield of cereals/ha, but had almost no effect on the yield of pulse crops. In the same fields cereals will typically out yield pulse crops by a factor of 2 to 3. For example, average soybean yields in the USA are 3.3 T/ha, average corn yields are 10.5 T/ha. It turns out that increasing harvest index has been simple for cereals (grasses) but is really hard for other types of crops. No doubt more work could be funded to try and achieve this, but there is no guarantee that the same yield gains will occur.

Also there is the issue of protein content. Comparing chicken to chickpeas, you need to consume roughly 3 times in weight as much chickpeas to get the same protein content as chicken. Using the Australian Government Recommended dietary intakes as a place to argue from, if you shifted 1 person from meat to pulses for protein, you would have to grow 0.26 extra T/year of chickpeas to accommodate that. Moving just 1 million people from meat to chickpeas would require an extra 260,000 T of chickpeas/year or an extra 79,000 ha of chickpea production. That is assuming the chickpeas could be grown on the best arable land. If they were produced on new marginal land, more area would be needed.

This is interesting, and I don't know much about agronomy etcetera. But even without improvement, pulses yield similar amounts of protein per unit area to cereals, and far more than meat, with beef being roughly the least efficient producer of protein per unit area. (The results are qualitatively similar for calories per unit area)

You don't seem to mention what happens to the corresponding land used to produce all those chickens and their feed. You could grow chickpeas on it and have some left over, according to the models.

Much of the argument that is often made about growing crops to feed animals and how wasteful that is, is based on North American production systems where a large amount of grain is fed to animals. In most other areas of the world, this is not the case. Animals are pasture grazed, fed on crop residue and only occasionally supplemented with grain (many times not at all). Removing animal production would free up grain for human consumption in North America, although by less than people claim (see my comments above), but would have negligible impact in other parts of the world and increase the area required to grow protein crops. In many places, animal production occurs on land that is not suitable for crop production, so this land would have to come from somewhere else.

Yes, large areas marginal land would probably revert to forest, which would be a Good Thing, even if it required the loss of a smaller area of forest to increase crop production. The people modelling global land use requirements all seem to conclude that plant-based diets would be a much more efficient use of the world's area, even if it would require some reconfiguration of where things are.

However, studies like this (my bold):

We find that, given the current mix of crop uses, growing food exclusively for direct human consumption could, in principle, increase available food calories by as much as 70%, which could feed an additional 4 billion people (more than the projected 2–3 billion people arriving through population growth). Even small shifts in our allocation of crops to animal feed and biofuels could significantly increase global food availability, and could be an instrumental tool in meeting the challenges of ensuring global food security.

don't seem to share your conclusion that increasing total cropland area would be necessary. Their level of detail seems reasonable (41 major crops) and the FAO's data on allocation should be robust enough? What are we missing?

The recent EAT Lancet report, which was combining Rockstrom's planetary boundaries concept with optimising for nutrition, also came in favour of a largely plant-based diet.

While I am all for eating less meat, and particularly less red meat (actually I had my first steak (rare of course) in over 12 months on a recent holiday), meat free diets are not going to save the world from greenhouse gas emissions.

I was hoping to find a more recent set of data, but this is what an easy search came up with. This figure divides greenhouse gas emissions up by both gas and source. The main sources from animal agriculture alone (leaving aside the transport and crop production values, which ironically would both increase if eating animals was done away today) are CH4 from agriculture, where animals contribute about 2/3rds (the rest is from rice) and N2O from agriculture, where animals contribute about half. Compare those with the other sources and it becomes pretty obvious where the biggest gains could be had. That doesn’t mean the gains from reduced animal production should not be had, it is just they are much smaller than the average person thinks and won’t be nearly enough on their own.

For sure, changing agriculture is only part of the solution to climate change. It's probably more important in addressing the biodiversity crisis. As well as water resources, nitrogen pollution, phosphorus pollution, and some people would also highlight the ethical aspects, but I think that's an unnecessary derail

[Chris Preston]

Protein content of meat consumed ranges from roughly 22% for pork and fish to 28% for chicken. Cured meats have less protein.

Protein content of cooked legumes ranges from roughly 4% for green peas to 11% for some of the more exotic beans. Protein content of cooked maize is about 3.5%. Tofu, the main way humans consume soybeans, is about 8% protein.

Most of the soybeans grown in the world (90%) are crushed for oil and then the meal is fed to animals. Roughly 90% of that oil is used for human consumption, the rest for biodiesel.

Details. They always spoil a good story.

[Bird on a Fire]

I'm still not getting it.

Protein content of meat consumed ranges from roughly 22% for pork and fish to 28% for chicken. Cured meats have less protein.

Protein content of cooked legumes ranges from roughly 4% for green peas to 11% for some of the more exotic beans. Protein content of cooked maize is about 3.5%. Tofu, the main way humans consume soybeans, is about 8% protein.

Yes, you would have to eat more grams of legumes than grams of chicken in order to achieve the same level of protein intake. Such quantities are well within the realms of possibility, and indeed within normal meals.

Most of the soybeans grown in the world (90%) are crushed for oil and then the meal is fed to animals. Roughly 90% of that oil is used for human consumption, the rest for biodiesel.

I haven't looked in detail into how the FAO statistics are collected - if a batch of soy was pressed for oil for human consumption, presumably that would be counted as 'destined for human consumption', with 10% waste subsequently fed to animals?

[Chris Preston]

However, studies like this (my bold):

We find that, given the current mix of crop uses, growing food exclusively for direct human consumption could, in principle, increase available food calories by as much as 70%, which could feed an additional 4 billion people (more than the projected 2–3 billion people arriving through population growth). Even small shifts in our allocation of crops to animal feed and biofuels could significantly increase global food availability, and could be an instrumental tool in meeting the challenges of ensuring global food security.

don't seem to share your conclusion that increasing total cropland area would be necessary. Their level of detail seems reasonable (41 major crops) and the FAO's data on allocation should be robust enough? What are we missing?

We are good at growing calories. that is not at issue. We are much less good at growing plant protein. Increasing calorie production by 70% won't solve that issue. Many of the analyses I see just bundle all grains together and consider them interchangeable. We have massively increased cereal yields, grain legume yields are lagging.

[Bird on a Fire]

However, studies like this (my bold):

We find that, given the current mix of crop uses, growing food exclusively for direct human consumption could, in principle, increase available food calories by as much as 70%, which could feed an additional 4 billion people (more than the projected 2–3 billion people arriving through population growth). Even small shifts in our allocation of crops to animal feed and biofuels could significantly increase global food availability, and could be an instrumental tool in meeting the challenges of ensuring global food security.

don't seem to share your conclusion that increasing total cropland area would be necessary. Their level of detail seems reasonable (41 major crops) and the FAO's data on allocation should be robust enough? What are we missing?

We are good at growing calories. that is not at issue. We are much less good at growing plant protein. Increasing calorie production by 70% won't solve that issue. Many of the analyses I see just bundle all grains together and consider them interchangeable. We have massively increased cereal yields, grain legume yields are lagging.

That's why I posted a link to yields of protein per area. (I know I've seen better graphics than that, but google fu is failing me, and TBF it might have been in a book) But the figures seem to show that plants per area produce more protein.

The studies I've seen certainly distinguish between grains and legumes, and indeed between rice, wheat and corn at the very least, though legumes might all be bunched together.

[Bird on a Fire]

However, studies like this (my bold):

don't seem to share your conclusion that increasing total cropland area would be necessary. Their level of detail seems reasonable (41 major crops) and the FAO's data on allocation should be robust enough? What are we missing?

We are good at growing calories. that is not at issue. We are much less good at growing plant protein. Increasing calorie production by 70% won't solve that issue. Many of the analyses I see just bundle all grains together and consider them interchangeable. We have massively increased cereal yields, grain legume yields are lagging.

That's why I posted a link to yields of protein per area. (I know I've seen better graphics than that, but google fu is failing me, and TBF it might have been in a book) But the figures seem to show that plants per area produce more protein ETA than meat, and the Lancet commission report by actual dieticians endorses a drastic reduction in meat consumption.

The studies I've seen certainly distinguish between grains and legumes, and indeed between rice, wheat and corn at the very least, though legumes might all be bunched together.

[Chris Preston]

I'm still not getting it.

Protein content of meat consumed ranges from roughly 22% for pork and fish to 28% for chicken. Cured meats have less protein.

Protein content of cooked legumes ranges from roughly 4% for green peas to 11% for some of the more exotic beans. Protein content of cooked maize is about 3.5%. Tofu, the main way humans consume soybeans, is about 8% protein.

Yes, you would have to eat more grams of legumes than grams of chicken in order to achieve the same level of protein intake. Such quantities are well within the realms of possibility, and indeed within normal meals.

It is not that consuming enough legumes to get protein is not within the ability of people to do. It is being able to grow enough of the stuff to do so. The world population has exploded primarily because of the massive increase in production of cereals. These provide easy to produce, ship and consume calories to people, but not enough protein.

India grows 18.5 million T of pulse crops on 24 million ha, but still needs to import another 5 million T. India on the other hand is self-sufficient in wheat.

Most of the soybeans grown in the world (90%) are crushed for oil and then the meal is fed to animals. Roughly 90% of that oil is used for human consumption, the rest for biodiesel.

I haven't looked in detail into how the FAO statistics are collected - if a batch of soy was pressed for oil for human consumption, presumably that would be counted as 'destined for human consumption', with 10% waste subsequently fed to animals?

No. 90% of the total production is crushed. The oil component is roughly 50% of the total. The remainder is meal. About 97% of the meal is used for animal feed, the rest for human consumption. The remaining 10% of soybeans are used for a variety of other purposes: whole cooked soybeans, tofu production (although soybean meal is also used) and direct feeding to animals.

[Chris Preston]

However, studies like this (my bold):

don't seem to share your conclusion that increasing total cropland area would be necessary. Their level of detail seems reasonable (41 major crops) and the FAO's data on allocation should be robust enough? What are we missing?

We are good at growing calories. that is not at issue. We are much less good at growing plant protein. Increasing calorie production by 70% won't solve that issue. Many of the analyses I see just bundle all grains together and consider them interchangeable. We have massively increased cereal yields, grain legume yields are lagging.

That's why I posted a link to yields of protein per area. (I know I've seen better graphics than that, but google fu is failing me, and TBF it might have been in a book) But the figures seem to show that plants per area produce more protein ETA than meat, and the Lancet commission report by actual dieticians endorses a drastic reduction in meat consumption.

In many areas, animal production occurs on land not useful for growing crops.

The studies I've seen certainly distinguish between grains and legumes, and indeed between rice, wheat and corn at the very least, though legumes might all be bunched together.

How? Substitution for total weight of production? Substitution for total calories? Substitution for consumed protein requirement?

[Bird on a Fire]

Apologies for f.cking up the editing there - how embarrassing.

It is not that consuming enough legumes to get protein is not within the ability of people to do. It is being able to grow enough of the stuff to do so. The world population has exploded primarily because of the massive increase in production of cereals. These provide easy to produce, ship and consume calories to people, but not enough protein.

India grows 18.5 million T of pulse crops on 24 million ha, but still needs to import another 5 million T. India on the other hand is self-sufficient in wheat.

Right. But how does this relate to the argument about diverting the crops that are already produced to feeding humans rather than feeding livestock? And what is the problem with switching production from cereal to legume (other than rice paddies, obviously)

Presumably we can at least agree that diverting human-edible crops to livestock is wasteful? And that it totals about a third of production globally.

Most of the soybeans grown in the world (90%) are crushed for oil and then the meal is fed to animals. Roughly 90% of that oil is used for human consumption, the rest for biodiesel.

I haven't looked in detail into how the FAO statistics are collected - if a batch of soy was pressed for oil for human consumption, presumably that would be counted as 'destined for human consumption', with 10% waste subsequently fed to animals?

No. 90% of the total production is crushed. The oil component is roughly 50% of the total. The remainder is meal. About 97% of the meal is used for animal feed, the rest for human consumption. The remaining 10% of soybeans are used for a variety of other purposes: whole cooked soybeans, tofu production (although soybean meal is also used) and direct feeding to animals.
[/quote]
My bad - I misread that. So how would that be accounted for in the statistics - 50:50, all human, or all animal?

In many areas, animal production occurs on land not useful for growing crops.

The graphic is broken down into 'cropland' and 'pasture'.

The studies I've seen certainly distinguish between grains and legumes, and indeed between rice, wheat and corn at the very least, though legumes might all be bunched together.

How? Substitution for total weight of production? Substitution for total calories? Substitution for consumed protein requirement?

The graphic is tons of protein per hectare of cropland.

The EAT Lancet study was about achieving 'healthy diets from sustainable food systems', and was a large (and fascinating) collaboration between dieticians and ecologists. They were mainly concerned about eliminating malnutrition in developing countries, and concluded that more whole grains and legumes, and less animal proteins and starchy tubers, were the answer. I really recommend having a look at it. They don't seem to be making any of the assumptions you seem to be suggesting everybody makes.

[Bird on a Fire]

Basically, Chris, what you seem to be saying appears to go against large swatches of the literature.

If there are any studies showing that meat production is genuinely important to achieving global food security and ending malnutrition, could you post one or two? I'd love to have a read rather than pestering you with questions.

[Bird on a Fire]

Yeah. But it really doesn't work as it means removing the major economic drivers and identities from large regions and even entire US states.

It might make it politically unpopular in those regions, but that doesn’t mean it doesn’t work.

Er, yes, that does mean it doesn't work. That's how Brexit and Trump and sh.t like that happens, which makes any kind of coordinated political action impossible.

You are right to mention political realities of course, but I also think it's important to know what an ideal looks like sometimes. I think the evidence is reasonably clear that decreasing livestock production is desirable for a host of reasons. I doubt that 100% is feasible even in the medium-term, but that seems to be the right direction of travel.

[bjn]

Protein content of meat consumed ranges from roughly 22% for pork and fish to 28% for chicken. Cured meats have less protein.

Protein content of cooked legumes ranges from roughly 4% for green peas to 11% for some of the more exotic beans. Protein content of cooked maize is about 3.5%. Tofu, the main way humans consume soybeans, is about 8% protein.

Most of the soybeans grown in the world (90%) are crushed for oil and then the meal is fed to animals. Roughly 90% of that oil is used for human consumption, the rest for biodiesel.

Details. They always spoil a good story.

As I said, first order approximations done with 15 minutes of googling. BOAF’s references have more detail which contradict your position.

As to your specific points, soybean meal is 43 to 50% protein by weight, the oil being very low in protein. So we are feeding the vast majority of the world’s soybean protein to animals. Same with maize.

How we eat plant derived protein is a different question to how much we can grow. We can grow more than enough.

[shpalman]

However, studies like this (my bold):

We find that, given the current mix of crop uses, growing food exclusively for direct human consumption could, in principle, increase available food calories by as much as 70%, which could feed an additional 4 billion people (more than the projected 2–3 billion people arriving through population growth). Even small shifts in our allocation of crops to animal feed and biofuels could significantly increase global food availability, and could be an instrumental tool in meeting the challenges of ensuring global food security.

don't seem to share your conclusion that increasing total cropland area would be necessary. Their level of detail seems reasonable (41 major crops) and the FAO's data on allocation should be robust enough? What are we missing?

We are good at growing calories. that is not at issue. We are much less good at growing plant protein. Increasing calorie production by 70% won't solve that issue. Many of the analyses I see just bundle all grains together and consider them interchangeable. We have massively increased cereal yields, grain legume yields are lagging.

Herbivores aren't creating protein out of thin air, they are eating plant protein. Some of those plants are not directly useful to humans (i.e. grass) but we should stop feeding so much human-useful plant protein to animals if we're so "less good at growing plant protein". Herbivores are an inefficient way of concentrating plant protein, not a way of creating protein.

(Some plants do indeed create protein out of thin air).

[Woodchopper]

However, studies like this (my bold):

don't seem to share your conclusion that increasing total cropland area would be necessary. Their level of detail seems reasonable (41 major crops) and the FAO's data on allocation should be robust enough? What are we missing?

We are good at growing calories. that is not at issue. We are much less good at growing plant protein. Increasing calorie production by 70% won't solve that issue. Many of the analyses I see just bundle all grains together and consider them interchangeable. We have massively increased cereal yields, grain legume yields are lagging.

Herbivores aren't creating protein out of thin air, they are eating plant protein. Some of those plants are not directly useful to humans (i.e. grass) but we should stop feeding so much human-useful plant protein to animals if we're so "less good at growing plant protein". Herbivores are an inefficient way of concentrating plant protein, not a way of creating protein.

(Some plants do indeed create protein out of thin air).

It seems to me that the issue is the extent to which soy meal contains human useful protein.

For humans, soy oil is very useful for food production and as fuel, and almost all soy meal is fed to animals. Soy meal contains protein, which for human consumption is turned into soy flour. Soy flour is used to make soy milk, texturized vegetable protein and other products. Other popular soy products like tofu aren't made from soy meal.

Instead of feeding it to animals, humans could directly consume soy milk, texturized vegetable protein and other products made from soy meal.

However, according to the Tesco website, all soy milk is more expensive than cow milk, most of them two to three times more expensive.

Likewise, textured soy protein can be purchased from here in bulk for £6.15 per kilo, whereas Tesco will sell you chicken for £1.60 per kilo.

To reprise the discussion in the thread about how the beyond meat plant based burger was far more expansive than one made out of cow and purchased in McDonalds, there are several reasons for soy products being so much more expansive.

Certainly economies of scale and subsidies may play a role, but overall, if in a competitive market, a product is several times cheaper per kilo, then I'd expect that the overall costs of production would be much less.

People don't eat raw soy meal. It appears that turning protein from soy meal into something that humans want to eat involves processing, which is usually costly and uses resources. That processing needs to be taken into account when comparing the relative efficiency of different forms of food production.

Or to put it another way. If the situation were reversed and soy milk or textured soy protein were several times cheaper than c cow milk or chicken then large numbers of people would rapidly switch over. It would be a very good idea to look at how costs are distorted by subsidies or tariffs, and the effect of economies of scale. But we should also consider the prospect that after including processing costs, chicken thighs or cow milk isn't a particularly inefficient way of converting soy meal into something edible for humans.

[shpalman]

It seems to me that the issue is the extent to which soy meal contains human useful protein.

For humans, soy oil is very useful for food production and as fuel, and almost all soy meal is fed to animals. Soy meal contains protein, which for human consumption is turned into soy flour. Soy flour is used to make soy milk, texturized vegetable protein and other products. Other popular soy products like tofu aren't made from soy meal.

Instead of feeding it to animals, humans could directly consume soy milk, texturized vegetable protein and other products made from soy meal.

However, according to the Tesco website, all soy milk is more expensive than cow milk, most of them two to three times more expensive.

Likewise, textured soy protein can be purchased from here in bulk for £6.15 per kilo, whereas Tesco will sell you chicken for £1.60 per kilo.

If I normalize for the actual protein content of 44.6g/100g for the soy and 17.7g/100g for the chicken thighs, I get £13.78/kg of protein for the soy and £9.04/kg of protein for the chicken thighs. I prefer to eat chicken breast though and that's way more expensive at £7.62 per kilo (priced according to demand, I suppose), 24.0g/100g protein content, making it £31.75/kg of protein. Back when I was still able to buy 7 kg tubs of unflavoured 80% whey for less than 16 Euros per kg including shipping, that would have been about £17/kg of protein, which is why I was buying huge tubs of unflavoured 80% whey (and I never noticed soy or pea-based protein powder to be an economical alternative)... I'd be into the idea of buying plain chunks of vegetable protein to use instead of chicken or mince though, in the things I cook for myself once a week and then freeze in portions to bring to work to eat for lunch.

(I'm not disagreeing with you, by the way, just adding my own data points.)

To reprise the discussion in the thread about how the beyond meat plant based burger was far more expansive than one made out of cow and purchased in McDonalds, there are several reasons for soy products being so much more expansive.

Certainly economies of scale and subsidies may play a role, but overall, if in a competitive market, a product is several times cheaper per kilo, then I'd expect that the overall costs of production would be much less.

People don't eat raw soy meal. It appears that turning protein from soy meal into something that humans want to eat involves processing, which is usually costly and uses resources. That processing needs to be taken into account when comparing the relative efficiency of different forms of food production.

Or to put it another way. If the situation were reversed and soy milk or textured soy protein were several times cheaper than c cow milk or chicken then large numbers of people would rapidly switch over. It would be a very good idea to look at how costs are distorted by subsidies or tariffs, and the effect of economies of scale. But we should also consider the prospect that after including processing costs, chicken thighs or cow milk isn't a particularly inefficient way of converting soy meal into something edible for humans.

Leaving aside the technically challenges of something like the impossible burger which is trying to imitate meat as closely as possible, rather than just being a soy-based product which is pleasant to eat in its own right, it's extremely interesting to me that processing soy protein through an animal seems to be more efficient and cost effective than processing it "artificially".

Like I said, I'm not disagreeing with you, I'm just interesting in the discussion.

[discovolante]

Milk from tesco appears to be about 65 to 70p per litre, you can get uht soy milk for 90p per litre. But it is still more, you have to know how to find it, and it is rubbish in coffee.

[Woodchopper]

Milk from tesco appears to be about 65 to 70p per litre, you can get uht soy milk for 90p per litre. But it is still more, you have to know how to find it, and it is rubbish in coffee.

Obviously different prices in the shops, but a basic large bottle of semi skimmed is listed online for 48p per liter.

There was one soy milk for £0.59 per litre, but the rest are between £0.85-£160 per liter. The cheaper one has a lower proportion of soy compared to water, so that could explain the difference.

Better to compare the website prices so as to have a comparable data source.

[Woodchopper]

Leaving aside the technically challenges of something like the impossible burger which is trying to imitate meat as closely as possible, rather than just being a soy-based product which is pleasant to eat in its own right, it's extremely interesting to me that processing soy protein through an animal seems to be more efficient and cost effective than processing it "artificially".

Like I said, I'm not disagreeing with you, I'm just interesting in the discussion.

Yes, its interesting and I don't have an explanation, and there are all sorts of things that can distort the market.

But overall, something that is much more efficient should be much cheaper.