The free lunch

I had hoped to be posting my review of The Omnivore’s Dilemma by now, but I’ve been a bit sidetracked by a calculation that’s just been begging to be done.

Michael Pollan writes that Polyface Farm produces what looks “an awful lot like the proverbially unattainable free lunch” (127). I’ve pointed out before that Polyface does use feed grains from other farms. What I want to look at is how Polyface’s animal product output might compare to using those grains to feed people. Because of the quality of the data, this calculation will unfortunately be only back-of-the-envelope quality, but I hope that it will still have some value.

I’ll first look at the grain used to feed Salatin’s broilers. The most recent printing of The Omnivore’s Dilemma states (page 222) that Salatin raises 12,000 broilers each year. (The productivity numbers in earlier printings were different. In fact, my copy even has different numbers in two different places.) According to Salatin’s Pastured Poultry Profit$, the broilers have a carcass weight of 4 pounds, and the chickens require 3 pounds of feed for each pound of carcass weight (page 185). That comes out to 144,000 pounds of feed each year. Salatin writes that he uses a feed consisting of 52% corn, 29% roasted soybeans, 11% crimped oats, and the rest consisting of fish meal, kelp and nutritional supplements. Since I’d like my estimate of the energy content of the feed to be conservative, I’ll ignore the calorie content of everything but the grains. Unfortunately, I wasn’t able to find good data for calorie content of feed grains, so I’ll be using data for food-grade grains. Using USDA data for corn, soy, and oats, one finds that a mixture of 0.52 pounds of corn, 0.29 pounds of soybeans, and 0.11 pounds of oats contains a little over 1700 calories. That comes out to 244,800,000 calories in the feed grain.

Now what about the productivity of Salatin’s farm? Here things get pretty tricky. There are several different kinds of animals, and each has several different kinds of meat with different calorie densities. For rabbits, I wasn’t able to find any useful data on carcass weights. Regardless, here is the calculation in tabular form. (References are at the end of the post. Where no reference is given, the source is the version of The Omnivore’s Dilemma available on Google Books.)

Food Quantity Energy density Total Energy (Cal)
Beef 25,000 lb 1163 Cal/lb 1 29,075,000
Pork 50,000 lb 1062 Cal/lb 1 53,100,100
Broiler hens 48,000 lb 2 615 Cal/lb 1 29,540,136
Stewing hens 2,400 lb 3 615 Cal/lb 4 1,476,000
Turkeys 12,000 lb5 541 Cal/lb 1 6,492,000
Rabbits 5,000 lb 6 617 Cal/lb 7 3,085,000
Eggs 360,000 eggs 135 Cal/egg 8 48,600,000
Total 171,368,236

With these very rough estimates, it appears that Polyface actually requires more calories in feed than it produces in food. As I’ve said, these estimates could be off, but I’ve tried to choose them to be favorable to Polyface. For the output, I’ve used energy data from David and Marcia H. Pimentel, which seems to be higher than USDA data for lean meats. (Pastured meat advocates like to say that grass-fed meat is lower in fat than grain-fed meat.) Also, I haven’t even attempted to consider grains fed to the turkeys, the pigs, or the layer hens, but this is likely substantial as well.

If it were the case that there were more energy in the feed grain than in the farm’s output, that would considerably weaken the environmental case for eating meat from a place like Polyface. While I’m still inclined to believe that Polyface meat is far better than factory-farmed meat, Pollan would have us believe that it’s environmentally better to eat Polyface meat than to eat grains because Polyface’s pastures capture solar energy that we wouldn’t otherwise be able to use. That would not make so much sense, though, if Polyface didn’t actually decrease the amount of grain we needed to grow.

One might argue that because Salatin’s cattle don’t eat grain, Polyface grass-fed beef is still relatively low input. But Pollan has made it very clear that each animal has important roles on the farm. The chickens, in particular, are responsible for fertilizing the pasture. Without the chickens and their feed, some other form of fertility would need to be imported from off the farm.

Because I’ve only done a rough calculation, I’m willing to believe that a more careful calculation might show that Polyface actually does produce more calories in food than it requires in grain feed. I welcome any efforts to tear this analysis apart or explain why it is irrelevant. Nonetheless, I think this rough calculation suggests, at least, that the Polyface meal is further from the “proverbially unattainable free lunch” than Pollan has told us. The question remaining is whether it is further from the free lunch than a plant-based meal.


1 David Pimentel and Marcia H. Pimentel. Food, Energy, and Society, Third Edition. Page 70, Table 8.3.
2 Based on 12,000 broilers (as stated in The Omnivore’s Dilemma) with carcass weight of 4 pounds (as stated in Pastured Poultry Profit$).
3 Based on 800 stewing hens (as stated in The Omnivore’s Dilemma) with carcass weight of 3 pounds (as stated in Pastured Poultry Profit$).
4 Unfortunately, I was not able to find a figure for the energy density of stewing hen meat, so I have assumed that it is the same as broiler hen meat.
5 Based on 800 turkeys (as stated in The Omnivore’s Dilemma) with carcass weight of 15 pounds (as stated in Pastured Poultry Profit$).
6 Here, I’ve had significant difficulty finding data. Pollan tells us that Polyface raises 500 rabbits. I’ve assumed a live weight at slaughter of 20 pounds because that seems very large. I’ve also taken into account that dressed weight is typically about 50% of live weight.
7 Based on USDA data for a composite of cuts.
8 This is another somewhat arbitrary number. I’ve chosen this number to be 150% of the USDA figure for a jumbo egg since Polyface eggs are supposedly richer than most.


  1. […] Pollan should choose not to look further. He also doesn’t bother to discuss the question of whether that feed grain might be more efficiently used to feed people directly. Either of these questions would be raised in a more fact-driven work, but there’s simply no […]

  2. pluralist said

    I think your analysis is fine: it just isn’t anything revelatory. My high school science classes taught us about tropic levels (energy flows up the food the chain in very small increments, So what this analysis really shows is that eating meat is more energy intensive than eating grain (and other plant based foods).

    • Adam Merberg said

      I think I’m making a more significant point than that. It’s well known that eating meat is generally less efficient than eating plants. However, Pollan makes the point that at Polyface Farm, the energy that’s wasted is mostly in the form of grass. Thus, he concludes that it doesn’t matter that we’re wasting some of that energy because we wouldn’t be able to use any of it it any other way.

      What I’ve attempted to do here is show that the input of grain to the system is significant, and that raising the animals does waste grain energy, even if you account for all the grass energy that we get “for free.”

      • alex said

        and this doesn’t even consider what could be grown on the land used to graze animals. Salatin says the land was a disaster when he bought it, but it’s hardly “marginal.” It could easily be used to grow bountiful vegan crops. And then there’s the huge man-made lake on his property being used as drinking water for all the animals… you really could write a small book poking holes in Salatin’s version of sustainability.

  3. Chris said

    Hmmmm…I appreciate what you’re doing here, Adam. I like Pollan a lot and his books have influenced…no, they’ve supported conclusions I had already made, but also like to do my own thinking about these issues.

    I would ask you to think about the nutritional differences between grains and meat. Many many people are finding that not only are grains, especially those containing gluten, are not a good food for their bodies. About 30% of Americans are genetically predisposed to gluten intolerance…I’ll dig up some sources for you on that when I have more time if you like. Many very sick people find that by going without gluten/grains, they experience tremendous healing. A lot of people, undiagnosed, but who have chronic pain, fatigue, allergies, arthritis, depression, anxiety, might find that by simply going grain/gluten-free, they could relieve themselves of their health problems.

    Grains contain no nutrients that cannot be found in meat and vegetables. We have no nutritional need for them. Vegans, on the other hand, have difficulty meeting all their nutritional needs and there are zero long-standing traditional cultures that subsist on a 100% vegan diet.

    My point is, that just like the grass that cattle eat is not wasted (because we couldn’t eat it anyway), perhaps the grain fed to chickens is not wasted, or best for human consumption. A diet high in protein from pastured animals, wild caught seafood, eggs from pastured chickens, full-fat raw dairy products from grassfed animals, colorful seasonal vegetables, fermented foods, fruits, and little/no sugar & grains, seems far more healthy to me than one full of hard to digest grains.

    • Adam Merberg said

      I’ve thought about the nutritional differences plenty, but I didn’t write about them because Pollan chose to make an environmental argument for meat-eating. If he had made a nutritional argument, I would have addressed that.

      It’s true that grains and meat are nutritionally different. I’m not sure why you bring gluten into this discussion when oats, corn, and soybeans don’t naturally contain any. One wouldn’t necessarily need to eat grains in place of meat, either. The resources that are used to grow grains to feed chickens could be used to grow various other crops. (The professional ironman triathlet Brendan Brazier promotes a vegan diet which is almost entirely grain-free.)

      We have no nutritional need for meat. It’s true that there are no long-standing traditional vegan cultures, but our lifestyles are without longstanding precedent in many other ways. It should also be mentioned that there are many healthy vegans today.

    • Jess said

      Hi Chris, had to butt in and shoot down some myths you just shat out.


      True nuff. But not all grains are created equal and not all grains contain gluten. Rice (of which there are endless varieties), quinoa (a complete protein), amaranth, millet, sorghum, teff, and buckwheat, just to name a few, do not contain gluten and are normally safe for celiacs.


      That’s funny, I have no problem meeting my nutritional needs at all. In what context are you talking about? If its sustainability, I think you will find that vegetarian farming practices are generally far more sustainable than those producing meat.
      Is there a particular vitamin or amino acid that you assume vegans have “difficulty meeting”? please try to be specific so I can be specific in my rebuttal! = )

      • Adam Merberg said

        Thanks for the support, but do realize you’ve replied to a comment that’s more than a month old. I doubt Chris checks this page regularly. My reply didn’t get any response. I didn’t get around to replying for more than a week. I guess it’s hard for me to be too enthused about answering everybody who comes along spouting Weston A. Price Foundation talking points.

  4. […] like any current meat producers are able to do it yet. “mathematician Adam Merberg performs calculations which suggest that Polyface [a 'humane' farm] requires more calories in feed (for the chickens) […]

  5. […] says that monocultures are bad without mentioning the corn, soy, and oats that feed the chickens and pig on Polyface Farm, his model of polyculture (featured prominently in […]

  6. […] production of pork and poultry tends to include some grain feed, even on small farms. Indeed, my calculations have led me to believe that Polyface Farm (presented in The Omnivore’s Dilemma as a model for […]

  7. Ahimsa said

    Granted, the comment to which I’m replying is old, but there may be latent lurkers… The assertion that all nutrients in grains are also found in meat is false, if one considers fiber a nutrient. I do, since it seems to be an essential input to good health.

    Since we’ve barely scratched the surface of phytonutrients in plants, there may be some undiscovered ones in grains that turn out to be vitally important.

    Of course, a diet consists of more than grains, so there’s no burden on grains to have an identical nutritional profile to meat or to supply all the nutrients one needs in their diet.

    Also, I must second the Jess’ comment: it’s not hard to meet nutritional needs on a moderately well-balanced vegan diet. I consider b12 supplementation to be easy. Of course, many meat-eaters are short on nutrients, including b12, and take supplements.

    • Eric B. said

      Jess, As a farmer, what’s notable to me about your suggested list of grains/pseudo-grains is that it’s a list made up entirely of things that no other small farmer or homesteader I know has found a way to get from field to table in my pretty average American location (North Carolina). Does anyone know of any small-scale, local sources for any of those crops? I’d love to figure out a way to grow several of them and get them to my table, but I think, apart from heavily industrialized ways of doing thing, that animal products (esp. dairy from grass-fed animals) are a lot easier and more efficient than trying to grow grains or oil seeds.

      Adam, I think your critique is very fair, however. I would place a lot of weight, too, on the fact that Salatin’s “beyond organic” meat is not just based on an unsustainable grain transfer, but that he’s buying grain grown with no limits on chemical pesticides or synthetic fertilizers. I think, generally speaking, good use of animals should mean that any potential food used as feed should be more than offset by feeds not suitable for food (e.g. grass), such that potential food spent on animals leads to a net gain of food, not as loss as you demonstrate is the case with Salatin.

  8. […] 9 people for a year. But on his blog Say What Michael Pollan, mathematician Adam Merberg performs calculations which suggest that Polyface requires more calories in feed (for the chickens) than it produces in […]

  9. […] I wasn’t even planning on going. For all my interest in asking Salatin what he thought of my calculations that suggested that his farm was less efficient than feeding grains to people, I didn’t […]

  10. Bob said

    Curious if any of you have read the vegetarian myth? I use to be a vegetarian for 9 years until I bought some land and wanted to farm organically. I went to town to get organic fertilizer and became enlightened to find out that vegans eat animals every time they eat vegetables. The fertilizer is bone meal, blood meal, and best of all animal crap.

    For a vegan to claim that nothing had to die for their meal the vegetables needed to be grown with chemical fertilizers. but we won’t even go into the deaths from petroleum in the gulf.

    So if you are a vegan you better keep your head in the sand because you are eating animals whether you like it or not.

    • Adam Merberg said

      I’ve read The Vegetarian Myth. Lierre Keith raises some good points about our relationship with nature, but her solution is misguided. She writes about the destructiveness of grain agriculture before putting forth Polyface Farm as a solution. The implication of this calculation is that we’d actually have to grow less grain to feed people a grain-based diet. In other words, everything bad she says about grains makes a stronger case for eating grain than for eating Polyface meat.

      While there are farms that don’t use animal inputs, I certainly wouldn’t claim that nothing had to die for my meal. I will, however, point out that the role of animals is not one of making nutrients but of concentrating them. If we look at Polyface Farm, the nutrients in the manure are mainly imported from off the farm in the form of grain fed to the animals.

    • Fireweed said

      Hmmm. I’m a long time vegan and don’t use any animal inputs on the food I grow. It’s another myth to think that we need animals to eat the plants first…grow green manure cover crops and manage crop rotation properly and build your soil that way.

    • Doug said

      There is no need to use those fertilizers at all. Veganic compost and rock dust produce excellent results. No need for any animal products or any chemical fertilizers.

  11. […] summer, I published a post attempting a crude analysis of the inputs and outputs at Polyface Farm. My rough calculations led me to speculate that Polyface Farm may require more calories in grain […]

  12. […] same blog, “Say what, Micheal Pollan?” does calculations of energy inputs on a similar post. I’m pretty impressed. I haven’t seen anything like this before. Very […]

  13. […] 9 people for a year. But on his blog Say What Michael Pollan, mathematician Adam Merberg performs calculations which suggest that Polyface requires more calories in feed (for the chickens) than it produces in […]

  14. […] is “such a good cook”) a Polyface steak, which he’d have us believe is “an awful lot like the proverbially unattainable free lunch,” only to have her wrinkle her nose and push it away. And it wasn’t too long ago that […]

  15. Marie said

    Just to add an interesting point that no one mentioned. Grazing animals is what turned Salatin’s land from a worthless rocky mess to lush fertile land. That is absolutely impossible to do “vegan” style. Grain crops destroy fertility, you can’t continually crop land without removing vast quantities of nutrients, grazing animals improves the soil as well as provides food for us to eat. Adding chemical fertilizer does nothing to improve soil, it actually makes it worse, the only thing that works to improve soil is animals. I don’t know why vegans continue to ignore that fact. And, if they continue to eat vegetable based diets grown on more and more depleted soils they are going to get even more nutrient deficient, leading to chronic disease.

    • Adam Merberg said

      Doesn’t it concern you that Salatin’s farm requires so much grain, though? Those grains are surely coming from the depleted soils that you express concern about. Essentially, it’s not clear that Salatin’s animals are really building soil so much as helping to move fertile land from the grain farm to what was once “a worthless rocky mess.” If that’s the case, the best thing for soil may be to simply use less land agriculturally. On the other hand, maybe that’s not the case. My point in writing is to say that the question deserves to be addressed. Answering it will require actual evidence, not the talking points that you offer or the magical thinking found in Pollan’s writing.

      • Marie said

        His cows eat zero grains, not one seed passes thru those cows from birth to butcher. And that’s most of what he’s improved his pastures with. Leaving land to lie fallow isn’t a good idea either, land needs to be grazed and fertilized, else it just goes to ruin, grows weeds, brambles, brush, etc. Back many years ago there were herds of millions of buffalo to fertilize the land, now we have to mimic that to try to get it to return to the fertility that we had back then. On the grain issue, chicken’s have to eat some grain, and in my opinion, people are going to eat chicken, and so why not feed them healthy pastured meat instead of factory farmed. Salatin is not God, just a farmer trying to do the best he can, and doing a hell of a lot better than most of the others out there. And that’s what I am too, trying to improve my few acres of land the best I can, I accidentally came across this site in my research on pasture improvement. I’ve read a few of Pollan’s books, I think he’s got pretty good ideas, though I’m not really sure what your site is all about, I skimmed it a bit but haven’t really figured out what your goal is, other than trying to discredit what he says. And when your counting calories above, your assuming a calorie is a calorie, the carbohydrate calories coming from grain do not provide near the quality of nutrition than when it is changed in an animal to fat and protein calories. Oh well, I have better things to do than argue health and nutrition and farming, though they are three of the loves of my life 🙂

      • Mountain said

        Salatin’s farm doesn’t require any grain; he uses it for his chickens & pigs because it’s cheap & easy. But chickens & pigs (and other omnivores) can be raised very successfully on nothing more than food waste, weeds, and insects– all things that humans don’t eat. We haven’t worked yet with larger animals like goats or pigs, but we are raising chickens completely free-range (in the actual sense of the term, not the marketing version) on compost consisting of food waste from farmers markets, restaurants, and local schools– all of which would otherwise be headed for landfills. And, of course, they have access to all the weeds they want & all the insects they can catch.

        Point being, Salatin’s approach is much more sustainable than conventional animal farming, but he could much more sustainable. If you want to criticize him for not making more efforts toward sustainability, that’s fine. However, if you want to argue that raising animals is inherently unsustainable or inherently less efficient, and use Polyface as an example, then you’re just mistaken.

      • Adam Merberg said

        This post is about Salatin’s farm, not yours. The point is simply if you want to argue for meat production on environmental grounds, it isn’t clear that Salatin’s farm should be your model. Clearly he improves on conventional animal farming in some ways, but the way Pollan declares it to be a “free lunch” you’d think it’s obviously better than eating a plant-based meal. It isn’t obvious; maybe Polyface is better but it would require an argument with some quantitative sophistication to show that.

  16. I love the implication that because Salatin isn’t running a perfectly closed system, he’s somehow a fraud.

    Land ain’t a fungible commodity. That’s what keeps getting left out of the conversation when people talk about feeding grains to livestock vs people.

    I read “Topsoil and Civilization” and “Dirt: the erosion of civilizations” recently and both books beat this point to death. Farming doesn’t work well in all climates. The great plains supported bison just fine but now that the bison are gone, intensive agriculture is causing massive erosion.

    Sure, what Salatin is doing in some ways involves a zero sum, but in many other ways, he’s increasing the net carrying capacity.

    Great blog, by the way. I’m now a subscriber.

    • Adam Merberg said

      I’m not sure where you find the implication that “because Salatin isn’t running a perfectly closed system, he’s somehow a fraud.” All I claim is that those who argue that a perfectly closed system or producing a “free lunch” are wrong. There may be some instances where Salatin’s marketing is a little deceptive, but I haven’t argued against his style of farming, only the inaccurate information surrounding it.

      • OK, that’s fair. Let’s back up. I think I’m more questioning the notion that the land on Polyface farm + the land where he gets his feed from would necessarily be better used if it were all converted to producing grain.

        I’m throwing out the possibility that maybe agriculture is inherently destructive, no matter how organic you try to be.

      • Adam Merberg said

        I’m not arguing that his farm should be used to grow grain. The point is that he’s using quite a bit of grain, and I’m raising the question of whether it would be environmentally better to simply grow less of that grain, eat it, and leave Salatin’s land as wilderness.

        I agree that agriculture is probably inherently destructive, and the purpose of this post was largely to counter those who held up Polyface Farm as a counterexample.

      • Well, what about eating the wild animals off the wilderness? You’d be able to consume less grain that way.

  17. chefsnausy said

    One of the main points that Pollan makes by using Polyface as an example is that Joel uses fewer man-made “chemical” inputs, thereby creating a less damaging and more “sustainable” polyculture – and that should be viewed as a good thing, right? By the way, as a huge fan of Pollan I’m enjoying your counterpoints and observations – thanks for contributing to this very valuable discussion of our food system!

    • Adam Merberg said

      Thanks for the comment! It’s always nice when Pollan’s supporters stop by with an open mind.

      I don’t mean to say that Salatin’s farm isn’t impressive in some ways. However, I do think when we evaluate it, we need to look at the inputs that are used for growing the feed grain. After all, if we don’t include inputs used to grow feed, then the feedlots look a lot better, too.

  18. Carrie said

    Another thing to consider though, is the satiety of foods; I have switched to a low carb, high fat, moderate protein diet & eat much less calories since doing so. The longevity research done on mice found that were able to follow a calorie restricted diet with a similar diet. We can live off of much less calories than we do, but grains and beans trigger hormone imbalances of insulin and lepit that cause many us to overeat. I probably eat about 3/4 of calories I was consuming now.

  19. L&F said

    My comment concerns only feeding grain to chickens.

    It sounds like Salatin knows something about this practice. From this recent article:

    “In the perfect world, Polyface would not sell eggs. Instead, every kitchen, both domestic and commercial, would have enough chickens proximate to handle all the [food waste] scraps. This would eliminate the entire egg industry and current heavy grain feeding paradigm. At Polyface, we only purport to be doing the best we can do as we struggle through a deviant, historically abnormal food and farming system.”

    Cows are herbivores and not omnivores, so this doesn’t work for them, but it’s still a valid point with chickens.

  20. Afifah said

    Your main point is a good one, (accepting your rough calcs) and it would be better if Polyface farm phased out grain based feed ASAP. But are you suggesting that polyface farm couldn’t function without the grain input?

    In the UK we have many 100,000s acres of upland pasture, completely unsuitable for crops. The beef and lamb raised on it require almost zero feed input. One farm in the West Country I know of has zero input. Theydo t even have tractors, just a quad bike! Their animals are out on grass 24/7/365. Knepp estate in Sussex are allowing rare breeds to wander all over the estate, outdoors all year round, and are sustainably “hunted” like game animals. My local farm Goodwood, grows all of its animal feed, importing none.

    The idea that, environmentally, it would be better to eat the grain directly is based on the dubious assumption that ‘calories per acre’ is the key measure of sustainability. When we swap from pasture to grain we push out the wild plants and animals that occupy that acre – more calories per acre = fewer species per acre!

    Polyface needs to ditch the grains…(and so do we)… We need to eat 100% pastured meat. It exists, and if we eat it we preserve an ecosystem. Eat grains and we destroy a meadow food web.

    Polyface could feed their animals hay instead of grains.
    Most ruminants can be fed on hay. Hay can be sustainably harvested from meadows that have a very high biodiversity index.

  21. […] 9 people for a year. But on his blog Say What Michael Pollan, mathematician Adam Merberg performs calculations which suggest that Polyface requires more calories in feed (for the chickens) than it produces in […]

  22. […] candidate in mathematics at University of California, Berkeley, demonstrates in detail on his blog, is that Polyface Farm isn’t really sustainable, largely due to it’s need to import grain to […]

  23. […] can. And such farms still use resources that could feed hungry humans. Mathematician Adam Merberg published rough calculations suggesting a well-known free-range farm uses more calories in feed than it produces in food; and […]

RSS feed for comments on this post · TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: