TalkVeganToMe

Download our iOS/Android app now!

Popular Fact

Land required to feed 1 person for 1 year:
Vegan: 1/6th acre
Vegetarian: 3x as much as a vegan
Meat Eater: 18x as much as a vegan
- Cowspiracy

This cites EarthSave.

Reality

We regard the quality of sources on the dietary comparison as poor (see below for why!) and suggest the following instead:

Beef consumed in the United States requires around 160x as much land as potatoes, wheat, or rice for the same number of calories, and nearly 100x as much land for the same amount of protein.
- TalkVeganToMe

Scroll down to the section about ‘Eishel, Gordon, et al.’ for the original source of the the numbers above.

Original Sources

EarthSave

Amount of land needed to feed a pure [vegan] for a year: ⅙ acre
Amount of land needed to feed a meat-eater for a year: 3¼ acres (about 20 times as much)
- EarthSave

Their source for this statistic is ‘Robbins, John. Diet for a New America, StillPoint Publishing, 1987, p. 352’.

John Robbins’ Diet for a new America

To supply one person with a meat habit food for a year requires three and a quarter acres. To supply one lacto-ovo vegetarian with food for a year requires one-half acre. To supply one pure vegetarian requires only one-sixth of an acre. In other words, a given acreage can feed 20 times as many people eating a pure vegetarian diet-style as it could people eating the standard American diet-style Diet for a new American p344 as PDF

This in turn cites - Aaron Altschul - Proteins: Their Chemistry and Politics - 1965 - p264.

Aaron Altschul - Proteins: Their Chemistry and Politics - 1965

Due to its age, this book was quite hard to track down. Fortunately we were able to secure a copy from the British Library.

A meat and milk-based diet may require as much as 3 1⁄2 acres to provide one Standard Nutrition Unit.
[…W]hile six to seven standard Nutrition Units of a rice-and-beans diet in Japan are produced per acre. Therefore, depending upon the type of diet, the capacity of the land to support a population ranges from 20:1.
- Aaron Altschul - Proteins: Their Chemistry and Politics - 1965 - p264.

For a source, this cites ‘Land Use and Food Production - Stamp, L.D, in Hunger: Can It Be Averted? 1961’

Land Use and Food Production - Stamp, L.D, - Hunger: Can It Be Averted? - 1961

Again this book required a trip to the British Library, and even they had to pull it from their offsite location in Yorkshire to bring it to the reading rooms.

Some diets (meat and milk for example) are very extravagant of land; it may require as much as 3.5 acres to produce one [Standard Nutrition Unit (1,000,000 calories)] whereas a mainly wheat-bread basic diet requires under efficient cultivation only 0.25 acre, whilst the very intensive rice culture together with oceans in Japan produces 6-7 Units per acre. In ‘carrying capacity’ there is thus a range from at least 20 to 1.
- Land Usage and Food Production - Stamp, L.D. - Hunger: Can It Be Averted? - 1961 - p38

It is not even vegan because it includes fish (as per the ‘together with oceans’ comment).

While there is no explicit citation, there is a note at the end of the article saying that ‘Stamp, L. Dudley - Our Developing World, Faber, 1960’ treats many of the topics mentioned above in more detail.

Stamp, L. Dudley - Our Developing World, Faber, 1960

On each acre of farmed land - essentially on a dietary base of rice, double or treble cropped where possible, beans, sweet potatoes and other high Calorie foods - Japan can produce 6 or 7 [Standard Nutrition Units] or support that number of people per acre. Under American conditions, and with a meat-milk-fruit diet essentially extravagant of land, it takes some 2 1⁄2 acres or more to produce one SNU or support one person. It [sic] we then wish to stimulate an argument we can then say Japanese agriculture is 15 to 18 times as efficient as American!
- Stamp, L. Dudley - Our Developing World - 1960 - p113

And here we have our original source. This page cites numbers from a table on the next page (p114)

USA: 175 million population, 3.5 acres cultivated land per capita, 0.4 SNU per acre
Japan: 91.5 million population, 0.15 acres cultivated land per capita, 6.5 SNU per acre
- Paraphrased version of table on p114 of Stamp, L. Dudley - Our Developing World

According to the table 0.4 people can be supported by an acre of cultivated land in the USA (i.e. it takes 2.5 acres to support one person) and 6.5 people can be supported per acre in Japan (i.e it takes 1/6th of an acre to support one person)

Although the table does not directly cite its sources elsewhere Dudley uses numbers on total food production from the United Nations’ Food and Agriculture Organisation from 1948-1952 (referenced in tables on p72-p77) and the total cultivated land acreage from the UN FAO’s ‘Yearbook’ from 1957 (referenced in tables on p40-p41).

I believe zir calculation went approximately as follows:

country: total_harvest_calories / 1 SNU (1,000,000 calories) / total_cultivated_acreage = number_of_people_diet_feeds_per_acre
USA: ? (not cited) / 1,000,000 / 192,500,000 (inferred) = 0.4
Japan: ? (not cited) / 1,000,000 / 13,725,000 (inferred) = 6.5
- Paraphrased version of table on p114 of Stamp, L. Dudley - Our Developing World

So the ‘vegan diet’ which takes 1/6th of an acre is arrived at by dividing the total calorie production of Japan immediately post WWII by the number of acres of cultivated land and then further dividing that by the number of calories it takes in a year to support one person.
And the ‘meat diet’ of 3.5 acres seems to have been a rounded up version of the same calculation for the United States.

Conclusion

The 18-20 times land usage statistic specifically is pretty untenable to quote so simplistically today for the following reasons:

  1. Comparing “average diets” is complex due to the huge variety in what people eat globally and even within the same country.
  2. This specific stat originally referred not to an average vegan diet but the average Japanese diet immediately post WWII
  3. The Japanese diet obviously included fish and almost certainly included at least some meat.
  4. Comparing meat and vegetable sources for efficiency needs to be heavily caveated with the regions the statistics were gathered from and the time period those statistics cover as these vary by region and year.
  5. The source for the food production stats are from the period 1948-1950.
  6. The original source does not unpack its exact calculations.
  7. There are much better, much more recent statistics we can draw upon.

We would suggest a much more supportable statement would be

Beef consumed in the United States requires around 160x as much land as potatoes, wheat, or rice for the same number of calories, and nearly 100x as much land for the same amount of protein.
- TalkVeganToMe

This information is based on the other source, ‘Eishel, Gordon et al.’ which you can find the details of below.

Eishel, Gordon, et al.

Compared with the average resource intensities of [potatoes, wheat, and rice] per megacalorie, beef requires 160, 8, 11, and 19 times as much land, irrigation water, [GreenHouse Gas], and [Nitrogen Fertilizer], respectively, whereas the four nonbeef animal categories require on average 6, 0.5, 2, and 3 times as much, respectively. - Eishel, Gordon, et al. “Land, irrigation water, greenhouse gas, reactive nitrogen burdens of meat, eggs and dairy production in the United States”. Proceedings of the National Academy of Sciences. Vol. 111 No. 33 June 2014

In simple terms, the above paragraph states that beef requires 160 times as much land as potatoes, wheat, or rice per 1,000 calories. The paper does not make any claims regarding the overall efficiency of an ‘average’ vegan or vegetarian diet with the variety of plants required to make up a nutritionally complete (not to mention interesting) diet, but the numbers are pretty staggering nonetheless.

Although not stated in so many words, they also provide the same statistics per kg of protein which shows that beef takes almost 100 times (98.2x precisely) as much land as rice, potatoes, or wheat to provide the same amount of protein.

This is calculated by dividing the mean beef kg protein per square metre Eishel Gordon Et Al Supporting Information Sheet > Dataset S1 Excel Spreadsheet > PartitioningPerProtein and dividing it by the average of rice, wheat, and potatoes’ kg protein square metre numbers. You can see in the PartitioningPerCalorie sheet that this is the same calculation performed for their 160x as much land per calorie quote.

Possible Objections

The article also goes on to address the possible objection of the numbers being ‘inflated’ by the cattle grazing land which is otherwise unsuitable for cultivation.

A possible objection to the above conclusion is that beef production partly relies on pastureland in the arid west[ern United States], land that is largely unfit for any other cultivation form. [… T]he objection ignores other societal benefits those arid lands may provide, notably ecosystem services and biodiversity. It further ignores the ≈0.16 million km2 of high-quality cropland used for grazing and the ≈0.46 million km2 of grazing land [..] that can thus be diverted to food production. Even when focusing only on agricultural land, beef still towers over the other categories. This can be seen by excluding pasture resources and summing only crops and processed roughage (mostly hay and silage, whose production claims prime agricultural land that can be hypothetically diverted to other crops). After this exclusion, 1 Mcal of beef still requires ≈15 m2 land (Fig. 2A), about twofold higher than the second least-efficient [animal product] category.

In simple terms, even if you exclude the land the cattle grazes on it uses 15 square metres of land per thousand calories to raise beef (36.5 times as much as potatoes)! Potatoes by comparison use 0.41 square metres per thousand calories - Referenced Supporting Information PDF - S19.

Alternative Presentations

These same statistics can be found in a myriad of different forms, you can find some of them below.

A Well Fed World - Why we need plant based approaches

A Well Fed World in Collaboration with Plantrician Project produced a graphic which states:

[On a] standard American diet: 2 football fields (1.3 acres each) feeds 1 person per year
[On a] plant based diet: 2 football fields feeds 14 people per year

This originates from Plantrician Project - Food Math 101

Plantrician Project - Food Math 101

In order to produce the Standard American Diet (SAD), heavily comprised of animal protein and dairy, it’s estimated that the equivalent of 2 football fields are required per person per year—with much of this allocated to growing the crops to feed the animals. Source: Lappe, Frances Moore. Diet for a Small Planet. 1982, page 69.
Her numbers were 3.25 acres which equals 2.5 football fields; so we used a more conservative two football fields assuming that some agricultural efficiencies have been gained since 1982.
[…]
By contrast, on just one football field of arable land, it’s estimated that food can be produced to feed 7 people for an entire year when they are consuming a predominantly plant-based diet. On the same amount of land that it requires to feed one person the Typical Western Diet (TWD), you can feed 14 people the vegan diet. Source: Diet for a Small Planet, Frances Moore Lappe. 1982, page 69.
Her number was 1⁄6 of an acre for one vegan. Dividing two football fields (2.6 acres by .166 = 16 people. Again, we used the more conservative 14 people on two football fields.
- Plantrician Project - Footnotes

Diet for a Small Planet, Frances Moore Lappe, 1982

Page 69 of Frances Moore Lappe’s 1982 edition for Diet for a Small Planet doesn’t actually contain any assertions about acreage that a vegan diet. The closest it comes is:

16 pounds of grain has twenty-one times more calories and eight times more protein
Diet for a Small Planet, Frances Moore Lappe, 1982, p69

But, here at TalkVeganToMe we are familiar with this page, because it is included as part of the compound citation for 1/6th of an acre in John Robbins’ Diet for a New America when ze states:

To supply one pure vegetarian requires only one-sixth of an acre. …
John Robbins’ - Diet for a new America, p344 as PDF

And when you look up reference 5 for chapter 12 you get:

5 See citations in note 3.

Which takes you to the folowing rather annoying compound citation.

3 Frances Moore Lappé, Diet for a Small Planet, 10th ed. (New York: Ballantine Books, 1982), 69; Aaron Altschul, Proteins: Their Chemistry and Politics (New York: Basic Books, 1965), 264; Folke Doyring, “Soybeans,” Scientific American, February 1974.

The first citation being Frances Moore Lappé, Diet for a Small Planet, but the correct citation for the 1/6th of an acre statistic being Aaron Altschul, Proteins: Their Chemistry and Politics

So we suspect that this an honest mistake on the part of the Plantrician Project; in providing an original source for the statistic they accidentally used the wrong citation from John Robbins’ Diet for a new America. Unfortunately neither source is the original, and the original source does not pertain to a vegan diet at all, instead referring to post-WWII Japan (see Stamp, L. Dudley - Our Developing World, Faber, 1960).

Citations

Article Contributors

Sam Martin
Author