Protein myths debunked

FACT: Vegetables have plenty of protein, and they're complete proteins as well

by Michael Bluejay • Last update: November 2024

We need only 2.5 to 11% of our calories as protein, according to peer-reviewed research and the official recommendations.^2,3,4  That amount is easily supplied by common vegetables.⁵  Vegetables average around 22% protein by calorie, beans 28%, and grains 13%,⁵ as shown in the chart.

The U.S. government's recommendation is 5-11%, based on various factors.³  The World Health Organization (WHO) recommends a similar amount.⁴  And these recommendations are padded with generous safety margins, to cover people who need more protein than average.  The National Institutes of Health and WHO's recs are intended to cover 97-98% of the population.  That means the overwhelming majority of people need less than the official recs, even though the official recs are still easily supplied with plant foods.  The median person needs only 80% of the official rec.⁴

In any event, whether you think our needs are closer to 2.5% or 11%, you can see from the chart that it's nearly impossible to fail to get enough protein, provided that you make sure to eat food.  Every single whole plant food has more than 2.5% protein, and every group averages at least 11% except for fruit.  Protein is one of the easiest nutrients to get.

The figures for food are from the bible of nutrition data, the USDA FoodData Central database.⁵  The science is exceptionally clear, for anyone willing to look.  (Unfortunately, many doctors and dietitians haven't looked.)

So plant foods easily supply our protein needs.  The truth is that if you're eating food, you're eating protein—and almost certainly more than enough.

It's meaningless to talk about a "source of protein", since all whole foods have plentiful protein.  In other words, every whole food is a "source of protein".  You don't have to eat certain, special foods to get protein. You just have to eat any whole food.  That's it.

This isn't just theoretical, either.  A massive review of 64 studies over twenty years concluded that "protein intake...in people following plant-based diets [was] well within recommended intake levels."  Vegan diets averaged 13% protein (about double the typical need), compared to 16% for meat-eaters. (Nutrients 2022)  So, that's what the science says.  It's probably not what you'll hear from your doctor or dietitian because, again, many health professionals haven't looked up what the science actually says.

Of course, many health professionals have looked up the science, and those who do say the same things you're reading in this article.  Take Marion Nestle, Ph.D, chair of the Department of Nutrition at New York University:

"We never talk about protein anymore, because it's absolutely not an issue, even among children.  If anything, we talk about the dangers of high-protein diets. Getting enough is simply a matter of getting enough calories."⁹

It's true that meat has more protein than vegetables, but the amount in vegetables is already much more than you need.  The extra protein in meat isn't better, it's useless.  If you're shopping for a car and one goes 200 miles an hour and the other goes 300, it doesn't matter, since the maximum speed limit anywhere in the U.S. is 85 mph.  Two hundred mph is more than enough for a car, and 22% protein from vegetables is more than enough when your protein needs are only 2.5 to 11%.

Oh, but you've heard that plant protein is "incomplete", right?  Well, that's not true either.  Let's have a look....

Vegetables are complete proteins

We've all heard that plant protein is "incomplete" compared to meat protein, and that plant foods have to be carefully combined to make a "complete" protein.  But that's an urban legend that was never based on science.  The Academy of Nutrition and Dietetics (AND) abandoned that idea decades ago.  Susan Havala Hobbs, Ph.D, R.D. describes how the AND discarded the protein combining myth:

There was no basis for [protein combining] that I could see.... I began calling around and talking to people and asking them what the justification was for saying that you had to complement proteins, and there was none.  And what I got instead was some interesting insight from people who were knowledgeable and actually felt that there was probably no need to complement proteins. So we went ahead and made that change in the paper.  [The paper was approved by peer review and by a delegation vote before becoming official.]  And it was a couple of years after that that Vernon Young and Peter Pellet published their paper that became the definitive contemporary guide to protein metabolism in humans.  And it also confirmed that complementing proteins at meals was totally unnecessary.¹⁶

There's a very easy way to see the completeness of plant proteins, that almost no nutrition writers have bothered to do:  Look at what's actually in the food!  It's not like this is a secret; that data has been publicly available from the USDA for decades, and now the USDA's database is even online.⁵  Below is what it looks like when you actually look up the numbers.

Vegetables are complete proteins
Amino acid >>	Isoleu- cine	Leucine	Lysine	Phenylalaline +Tyrosine	Methionine	Threonine	Tryptophan	Valine	Histidine
Human Need
Brown Rice
Tomatoes
Potatoes
Green Peppers
Corn
Lettuce (iceberg)
Celery
Cucumbers
Oats
Carrots
Broccoli
Pinto Beans

Amino acid need from the World Health Organization⁴, food composition from the USDA nutrient database⁵.
Analysis is for each individual food all supplying calorie needs (closest to the "low active" category for a 5'11" 181lb. 25BMI male, as per the FDA).³

So when we compare the actual requirements to what plant foods actually contain, we find that basic plant foods aren't incomplete at all.  They have every essential amino acid, in excess of what we need.  It might not surprise you that beans are a complete protein by themselves, but even carrots are a complete protein.  Tomatoes are a complete protein.  Celery is a complete protein.  Even iceberg lettuce is a complete protein.

(Those who would object that we can't eat enough lettuce to satisfy our protein needs are wildly missing the point.  The point of using a day's worth of calories for a single food is simply to show mathematically how the food measures up, not to suggest that anyone could or should eat only a single food.  These plant foods are complete no matter how much or how little of them you eat.  That is, if only 1% of your diet is lettuce, then lettuce supplies more than 1% of your protein and amino acid requirements.)

A 2015 study analyzed the dietary intake and blood amino acid levels of various groups, and found that vegans met the RDA for each and every amino acid. (Jack Norris, R.D.)

Interestingly, the amounts for "Need" in the table are twice what they were until recently. The original recommendations in the WHO's 1973 and 1985 reports were based on William Rose's pioneering work in the 1950's, and were considerably lower.¹³  Rose determined the levels needed by his subjects by intentionally feeding them diets with a synthetic mixture of declining levels of amino acids until they became deficient.  After finding the highest amount needed by any subject, he then doubled that figure to arrive at his recommendation.¹⁴  And the current WHO recommendations have doubled their earlier figures again.  And even with all these increases, individual plants still measure up as fully complete.

Experts confirm that plant proteins are complete

Besides the AND, other medical and nutrition professionals who have actually looked at the science know that there's no need to carefully combine proteins.  For example:

Dennis Gordon, M.Ed, R.D.:

[C]omplementing proteins is not necessary with vegetable proteins. The myth that vegetable source proteins need to be complemented is similar to the myths that persist about sugar making one's blood glucose go up faster than starch does. These myths have great staying power despite their being no evidence to support them and plenty to refute them.¹⁵

Jeff Novick, M.S., R.D.:

Recently, I was teaching a nutrition class and describing the adequacy of plant-based diets to meet human nutritional needs.  A woman raised her hand and stated, "I've read that because plant foods don't contain all the essential amino acids that humans need, to be healthy we must either eat animal protein or combine certain plant foods with others in order to ensure that we get complete proteins."

I was a little surprised to hear this, since this is one of the oldest myths related to vegetarianism and was disproved long ago.  When I pointed this out, the woman identified herself as a medical resident and stated that her current textbook in human physiology states this and that in her classes, her professors have emphasized this point.

I was shocked.  If myths like this not only abound in the general population, but also in the medical community, how can anyone ever learn how to eat healthfully?  It is important to correct this misinformation because many people are afraid to follow healthful, plant-based, and/or total vegetarian (vegan) diets because they worry about "incomplete proteins" from plant sources. ...[I]f you calculate the amount of each essential amino acid provided by unprocessed plant foods...you will find that any single one...of these whole natural plant foods provides all of the essential amino acids....

Modern researchers know that it is virtually impossible to design a calorie-sufficient diet based on unprocessed whole natural plant foods that is deficient in any of the amino acids.  (The only possible exception could be a diet based solely on fruit.)¹⁷

John A. McDougall, M.D.:

Many people believe than animal foods contain protein that is superior in quality to the protein found in plants.  This is a misconception dating back to 1914, when Osborn and Mendel studied the protein requirements of laboratory rats.[11]... Based on these early rat experiments the amino acid pattern found in animal products was declared to be the standard by which to compare the amino acid pattern of vegetable foods.  According to this concept, wheat and rice were declared deficient in lysine, and corn was deficient in tryptophan.  It has since been shown that the initial premise that animal products supplied the most ideal protein pattern for humans, as it did for rats, was incorrect.... It is clear that even single vegetable foods contain more than enough of all amino acids essential for humans.... Furthermore, many investigators have found no improvement by mixing plant foods or supplementing them with amino acid mixtures to make the combined amino acid pattern look more like that of flesh, milk, or eggs.[35-44]... Nature has designed vegetable foods to be complete.  If people living before the age of modern dietetics had had to worry about achieving the correct protein combinations in their diets, our species would not have survived for these millions of years.¹⁰  A careful look at the founding scientific research and some simple math prove it is impossible to design an amino acid–deficient diet based on the amounts of unprocessed starches and vegetables sufficient to meet the calorie needs of humans. Furthermore, mixing foods to make a complementary amino acid composition is unnecessary.³⁴

Andrew Weil, M.D.:

You may have heard that vegetable sources of protein are "incomplete" and become "complete" only when correctly combined.  Research has discredited that notion so you don't have to worry that you won't get enough usable protein if you don't put together some magical combination of foods at each meal.¹⁸

Charles Attwood, M.D.:

Beans, however, are rich sources of all essential amino acids.  The old ideas about the necessity of carefully combining vegetables at every meal to ensure the supply of essential amino acids has been totally refuted.¹⁹

The original source of the protein combining myth recants!

Interestingly, it's very easy to trace the protein combining myth to its original source: A bestselling book called Diet for a Small Planet, in 1971.  The author, Frances Moore-Lappé, wanted to promote meatless eating because meat production wastes horrific amounts of resources.  But she knew her readers would think you couldn't get enough protein from a vegetarian diet, so she set out to reassure them, by telling them that if they carefully combined various plant foods, like rice and beans, the supposedly inferior plant proteins would become just as "complete" as the ones in meat.

Lappé got her idea from studies that were done over 100 years ago, on rats.  The researchers found that rats grew best when the proteins in their diets were in the same proportions as found in animal foods.  From this finding, animal proteins were arbitrarily labeled first-class while plant proteins were deemed inferior.  The problem with this conclusion is that rats are not simply smaller versions of people.  Baby rats actually need a higher percentage of protein than do baby humans, because they grow a lot faster.  Humans grow slowly.  It takes a human baby half a year to double its birth weight.  A rat does it in only four and a half days.¹¹  So clearly rats are going to need more protein.  In fact, rat milk is a whopping 49% protein¹²—much higher than the mere 6% found in human mother's milk.

Lappé's idea of protein combining spread like wildfire.  Soon the National Research Council and the AND, without bothering to verify the hypothesis, jumped on the bandwagon by saying that plant proteins were inferior and had to be combined.¹⁰

But it wasn't long before Lappé realized her mistake, and owned up to it.  In the 1981 edition of Diet for a Small Planet, she recanted:

In 1971 I stressed protein complementarity because I assumed that the only way to get enough protein...was to create a protein as usable by the body as animal protein.  In combating the myth that meat is the only way to get high-quality protein, I reinforced another myth.  I gave the impression that in order to get enough protein without meat, considerable care was needed in choosing foods.  Actually, it is much easier than I thought.

With three important exceptions, there is little danger of protein deficiency in a plant food diet.  The exceptions are diets very heavily dependent on [1] fruit or on [2] some tubers, such as sweet potatoes or cassava, or on [3] junk food (refined flours, sugars, and fat). Fortunately, relatively few people in the world try to survive on diets in which these foods are virtually the sole source of calories.  In all other diets, if people are getting enough calories, they are virtually certain of getting enough protein."²⁰[emphasis in original]

It says a lot about Moore-Lappé's integrity that she owned up to the very mistake that made her a household name.  (It's also worth noting that her book pretty much single-handedly jump-started the vegetarian movement in the U.S. in the 1970s.)

In any event, if you came to this page with the idea in your head that plant proteins have to be combined, I hope it means something to you that the person responsible for that idea being in your head in the first place said that she was wrong.

What's truly ironic is that everyone who has the mistaken idea about protein combining got it from Moore-Lappé, directly or indirectly, but she took it back.

What's really crazy is how many people cling to the protein-combining myth even after learning that the very source of that myth admitted that it's not true.  If people believed her when she made the original claim, then why won't they believe her when she said her claim was wrong?

Digestibility is not a problem

Some critics have screamed at me that plant protein isn't digested as well as animal protein.  Well, say it, don't spray it.  Once again, these critics haven't bothered to look up the numbers.

The protein in beef and fish in 94% digestible.  That's actually less than the digestibility than plant foods like white flour (96%) and peanut butter (95%).  Peas, rice, whole corn, soy flour, oatmeal, and whole wheat flour aren't far behind (86-88%).  Beans, despite their high protein content, are a bit further down on the digestibility scale (78%).³  (By the way, the WHO report didn't list other vegetables, or I would have listed them here.)

This shows that digestibility isn't a problem at all, in practical terms.  Plant foods still provide more than enough protein, even after considering lower digestibility.  From the numbers above, the protein in meat is digested 20.5% better than that of beans.  If we take someone with a higher than average need for protein (10% of calories), and add 20.5% to that figure to account for lower digestibility, they now need 12.5% protein instead of 10%.  And again, grains average 13% protein and vegetables average 22%—far more than enough.

Protein quality is not a problem

Some critics have pointed to various measures of protein quality, such as PDCAAS, which say that plant protein is inferior.  Such critics are missing the obvious:  The quality measures are mostly based on the amounts of amino acids in foods,²¹ and I've already explained in detail, with a nice chart, using numbers from official sources, that vegetables absolutely contain as much or more than you need of each individual amino acid.  That is, plant foods provide more than enough protein even after you account for any differences in digestion or protein quality.  Your body doesn't care whether the protein quality of what you're eating is "very high" vs. simply "high".  It's concerned only that you eat enough.  As long as your body is getting as much protein as it actually needs, it doesn't matter what form the protein comes in.

Critics are confusing more with better.  Yes, animal foods have more protein, but that's not a benefit.  There's absolutely no advantage to eating way more protein than your body can use.  If you need 2500 calories a day, would you be healthier with 3000 calories a day?  No.  In fact, eating that much more than you need would be detrimental to your health.  The same is true of eating too much protein.  Excess protein intake has been linked to bone loss, osteoporosis, kidney damage, kidney stones, immune dysfunction, arthritis, cancer promotion, low-energy, and overall poor health.²²  The science on this is very clear.

Vegan diets supply plenty of protein for building muscle

Vegan bodybuilders shatter the myth that vegans are skinny and malnourished.  (Pictured: Avi Lehyani, anonymous, Ryan Wilson, Robert Cheeke)

Plant foods supply plenty of protein even for athletes and those trying to build muscle.  In a study of older adults doing either lower-body or whole-body resistance training increased their muscle strength and mass on the US RDA for protein of only 0.36 g per lb. of body weight.²⁵  For a 120-lb. person eating 2000 calories or a 180-lb. person eating 2500 calories, that's 8.6% to 10.4% of their diets as protein.  And remember, vegetables average 22% protein and beans 28%.

Another study suggested that established bodybuilders need around 0.48 g of protein per pound of body weight per day (1.05 g/kg).²⁶  (Incidentally, it also found that bodybuilders required 1.12 times and endurance athletes required 1.67 times more daily protein than sedentary controls.)  For an 180-lb. athlete the 0.48 g/lb. figure is 90 grams (360 calories from protein).  For a 3000-calorie diet, that's 12% of calories from protein. And again, vegetables average 22% and beans 28%.

Those starting a muscle-building program may need more protein, 0.77 g/lb. (1.7 g/kg).²⁷  For a 180-lb. athlete that's 139 grams (556 calories).  On a 3000-calorie diet, that's 18.5%, still less than supplied by common vegetables.

If the athlete eats more than 3000 calories a day, or weighs less than 180 lbs., then the percentage of protein required goes down even more.

In 2009 three major health organizations endorsed the 0.5 to 0.8 g/lb. (1.2-1.7 g/kg) figures above (American Dietetic Association, Dietitians of Canada and the American College of Sports Medicine)²⁸

More is not better. As one paper said, "Ingesting more protein than necessary to maintain protein balance during training (e.g., >1.8 g/kg/d) does not promote greater gains in strength or fat-free mass."²⁹.  One study of 60 elderly men doing power training showed that supplementing with whey or casein didn't lead to muscle improvement versus no supplementation.³²

Jack Norris, RD points out that nutrient recommendations are always "padded" with safety margins. That is, most people need less:

Considering the information reviewed above...it seems reasonable to conclude that the protein needs of most vegan bodybuilders are somewhere between 0.8 and 1.5 g/kg (0.36 and 0.68 g/lb) of body weight....

The Food and Nutrition Board, which sets the RDA, reviewed Lemon et al.'s study and others and concluded there is no sufficient evidence to support that resistance training increases the protein RDA of 0.80 g/kg [0.36 g/lb] for healthy adults.³⁰

For more on protein and muscle-building, see my separate article on Protein & Strength.

Is it possible to be protein deficient?

Experts know (and have said) that it's essentially impossible to design a diet based on whole foods that's deficient in protein.  But there are two cases where it's possible to be deficient.

The first is a diet that's not based on whole foods, meaning lots of processed foods.  Foods (or drinks) with lots of fat or sugar have lots of calories but less protein.  This doesn't mean that every diet that includes processed foods is protein-deficient, but it's possible.  It's much more likely that a diet high in processed foods would be deficient in other nutrients.

The second is a diet that doesn't provide enough calories could be protein-deficient, but not necessarily.  For example, if someone reduced their calories by a whopping 20%, they'd need 25% more protein as a percentage of calories.³³  Though if someone normally needing 8% protein cut their calories and needed 25% more, then they'd need 10% protein, which is still less than the 13% in grains, 22% in vegetables, etc.

Other objections

I get lots of misinformed objections to this article, but some of it is really wacky.  One particular objection is that my use of recommendations from the World Health Organization (WHO) is wrong, because the WHO's recommendations are supposedly designed only to prevent extreme malnourishment among impoverished third-world residents.  Such critics have apparently never actually read a WHO report, since such reports say the exact opposite.  For example:

• "The levels of energy intake recommended by this expert consultation are based on estimates of requirements of healthy, well-nourished individuals."³¹ (emphasis in original)
• "[T]he objective of this report is to make recommendations for healthy, well-nourished populations..."³¹

It's really meat that's incomplete

It's kind of silly to single out protein, just one of the many nutrients, just so we can declare plant proteins to be incomplete (although they're not).  Why aren't we declaring meat to be an incomplete vitamin?  Because it is, you know.  For example, beef is completely devoid of Vitamin C, an essential nutrient without which you'd die.  And beef doesn't just have a lower level of this essential nutrient, it has zero. So why didn't the authorities ever caution us that we need to combine various foods to get a complete vitamin?

But actually, no combination of meat will make a complete vitamin, since every single kind of common meat has zero Vitamin C.  And it's deficient in other vitamins as well.  So while plants aren't actually deficient in protein, meat is definitely deficient in vitamins.  But I'm sure you never heard about vitamin deficiency in meat.  All you've heard about is the supposed deficiency of protein in plants.

And speaking about biases, the whole protein-combining idea supposes that vegetarians are eating just one food, which is allegedly incomplete.  Okay, how many people do you know who eat one food?  And since nobody eats just one food, the whole idea of protein combining would be unnecessary anyway, even if it were true.  So here again, what would be the point of harping on protein combining when it doesn't matter?

Using some common sense

The largest land animals in the world, elephants, are exclusively vegetarian.  They grow up to 10,000 pounds, by eating nothing but plant matter.  They couldn't grow so big if plants weren't loaded with protein.

Amazingly, many readers have protested this by saying "But we're not elephants!", as though they've made some sort of point.  If they mean to suggest that elephants don't need protein, they're wrong:  Every living creature on the planet does.  Elephants don't have some magical superpower which allows them to live and grow without eating protein.  They need it, eat it, and use it, like everyone and everything else.

Perhaps the point was supposed to be that elephants utilize protein differently?  Not in any meaningful way.  All protein, whether plant or animal, is broken down into the individual amino acids before the body uses it.  And that goes for any body, elephant, human, or otherwise.

Maybe the idea was that elephants get enough protein from plants only because they eat so much?  No, because once you adjust for body weight, elephants eat less than we do.  Per 100 lbs. of body weight, Americans eat about 3 lbs. of food per day, while elephants eat only 1.9. ²³

And elephants aren't the only huge vegetarian animals roaming the planet.  There are also horses, camels, giraffes, elk, rhinos, cattle, and more.  Clearly if these massive animals are eating only plants, then plants have more than sufficient protein.

Ignoring common sense

We've seen what answers we get when we use credible resources and common sense.  What happens if we just draw ridiculous conclusions?

I studied nutrition at the University of Texas at Austin, one of the largest universities in the country. They taught us that vegetarian or natural-foods diets were only for the stupid or naïve.  I remember in particular one page from our textbook, which had three pie charts, each showing the proportion of macronutrients in vegetables, meat, and a human being, respectively.  And by golly, the charts for the meat and for the human being were nearly identical!  The implication was that meat is an ideal food, because it matches our body type so well.

It was at that point I realized I paid too much for my college education.  Of course the charts were the same, because human beings are simply walking, talking meat.  A body is a body, after all.  By the logic presented in the textbook, if someone is white then they should eat white food.

And what about all the other animals?  Elephants, giraffes, and horses would all have a pie chart that looks exactly like the one in the book.  Shouldn't they be eating meat, too?  Stupid elephants!  If only they'd gone to college.  You know, I really should go find a horse and try to get her to eat some meat.  If she's not interested, I should show her the textbook so she can see that beef has the same makeup of her own body type.  If that doesn't convince her, I should at least implore her to combine her proteins to make a complete protein.

Changing our vocabulary

Now that we know the truth about protein, it's time to start changing the way we talk about it.  Here are some common ways in which the word protein is misused.

"Are beans a good source of protein?"

This question misses the point.  When all whole foods have plentiful protein, it's meaningless to talk about a "source" of it.  Food is a good source of protein.  We don't ask whether certain foods are good sources of calories, because all foods have calories.

"I try not to eat a protein until lunchtime."

That's pretty much impossible, unless this person is not eating at all, since all whole foods contain protein, and they generally contain more than what you need.

There is no food that is "a protein".  Foods contain various nutrients, and protein is just one of them.  It's disparaging to food to identify it by just one of the nutrients it contains.  Would you call peanut butter a protein?  Because it's actually mostly fat, much more fat than protein.  Would you then call it a fat?  Why so, since it has more protein than you need?  We have to stop equating foods with individual nutrients.  Foods contain nutrients, they contain a collection of nutrients (not just one), and in any event, all whole foods contain protein.

"Try to eat three servings of lean protein."

There is no such thing as "lean protein". There is only food, all of which contains protein.

We know that all food has protein, so something called "lean protein" should at least be low in fat, right?  Think again.  The National Institutes of Health says "lean protein" can have up to 3 grams of fat per 55 calories.²⁴ That's up to 49% fat!  Forty-nine percent is lean to these people?!  Lean would be something like 10%.  Half-fat would not.

So "lean protein" is wrong on two counts.  First it assumes that only certain foods have protein, or that a food can be "a protein" (rather than being food), and second, "lean protein" isn't even lean.

Comparing infants' protein needs with adults'

Human breast milk is a mere 5.9% protein, when our protein needs are highest, because we're growing faster than we ever will again.  We might therefore conclude that we'll need less than 5.9% protein in adulthood.  However, as The Vegan RD points out, a comparison to babies on a percentage-of-calories basis doesn't work, because babies consume lots more calories than adults, relative to their body size.  A six-month old baby needs 37-49 calories per lb. of body weight per day, which would be like a whopping 5600-7400 calories/day for a 150-lb. person.  Adjusting for a baby's voracious appetite, a baby's protein intake would look more like 13% of calories.*  So it's not right to conclude that we need less than 5.9% protein as adults.  But we could certainly conclude that we need less than 13% of calories as protein, and that's in fact what the official sources say.  Even so, common vegetables average more than 13% protein, even after adjusting for bioavailability.

*NIH says that a six-month-old weighs about 16.7 lbs. and needs 9.1g/d of protein.  That would be 85.6g/d for a 150-lb. person.  For a 2650 calorie/day diet, that would be (85.6 x 4) ÷ 2650 = 12.9% of calories as protein.

Related

• Dr. McDougall on protein
  
• The Milk Letter: A message to my patients, by Robert M. Kradjian, M.D.

²⁸ Nutrition and Athletic Performance, joint position of the American Dietetic Association, Dietitians of Canada and the American College of Sports Medicine (March 2009)
²⁹ Effects of protein and amino acid supplementation on athletic performance, Kreider RB, Sportscience 3(1), 1999
³⁰ Vegan Weightlifting: What does the science say?, Jack Norris, RD, Vegetarian Journal (2003, Issue 4)
³¹ Human Energy Requirements, Report of a Joint FAO/WHO/UNU Expert Consultation, Chapters 2 & 4, October 2001
³² Effects of slow versus fast-digested protein supplementation combined with mixed power training on muscle function and functional capacities in older men, M C Dulac et al, British Journal of Nutrition (2020 Jun 5)
³³ If you wondered why both figures aren't 20%, it depends on which direction you're looking towards.  For example, if two items cost $80 and $100, then the $80 item costs 20% less than the $100 item (20÷100), but the $100 item costs 25% more than the $80 item (20÷80).  For the example in which someone reduces calories by 20% but needs 25% more protein as a percentage of calories: 2500 kcal > 2000 kcal (20% fewer calories) • 8% protein of 2500 kcal = 200 kcal • 200 kcal ÷ 2000 kcal = 10% • 10%÷8% = 25%.
³⁴ Plant Foods Have a Complete Amino Acid Composition,John McDougal, M.D., Circulation, 25 un 2002

While I didn't cite this in the article, in How Much Protein is Needed? (PDF), Professor T. Colin Campbell agrees (p. 18) that looking at the percentage of calories from protein is preferable to looking at the number of grams, because grams will be different for different genders while percentages will be the same.  He also shows that the U.S. government recommendation for protein intake works out to about 9% of calories.