Having our fake meat and eating it too
Exploring the impacts of fake meats sent me down a deep and cavernous rabbit hole this past month. Loyal readers will know that I preach, perhaps ad nauseum, that reducing animal products in our diets is perhaps the most powerful action we can take to protect the biosphere. The efficacy of reducing meat and dairy seems to be a pretty straightforward conclusion given that producing beef requires 7 to 10 times the land, water, and carbon that it takes to raise an equivalent amount of plant protein and calories. So when I came across the results of a study by the highly respected Lancet Commission on the impacts of vegetarian diets I was rocked. Led by 37 world leading scientists, the EAT-Lancet report asked the question “Can we feed 10 billion people a healthy diet in 2050 and remain within planetary boundaries?” Planetary boundaries are estimates of the scale of perturbations that global life sustaining systems can withstand - e.g. how much carbon can we (relatively) safely emit. Imagine my shock and horror, when I read that the EAT-Lancet study predicts that to feed the global population in 2050 with a nutritious vegetarian diet would use about as much land as a business as usual world, with both futures exceeding the planetary boundary for land use by about 60%. It literally sent me into a tail spin. I spent the better part of two weeks walking around under a cloud, confused and digging for an explanation. Was it perhaps the case that I was wrong to believe that reducing our meat intake could solve a suite of environmental problems?
I finally found a partial explanation. In EAT-Lancet’s business as usual study (BAU), eating patterns from today are extrapolated to 2050. Today, about 800 million people suffer from inadequate nutrition and in the BAU study there will continue to be a large proportion of undernourished people. In the vegetarian 2050 scenario, however, everyone has ample nutrition. So the veggie future has to provide more food than the BAU future, explaining part of the larger than expected land use in the veggie world. But a rough and ready calculations (based on food production using 1/2 the arable land surface today, half of which is currently used for beef, plant proteins requiring 1/7 the land of beef, and 35% more food altogether to feed everyone well in the veggie world) estimates that a vegetarian 2050 world would need 11% less land than is currently farmed, not the 65% more land predicted by the EAT-Lancet study. Even more surprising in the study is that the veggie 2050 world uses as much land to feed its people as a third future scenario - an omnivorous 2050 world. In the omnivorous 2050 world, everyone has ample nutrition and eats 2 and half ounces of meat a day.
That the EAT-Lancet concludes that a veggie 2050 world and a omnivorous 2050 world use the same land makes no sense to me. There are countless estimates of how much land different crops take, and luckily enough for our purposes, Poore and Nemecek have recently compiled most of them into a dataset that includes 38,700 farms representing 90% of global calorie and protein production. They found that even the lowest impact animal products almost always have greater impacts than any plant protein substitutes. And the EAT-Lancet study is well aware of this as shown in their Figure 4, although this figure has a crucial missing entry: how much land is required to grow nuts. The only way I can see to come to the conclusion that BAU, veggie, and omnivorous 2050 diets use the same amount of land is if meat is only chicken and fish (which is given a land use of zero), and veggie protein is only the land hungry veggie protein optins of dairy, eggs, and nuts. More research has convinced me that there is some unexpected assumption in the EAT-Lancet study that differs from those used in other studies or my rough and ready calculation that find that switching to a global vegetarian diet would free up large proportions of land. Take, for instance, this study which finds that if we switched to a global vegetarian diet, the freed up farmland would absorb 332-547 gigatons of carbon dioxide through natural restorative processes. That’s 50-85% of the carbon reduction we need by 2050 to keep warming below 1.5 degrees.
Although I’ve been unable to solve the land use mystery of the EAT-Lancet study, the dive into the topic served to inform the fake meat story, about which there are two very different stories. First is the story promoted by the producers of fake meat that Beyond Beef and the Impossible Burger use something like 1/10th the land and water while only emitting 1/10th of the carbon. This could be massively helpful given that meat production drives at least half of humanity’s land use, upwards of 80% of freshwater withdrawals, and emits enough carbon to alone propel us above 1.5 degrees C of warming. But there is also another story around fake meats: that they are not as environmentally friendly as they appear. I followed the reference train of one of these claims to a paper by Adesogan et al. But remarkably, Adesogan et al. say nothing at all about fake meats. Rather, the paper is a defense of raising livestock, especially for the rural poor, concluding that “recent attention and focus on the negative impact of livestock production skews this discussion.” To support their conclusions, Adesogan et al attack the EAT-Lancet study. Their first complaint is that the EAT-Lancet study ignores undernourished populations. We know from the above that that is incorrect. Both the veggie and the omnivore 2050 diets in the Lancet study are based on 10 billion people having adequate nutrition. The reason Adesogan et al missed this point is because the Lancet paper did not state this upfront. My feeling is that feeding everyone adequately it is so fundamental to the Lancet commission’s mission that they didn’t think it needed stating. Rather, this pops out of the details of the model which took some digging to uncover. And so Adesogan et al, as well I, initially, read the veggie and the BAU scenarios as equivalent.
In further support of their conclusion that the impacts of livestock are over estimated, Adesogan et al also scolded the EAT-Lancet study for ignoring the variability in livestock practices. This one is oddly and obviously incorrect. The Lancet study explicitly states that the variability in the environmental impacts of livestock demonstrates a massive opportunity to reduce negative impacts. Indeed, one of the Lancet’s primary conclusions is that agricultural best practice, alongside halving food waste and dietary change, are all three essential for staying within the safe operating space for the climate, land use, and water use. No one behavioral change can solve all three of these environmental problems. Using Adesogan et al as a reference that fake meats aren’t so environmentally friendly is sloppy and promotes inaccurate views.
Perhaps unsurprisingly, much of the negative press about fake meats comes from the livestock trade. I suppose the livestock industry, as a business, is right to be frightened. In 2020, the sales of plant based meats exceeded a billion US dollars and it is increasing every year. But wouldn’t it be glorious if instead of following the oil and tobacco industries’ play book, the livestock industry instead did the right thing? For fake meats could well be a panacea to drastically reduce our negative impact on the environment.
According to the Impossible Burger company - their burger creates only 11% of the greenhouse gas emissions which a beef burger emits, while using only 4 percent of the land, 13% of the water and 8% of the nutrient runoff that leads to dead zones compared to the outputs of a beef burger. Similarly vastly reduced impacts are reported for Beyond Beef - a pea based fake meat. A note of caution however: these estimates come from studies commissioned by the companies that produce the products. That’s always a little worrisome. Certainly, we know that the agricultural impact of plant based meats must be less than that of animal based meats, merely because for every pound of body weight that a cow puts on, it needs to eat 6-10 times as many pounds of feed. So eating fresh veg undoubtedly has far less of an environmental impact than meat does. But that’s not the whole story for plant based meats. After harvesting - plant based meats require a great deal of processing to get them into meat like form - they go through protein isolation, texturizing processes like kneading and mixing, and flavor adjustments. What emissions are created to make an Impossible Burger, how much river pollution arises from producing a Beyond Beef patty? We have the companies own estimates, but we need independent confirmation of these impacts and for that the recipes of the fake meats need to be made public.
There are other issues about plant based meats that need more attention as well. Firstly, fake meats cost about 40% more than the products they replace. Of course, meat prices are artificially low for various reasons, such as subsidies, and the cost of fake meats are likely to drop as they develop. But for the time being, there is no doubt that many people will not feel comfortable spending an additional 40% to buy a product replacement.
In addition to dramatically reducing our diet’s environmental impact, fake meat could be more healthy than animal meat. Saturated fats can be removed in processing and essential nutrients can be added in. It is also possible that fake meats can reduce our overuse of antibiotics in the livestock trade and reduce the risk of zoonotic disease. But there are nutritional questions, as well. On the surface, macro nutrient scale, plant based meats look amazing. An Impossible Burger has about as much protein, gram for gram, as a beef burger. The beef burger has more calories and fat and less B12 than an Impossible Burger. However, that is not the whole story. Plant based proteins may not be as digestible as animal based ones. Apparently plant proteins are globular in shape, have sheeting on the outside, and are interwound with fibers, all of which makes it harder for our digestive systems to extract the proteins. Indeed, many people find Quorn, for example, upsets their digestive tract. And as a general rule we have found that processed foods are less healthy than whole foods.
My gut feeling is that on the whole plant based, fake meats will have some health issues but will be more healthy than red meat - which of course is linked to enhanced mortality and morbidity issues. And the same will be true of their environmental effects - less bad than raising cattle for slaughter but still having some negative eco impacts. And of course, fake meats alleviates the concerns about the 50 billion animals which are slaughter every year to feed our meat habit.
The clear winner in the protein wars in all three categories - health, environment, and morals - is - drum roll please - none of the above. Rather, we could all do with concentrating on whole fruits and vegetables, and less processed foods. The rush to provide alternatives, be it as electric cars, eco-friendly fashion, or plant based meats, are all capitalist solutions. We simply need to reduce consumption, including processing. And I also feel compelled to comment on how my journey down the fake meat rabbit hole reminded me of how difficult it is to understand just exactly what other people are saying - be it in scientific papers, newspaper articles, or casual encounters with our friends and family. For a dose of goodness, may I encourage you to listen to Pops.