By Jeanne Yacoubou, MS

A growing body of research shows that a vegan diet has a substantially lower ecological (or environmental) footprint than a meat-based or even a vegetarian diet. For a recent review, see our article from 2021.

Adding to the evidence in favor of plant-based diets is a 2022 article published in the Proceedings of the National Academy of Sciences by Michael Clark and colleagues. In this blog post, The VRG offers article highlights, pros and cons, and suggested ways you can use this information in your daily life. (Note: For other comments, please see Reed Mangels’ post on this article.)

Estimating the environmental impacts of 57,000 food products: Article Highlights

Here’s a summary of the major conclusions from this research.

• More nutritious foods tend to be more environmentally sustainable, that is, have fewer
• ecological footprints).
• Like-for-like substitutes (for example, meat analogs for meat) can have highly variable environmental and nutritional impacts.
• Research methodology was based on percent composition of ingredients from package labels or estimated from similar products; relied on reputable nutrition and environmental databases to determine impacts.
• Foods with low environmental impacts tend to have water as a major ingredient (sugary drinks) or plant-based ingredients (chips, crackers).
• Foods with high environmental impacts tend to have animal ingredients (beef jerky).
• To test the algorithm’s accuracy, the researchers compared the environmental impact scores generated using the methodology described above to scores estimated when ingredient composition in the same products was not known. The estimated environmental impact score was within 10% of the known score for 66% of products; within 25% for 85%. From this, the authors concluded that their method was sound.
• According to the authors, here is the listing of food products from a major UK retailer ranked with those with the lowest environmental impact to those with the highest:
  • Sugary drinks/beverage with mostly water
  • Vegetables, snacks (e.g., chips), dairy and meat alternatives, some cereal grains, and breads
  • Desserts (e.g., cakes), other cereals and breads, and prepared foods (e.g., pizzas)
  • Nuts, sweet and savory spreads, cheese, fish, and some meats (pork and poultry)
  • Beef and lamb products
• Researchers stated that their analysis revealed that a “…lack of ingredient sourcing information is a potential limitation.” So, to calculate more accurate environmental impact scores, they called for greater transparency from companies about their ingredient sourcing.
• When pairing environmental impact with nutritional impact (using a system called NutriScore), they found that, like their previous research on single-ingredient foods, multi-ingredient healthy foods generally have low environmental impacts compared to less healthy foods with multiple ingredients. Here is the breakdown considering both environmental and nutritional impacts. (Note: low nutritional impact means the food is not nutritious.)
  • Foods with low environmental impact/high nutritional impact: fruits, vegetables, salad, breakfast cereals, some breads, and meat alternatives (e.g., tofu, vegan sausages)
  • Foods with high environmental impact/low nutritional impact: cheese, chocolate, savory pies, and quiches
  • Foods with high environmental impact/high nutritional impact: fish and seafood, nuts, and some prepared meals. They also included beef and lamb in this category, stipulating that beef leads to negative health outcomes in “high-income and high-consuming contexts,” but “…animal-based foods in food-insecure contexts can be integral to nutrition security.”
  • Foods with low nutritional impact/low environmental impact: sweet cakes and pies, sugary drinks, frozen desserts, and table sauces
  • Higher-environmental impact breakfast cereals: granola or those containing chocolate
  • Lower-environmental impact pot pies: “…predominantly vegetarian or vegan.”
• When looking at the claim that “Replacing meat, dairy, and eggs with plant-based alternatives could have large environmental and health benefits in places where consumption of these foods is high,” to see if it were true, they discovered (when considering these examples):
  • Environmental impact of sausages: beef or lamb sausages had on average a 240% higher impact than pork sausages, which had a 100% higher impact than chicken and turkey sausages, which had a 170% higher impact than vegan and vegetarian sausages.
  • Health impact of sausages: beef, lamb, or pork sausages had a 20% higher nutritional impact than chicken and turkey sausages, which in turn had a 75% higher impact than vegan and vegetarian sausages.
  • Environmental impact of pesto: “Nuts were determinants of high environmental impacts, while dairy was a driver of poor nutrition quality.”
  • Environmental impact of lasagna: “…beef lasagna having the highest impact, pork and poultry lasagna having intermediate impacts, and vegetarian and vegan lasagna having the lowest impacts. There were no significant differences in the nutrition impact of different types of lasagna.”
  • Environmental and health impacts of cookies: “…chocolate was a key determinant of both environmental and nutrition, with cookies containing chocolate having, on average, a 13% worse nutrition composition and a 46% higher environmental impact.”
• In considering the effect of changing ingredient sourcing on impacts, (for example, changing country of origin or farming method), the authors said it was “…unlikely to result in meat-based sausages and lasagna having lower environmental impacts than vegetarian and vegan products, but it could result in nut-based pesto and chocolate cookies having lower environmental impacts than their counterparts.”
• The researchers concluded that “…across retail categories, many of the most nutritious food (but not drink) categories are also among the most environmentally sustainable.” This means there “…does not need to be a tradeoff between nutrition and environment.”

Estimating the environmental impacts of 57,000 food products: Pros of Article

Here are some pros of the article:

• Breadth. The article provides a good snapshot of the multitude of food products on the market today by considering all major food categories in the UK and Ireland.
• Depth. By considering four major environmental parameters (greenhouse gas emissions, scarcity-weighted water use, land use, and aquatic eutrophication potential) and nutritional value as determined by NutriScore, the article provides a comprehensive look at the environmental-nutritional impacts of various foods.
• Statistical analysis. The article’s multiple statistical analyses allow the researchers to make trustworthy conclusions about the certainty of their findings.
• Helpful information for people who eat mostly processed food purchased in UK and Irish grocery stores. Unlike most research on this topic, this article considers the environmental impacts of all ingredients in store-bought food products based on their percentage in the product and country of origin (when possible).

Estimating the environmental impacts of 57,000 food products: Cons of Article

Here are some points we hope the authors will consider for future work on this topic. The VRG reached out to the researchers with questions related to these points, but did not yet receive a reply.

• Assumption that personal carbon footprint matters. An idea created by a fossil fuel corporation as a marketing tactic personal carbon footprint shifts responsibility and the spotlight away from companies supplying climate-damaging products and making tax-free, heavily subsidized huge profits from them, and places it markedly and unwittingly on individuals. While personal food choices have different environmental impacts, a focus instead on the current agricultural industry’s environmental impact would make the need for systemic change readily apparent if human civilization is to survive the 21^st
• Narrow focus The article considered food products in the United Kingdom and Ireland only. We’d like to see how applicable these conclusions are to other countries, including the United States and other major carbon emitters (e.g., China).
• Unclear why the researchers used France’s NutriScore. The researchers are based in the UK. NutriScore is based on the British Food Standards Agency’s nutrient profiling system.
• Placed equal weight on the four environmental indicators used in their algorithm. While greenhouse gas emissions, water stress, land use, and eutrophication potential are all relevant parameters when determining environmental impact, it is not true that all are equally important for a given ingredient in all cases. The following series of graphs on milks is a case in point (almond and rice milks). An analysis reflecting this fact would be more realistic.

Restricted number of environmental metrics. Although the list is long of what the authors could have chosen, inclusion of differences in pesticide use and deforestation rates to produce certain foods will enhance knowledge of the specific environmental costs to produce some foods. Although these costs might be embedded in the chosen parameters, delineating it further may influence someone’s decision to purchase certain foods as well as call needed attention to the food category as environmentally damaging in those ways. For instance, since palm oil production necessitates mangrove forest destruction, foods containing ingredients derived from palm oil should receive higher environmental impact scores compared to foods that don’t use palm oil-derived ingredients.

• Did not consider the deep uncertainty inherent in the effects of the climate crisis on the nutritional value of foods or the likelihood of their cultivation. Investigators are just beginning to assess or predict how the climate crisis will impact crops and the nutritional profiles of foods. Also, as soil is degraded by industrialized agriculture, nutritional values of foods are adversely affected. Considering the statistically relevant notion of deep uncertainties in issues related to the climate crisis would likely reduce the confidence levels of the authors’ conclusions.
• It may be cheaper to eat at restaurants than at home. According to a 2022 analysis, food inflation is greater at home than at restaurants. So grocery store information is not helpful. Analyzing data from popular restaurants would be more useful.
• Not helpful for individuals on a whole foods diet. If you are on a whole foods diet, Poore & Nemecek’s 2018 article is more informative. org presents this information, as well as data on food products from Clark, et. al.’s article reviewed here. Helpful tables covering the environmental impacts looking at single metrics (for example, carbon footprint) or composite impacts of four environmental parameters are available.

How You Can Use the Information from Estimating the environmental impacts of 57,000 food products

• If you’d like to consider the impact on the environment that your food choices have, you may use this article as a:
• General guideline when shopping in a grocery store
• Validation for your vegan food choices
• Way to assess the nutritional value you obtain from foods in light of their environmental costs
• Cross check on the criteria of a growing number of ecolabels appearing on food items

The contents of this posting, our website, and our other publications, including Vegetarian Journal and Vegan Journal, are not intended to provide personal medical advice. Medical advice should be obtained from a qualified health professional. We often depend on product and ingredient information from company statements. It is impossible to be 100% sure about a statement, info can change, people have different views, and mistakes can be made. Please use your best judgment about whether a product is suitable for you. To be sure, do further research or confirmation on your own