Greenhouse gas emissions for common protein sources, renormalized

January 5, 2020

Back in 2011, Environmental Working Group and CleanMetrics released a guide that helps meat eaters understand the climate impact of meat consumption (The Meat Eaters Guide to Climate Change + Health). The report includes CO₂ equivalent estimates for a selection of common protein sources, and a detailed report explaining their methodology.

As a guilty meat-eater, I took to the report with interest, but I wasn’t entirely convinced by the way this study had normalized the CO₂ equivalent estimates. These are presented as kilograms of CO₂ equivalents per kilogram of food consumed (eCO₂/kg). This can lead to some seemingly strange conclusions. For example, eating cheese seems to be much worse for the environment than drinking milk (13.5 vs 1.9 eCO₂/kg). But, cheese is basically concentrated milk. I might drink 300 g of milk during a normal lunch, but I wouldn’t eat 300 g of cheese. It doesn’t quite make sense to me to think in terms of kilograms consumed when actually planning a meal.

So, I normalized the estimates again by protein and calorie content, using data from the USDA’s Food Data Central database. Protein-normalization creates estimates that, in my view, are more relevant to actual life choices (see green series in the chart below). Cheese when normalized by protein content has about the same CO₂ burden as milk. Surprisingly, chicken becomes comparable with vegetable-based protein sources, and is apparently a better option than vegetarian protein staples like milk, cheese or eggs. The most consequential change a person can make is to reduce their consumption of beef and lamb.

I’ve included below an Excel file with the data I used and my calculations. There were admittedly a few somewhat arbitrary decisions I needed to make when selecting the USDA entry that best matched the food surveyed in the meat eater’s guide. For ‘dry beans’, for example, I’ve averaged black-eyed peas (a low protein bean) and kidney beans (higher protein).

Naturally, all this doesn’t take into account all the other important reasons that one might wish to avoid eating meat, particularly fundamental moral objections to killing animals for consumption.

UPDATE (9 February 2020): For a more rigorous analysis of this topic, based on more up-to-date data, see the recent post by Hannah Ritchie:

Ritchie, H. Less meat is nearly always better than sustainable meat, to reduce your carbon footprint. Our World in Data (9 February 2020)

In those results, animal-based protein sources are more consistently worse than vegetable sources. Again reducing consumption of beef and lamb is the biggest positive change that one can make. She also concludes that reducing consumption of meat is always better than just switching to sustainable meat options.

The underlying data are from Poore & Nemecek’s, Reducing food’s environmental impacts through producers and consumers (Science, 2018). These data do not take into account energy used in cooking and meal preparation. This might account for some of the differences relative to the EWG/CleanMetrics values that I used.

See also the recent blog by Adam Hardy, where he uses this issue to illustrate how difficult it can be for even well-informed consumers to make choices aimed at reducing their carbon footprint:

Hardy, A. Go vegan to stop climate change? It should be obvious. Total Carbon Rationing (24 January, 2020).

This work is licensed under a Creative Commons Attribution 4.0 International License.