“Plant-based diets reduce emissions”

Evidence across methods

Several study designs reach the same directional finding. A life-cycle meta-analysis consolidated five environmental indicators across 38,700 farms and found large variation among producers, while reporting that even low-impact animal products typically exceeded vegetable substitutes on the assessed impacts (Poore & Nemecek, 2018). A spatially explicit global food-production model estimated that animal-based foods, including livestock feed, accounted for 57% of food-production greenhouse-gas emissions, compared with 29% for plant-based foods and 14% for other uses (Xu et al., 2021). These studies evaluate production systems rather than directly measuring emissions from individual consumers (Poore & Nemecek, 2018; Xu et al., 2021).

Observed dietary data produce a similar pattern when linked to life-cycle inventories. A UK study connected food-frequency data from 55,504 vegans, vegetarians, fish-eaters, and meat-eaters to a review of 570 life-cycle assessments covering more than 38,000 farms in 119 countries (Scarborough et al., 2023). Modelled vegan dietary impacts were 25.1% of high-meat diets for both greenhouse-gas emissions and land use; the corresponding estimates were 46.4% for water use, 27.0% for eutrophication, and 34.3% for potential biodiversity loss (Scarborough et al., 2023). At least 30% differences between low- and high-meat groups were reported for most indicators, supporting the stub's statement that partial reductions can be meaningful (Scarborough et al., 2023).

Why ruminant foods matter

The difference is not explained by transport alone. Animal-based food emissions include enteric methane, manure, feed production, and land-use change, and the global model identified beef as the largest animal-based commodity contributor (Xu et al., 2021). Life-cycle impacts nevertheless vary substantially among producers of the same food, so a product category does not have one universal footprint (Poore & Nemecek, 2018). The evidence supports a larger average effect from replacing high-impact ruminant products than from replacing lower-impact foods, while retaining uncertainty about the exact result of any specific substitution (Poore & Nemecek, 2018; Xu et al., 2021).

Partial and region-specific shifts

A global modelling study compared a 2050 reference scenario with a dietary-guideline pattern, a vegetarian pattern, and a vegan pattern. Across those scenarios, transitions toward more plant-based diets reduced food-related greenhouse-gas emissions by 29–70%, with the size of the effect varying substantially by region and scenario (Springmann et al., 2016). This supports a graded relationship rather than an all-or-nothing threshold: flexitarian or guideline-aligned changes can reduce modelled emissions even when animal-source foods remain in the diet (Springmann et al., 2016).

Nutrition and livelihood context limits global generalization. A multidisciplinary review concluded that the amount and type of animal-source food compatible with health and environmental goals depends on local ecosystems, health priorities, and population needs (Beal et al., 2023). This is particularly relevant where current animal-source food intake is low or micronutrient deficiencies and food access are material constraints (Beal et al., 2023).