Is grass-fed beef better for the environment?

The objection

“Well-managed grazing builds soil carbon, and on regenerative farms it locks away more than the cattle emit. Grass-fed beef is part of the climate solution.”

The answer

This objection has more science behind it than most. Grassland soils hold enormous stores of carbon, overgrazing releases it, and careful grazing on degraded land can put some of it back [1]. Grazed and Confused? (2017), a review by 11 researchers led from the University of Oxford’s Food Climate Research Network, set out to test whether that outweighs what the animals emit [1].

Can better grazing make beef carbon-negative?

Better grazing cannot make the world’s grazing livestock carbon-negative, because the most it could store falls well short of what the animals emit. Drawing on the published global estimates, Garnett and colleagues (2017) put the potential of improved grazing management at 295 to 800 million tonnes of CO2-equivalent a year, the upper figure assuming a very ambitious effort, against about 1,320 million tonnes a year from cattle and other ruminants in grazing-only systems, including their supply chains and land-use change [1]. That offsets 20 to 60% of those emissions and 4 to 11% of all livestock emissions, leaving the grazing sector a net emitter even at the maximum [1]. Poore and Nemecek’s data from about 38,700 farms (2018) point the same way: improved pasture management cuts ruminants’ life-cycle emissions by 22% at most, and storing more carbon takes more land [2].

The gains also run out. Soil takes up carbon fastest in the first years after grazing improves, then slows until it reaches a new balance, perhaps after 30 to 70 years and in many places within 20, and a drought, a new owner or a return to poor management can release what was stored [1]. The animals’ methane has no such limit and continues for as long as they are on the land [1].

What about farms that report carbon-negative beef?

Individual farms can report carbon-negative beef, but the result depends on how degraded the land was to begin with and on how the soil carbon is shared out. Rowntree and colleagues (2020) studied a multi-species livestock farm in Georgia, in the US, on land converted from degraded cropland, the kind of soil Garnett’s review expects to gain most [1], and estimated that its soils gained an average of 2.29 tonnes of carbon per hectare a year across fields grazed for up to 20 years [3]. Without that gain, the farm’s poultry, pork and beef together emitted 20.8kg of CO2-equivalent per kilogram of carcass, against 11.9kg for conventional production [3]. With it counted, they came to 4.1kg, 66% less than conventional production, and the beef came out negative, at minus 4.4kg, only when all of the soil carbon was credited to the cattle [3]. The system needed 2.5 times as much land as conventional production, and the soil gains were estimated by comparing fields grazed for different lengths of time rather than by measuring the same fields over the years [3]. General Mills paid for the whole study, which began as data collection for the company’s internal use, and one of the authors was a General Mills employee [3].

Is grass-fed beef lower-carbon than grain-fed beef?

In most life-cycle assessments, which treat soil carbon as stable, grass-fed beef generally has a higher carbon footprint per kilogram than beef from cattle fed grain, because grass-fed cattle grow more slowly [1]. Garnett’s review (2017) reports this for pasture-based systems in general, compared with systems feeding grain and concentrates [1]. In the US, Hayek and Garrett (2018) calculated that producing all the country’s beef from grass-fed cattle would need a national herd of 100 million instead of 77 million and raise methane from the cattle’s digestion by 43%, while existing pasture alone could supply only 27% of current US beef [4]. Worldwide, the Oxford review judged that supplying all meat from grazing would need a massive expansion of grazing land at the cost of forest, and a massive rise in methane [1]. It also called it a mistake to suppose that grass-fed animals use spare land with no other use, since much grazing land could hold forest or wildlife instead and over a quarter of grassland could grow crops [1], which is the weak point in the marginal-land argument.

How does the lowest-impact beef compare with plant protein?

Even the lowest-impact beef emits more than 20 times as much as the average pea crop for the same protein, and needs more land. Poore and Nemecek (2018) gathered data from about 38,700 farms in 119 countries and found that the lowest-impact animal products typically do more damage than the plant proteins that could replace them [2]. Their figures per 100g of protein [2]:

Per 100g of protein	Greenhouse gases (kg CO2e)	Land (m² × years)
Beef from beef herds, average	50	164
Beef from beef herds, 10th percentile	20	42
Beef from dairy herds, 10th percentile	9.1	7.3
Peas, average	0.4	3.4

The 10th percentile is the footprint that only a tenth of the world’s production of that food comes in under, and beef from dairy herds is listed separately because its production is tied to milk [2]. Poore and Nemecek concluded that the environmental cost of ruminants turning grass into human food is immense under any production method in use today [2].