Red meat, methane and the ‘global average’ myth
The newly published US dietary guidelines have landed with a predictable thud. Critics accuse them of climate backsliding for encouraging higher protein intake, including red meat. Meanwhile, supporters of the new guidelines argue that protein density and metabolic health matter more than abstract carbon arithmetic. As usual, both sides reach for headline numbers such as emissions per kilogram, methane per calorie, livestock as a percentage of global warming etc., and end up talking past one another.
Many have also reported on the visual gimmick of inversing the famous food pyramid from the 1980s (heavily influenced at the time by America’s grain growers), but what we see is in many ways, more reminiscent of post-wartime food advice from over 80 years ago. That advice was to eat a broad selection of real food. This matters because what is determined to be the right way to eat in the United States has set the standards for policy all around the world. You might ask why… but that’s for another discussion.
The core problem we address here is less the debate, and more about how it gets framed by the use of misleading metrics.
Why our metrics mislead us
Most climate-nutrition arguments rely on global averages. These averages are tidy and quotable but they’re also almost entirely meaningless. Consider methane from cattle. Globally averaged figures blend together radically different systems: smallholder dairying in India and Pakistan, pasture-based systems in New Zealand and Ireland and highly industrial confinement systems in the United States and Spain. The resulting “average cow” exists nowhere on Earth. The figures describe no real farmer, no real food system and offer no helpful policy levers.
This matters because methane behaves very differently depending on how food is produced, not merely what food is produced.
In low-income countries, cattle numbers are large, milk yields are low and diets rely heavily on crop residues. Methane per litre of milk is high, but industrial inputs are minimal. These systems support livelihoods and food security, not agribusiness infrastructure. Telling them to “eat less red meat” is not a climate policy; it is a development failure.
Get James Capon’s stories in your inbox
Join Medium for free to get updates from this writer.
At the other extreme, high-income countries often produce milk and beef with surprisingly low methane per unit of output. This is the statistic most often cited in defence of industrial livestock. But that efficiency is achieved through dense confinement, imported feed, fertiliser-intensive cropping and large manure handling systems. Methane is diluted per kilogram of product, but the associated industrialisation per kilogram is high.
Global averages collapse these two realities into a single ‘moralised’ number; and in doing so, they reward industrialisation while penalising low-input systems.
Why global averages mislead — As shown in this graphic form, global methane averages flatten three very different realities into one number: how many cattle there are, how much methane they emit per unit of food, and how industrialised the production system is. When these are averaged together, smallholder systems with many animals and low yields (such as in India or Pakistan) are treated as equivalent to highly industrialised, high-yield systems in the US or Spain. The result is a metric that describes no real farming system on Earth, rewards industrialisation by making confinement look “efficient” and obscures where climate leverage actually lies. The bubble chart separates these dimensions, showing that absolute emissions, methane efficiency and industrial infrastructure are distinct problems each requiring distinct solutions.
Nutrition policy cannot be reduced to climate arithmetic alone
The new US guidelines implicitly acknowledge a different truth: nutrition policy cannot be reduced to climate arithmetic alone. Protein quality, affordability and micronutrient density all matter. Red meat remains one of the most bioavailable sources of iron, zinc and B-vitamins, particularly for children, older adults and lower-income households. Pretending otherwise in the name of climate virtue helps no one.
But this does not mean climate concerns should be ignored. It means they should be targeted properly. If the goal is to reduce livestock emissions, the leverage lies in system design: manure management, feed sourcing, fertiliser use, regenerative approaches and stocking density… not in blanket dietary condemnation. A grass-based system in Ireland, a feedlot in Texas and a smallholder cow in Bihar should not be treated as the same problem simply because the cattle all chew the cud.
The real danger of global averages is not that they are wrong, but that they are seductive. They offer what appears to be moral clarity without analytical responsibility. They allow us to argue about food choices while avoiding harder questions about industrial infrastructure. Quoting the organizational scientist Peter Senghe: “Learning to see the structures within which we operate begins a process of freeing ourselves from previously unseen forces and ultimately mastering the ability to work with them and change them”:
If the US dietary guidelines provoke anything useful, it should be this realisation: that climate policy fails when it targets biology instead of systems. We will not eat our way out of climate change with averages. We will only design our way out of it — one production system at a time.
