Eating less meat and dairy will almost certainly reduce emissions. What is less certain is where emissions will fall.
Scientists have established beyond reasonable doubt that foods of animal origin have greater environmental impacts than plant-based products. This is based on a couple of key principles. First, all energy enters the Earth system from the sun. Second, animals, unable to photosynthesise, get all their energy via plants. Finally, entropy implies losses in this energy transfer, meaning that sourcing energy from animals takes more resources than from plants.
In the UK, agriculture was responsible for 11% of greenhouse gas emissions in 2023. Animal agriculture contributes the vast majority of this, with 56% being methane emitted from the digestive systems of ruminant livestock – primarily cattle but also sheep, deer and goats. Once manure emissions and emissions from the use of fertilisers on grazing land and animal feed are added to this total, animal agriculture is responsible for the overwhelming majority of agricultural greenhouse gas emissions.
Once the outsized aggregate climate impact of animal-based food was made clear, scientists started drilling deeper into the question. The evidence is overwhelming: whether measured in terms of emissions per portion, per calorie, per gram of protein, or relative to vitamin and mineral content, plant-based food is generally less harmful to the environment than animal alternatives (Figures 1-2).
Figure 1. Greenhouse gas emissions from food production
Source: Doran-Browne et al, 2015
Note: A comparison of GHGE results using the metrics t CO2e/unit nutrient density, t CO2e/t product, t CO2e/ t protein and t CO2e/GJ. The values presented are an average figure for the enterprises (Avg and Top). The foods examined were wheat flour, milk with either 1.0 or 3.5 % fat, canola oil, lamb and beef (L lean, U untrimmed). Canola oil was excluded from the emissions/t protein metric due to canola oil containing no protein.
Figure 2. Environmental and nutritional impact scores of food products
Source: Clark et al 2022
Note: Environmental impact score and nutrition impact score per 100 g of multi-ingredient food products. Data were limited to products available for purchase from Tesco, and were categorised into Aisles using information from Tesco’s website. Colours indicate food types. Points indicate the environmental impact and nutrition impact scores of Aisles not denoted by a text label. When plotting, Aisles containing similar products were condensed for visibility and clarity.
Many spokespeople beyond the longstanding vegetarian and vegan movements have called for action to reduce consumption of animal products, from the late Pope Francis to Beyoncé and Arnold Schwarzenegger. Although successive UK governments have been coy on the subject, measures to reduce meat and dairy consumption have featured prominently in official independent advice from the National Food Strategy and the Climate Change Committee.
The economics of sustainable diets
If animal products have such high greenhouse gas emissions, it might seem redundant to ask whether reducing their consumption will reduce those emissions. But the economy is a complex system. Emissions arise from production, not consumption. The emissions impact of reducing consumption happens only indirectly. To untangle this complex question, economists have joined interdisciplinary efforts to envision how reductions in meat and dairy consumption might play out in the global food economy.
Simulations by the OECD and UN Food and Agriculture Organization (FAO) in 2019 found that a 10% global reduction in ruminant meat consumption could lead to a 15% reduction in agricultural emissions, as well as a 4% reduction in consumer food prices and a 5% reduction in agricultural incomes. Earlier work by Elke Stehfest and others applied global reductions in meat and dairy consumption in an ‘Integrated Assessment Model’ – the class of environmental economic model used in climate policy making. Their scenarios are much more dramatic, with the most moderate being a ‘healthy diet’ scenario where beef consumption falls by 48%, pork by 65% and poultry by 56%. They find this would approximately halve agricultural emissions and reduce the costs required to meet climate targets by 54%.
Modelling by the UK’s Department for Environment, Food and Rural Affairs (DEFRA) implemented very different scenarios and consequently came to very different results. A 2022 paper used the UK Agricultural Market Model to estimate the impact of unilateral (i.e. UK-only) reductions in meat and dairy consumption of between 10% and 50%. The team projected that a 10% reduction in beef consumption would lead to only a 1% reduction in beef production, and a 50% reduction in beef consumption would cause only a 5% reduction in output. This is because – in a country like the UK that is open to international trade – reductions in consumption lead to a large fall in imports and a small increase in exports rather than a change in domestic production. In their simulations, diet change only made substantial reductions to domestic emissions when paired with production-side policies like carbon taxes or land sparing.
What do different visions of the future suggest could happen?
In broad principles, these different modelling exercises come to similar conclusions. Yes, falls in demand for meat and dairy products are likely to reduce prices. Yes, at a global scale, falls in demand for meat and dairy products are likely to reduce production and associated emissions. But the different scenarios highlight a key uncertainty: where in the world these reductions in production and emissions will take place.
The trade components of the models used in these studies would be familiar to anyone with an undergraduate economics education: supply and demand models where elasticities (how much choices flex in response to changes in external conditions) estimated using historic data affect how the model adjusts. The only complication is that both supply and demand are split into home and overseas.
In DEFRA’s UK Agricultural Market Model, for example, UK farmers have very low elasticities of supply (how much they produce is not very flexible to external factors) compared with higher elasticities of supply across global markets. This means increases in demand tend to be met through increases in imports, and decreases in demand tend to reduce imports. This effect is stronger in the short term than the long term, as adjustment in UK supply happens slowly. Other key UK agricultural economic models, such as AfBINI’s FAPRI model, make similar assumptions.
With global climate treaties dividing up responsibility for emissions reductions on a territorial basis, the ‘where’ question has important consequences in climate politics. If reducing meat and dairy consumption reduces imports and thus emissions overseas, it helps the climate but does little to help the UK meet its UNFCCC and Climate Change Act targets (which correspond to territorial emissions). If reducing meat and dairy consumption does in fact reduce agricultural emissions on UK soil, it becomes a potentially important part of delivering UK carbon budgets.
That is the view of the UK’s climate watchdog, the Climate Change Committee, who argue for a 35% reduction in average meat consumption by 2050. Their analysis makes a much simpler assumption: the percentage of UK meat sourced from imports remains constant as consumption falls. This would be more likely under global reductions in consumption than under unilateral reductions, but even uniform global reductions may involve significant turbulence in the pattern of global trade.
Which vision should policy-makers favour?
Should reducing meat and dairy consumption be seen as a selfless, costless contribution to global emissions reduction, with minimal impact on UK agricultural production? Or is it a key means of driving down UK ruminant herd sizes and their associated emissions? The answers to these questions depend on diet trends overseas as well as in the UK.
The UK’s international trade in animal products isn’t as global as it sounds. The European Union (EU) is by far the UK’s main food trade partner, with nine EU countries responsible for two-thirds of UK food, feed and drink imports. For cattle products, the highest carbon food category, EU imports outnumber non-EU imports by ten to one. Of these European imports, the vast majority are from the Republic of Ireland: of the 241 kilotonnes of beef the UK imported in 2024, 185 kilotonnes – more than three-quarters – came from Ireland.
The relatively local nature of the UK’s meat and dairy trade network means ‘multilateral’ diet changes won’t be as tall an order as it sounds. A scenario where both British and Irish consumers reduce beef consumption in parallel would mean British farmers feel more pressure to reduce production, and the UK’s territorial emissions would reduce more than in ‘unilateral’ diet change scenarios. Trade deals with Australia and New Zealand have very slightly globalised the UK’s meat and dairy trade network, but only to countries with relatively high meat consumption where parallel diet trends could well be feasible.
Like with other aspects of the climate challenge, the uncertain location of potential impacts means that the transition to more sustainable diets is best delivered through multilateral co-operation. The good news for negotiators is that the relatively sparse nature of global food trade networks means only smaller coalitions (‘climate clubs’) are needed for agri-climate multilateralism to succeed.
What next for sustainable diets research?
For decision-makers – both politicians and readers deciding what to have for dinner – the message is simple: reduce meat and dairy consumption to confront climate emergency. There is no time to quibble about whether emissions reductions will happen on home soil or overseas.
Food modellers need to move from hypothetical, ‘for the sake of argument’ scenarios – uniform global diets or arbitrary 10%, 20%, 50% reductions – to co-producing viable diet transformation plans. These plans must scope how local food cultures can evolve and transform in co-production with civil society, health practitioners, cooks, nutritionists, social researchers and more.
Plans that are realistic about our changing climate will quickly come unstuck in conventional agricultural economic models based on brittle historic trends and ‘all else held equal’ analysis. With climate impacts affecting yields and conflict upending trade, researchers need models like the Future Food Calculator that take transformation as given.
The data to inform sustainable diets policy and research is still lacking. DEFRA’s decision not to investigate climate policies that rely on consumer behaviour changes means that UK-specific research on sustainable diets is poorly funded and sparse. UK agricultural production statistics include legacy commodities like flax but still do not monitor alternative protein production. Private bodies like the Good Food Institute attempt to fill the gap but end up being another partisan voice in a sensitive debate where independent research is needed to build trust.another partisan voice in a sensitive debate where independent research is needed to build trust.
Research on the environmental impacts of food still relies on averages across a limited number of static studies rather than a standardised, dynamic system of environmental impact monitoring. Without better monitoring, policies that incentivise more sustainable farming practices can only reward actions, rather than improvements in outcomes. This sort of monitoring is possible: government systems to monitor animal diseases can tell you where every single cow in the country has ever travelled. But at the moment the UK is poorly served by 81 private sector farm carbon calculators that produce emissions estimates that can vary by a factor of ten to one.
Where can I find out more?
- Roadmap for Resilience: A UK Food Plan for 2050. Agrifood for Net Zero Network+, 2025.
- Future Food Calculator: an interactive computer model of the UK food system.
- resourcetrade.earth: an online visualisation tool for global trade flows.
