Knowledge for better food systems

FCRN Blogs

Why the beef with UK livestock? The need to distinguish between local and global scales in discussions on food sustainability

April 6, 2020
Ellen Fletcher

Ellen Fletcher holds a BSc and MSc from Imperial College London, Centre for Environmental Policy. Her research has previously focused on urban agriculture (UA) where she has provided evidence on the importance of allotments as a form of UA in sustainable cities. Ellen is currently a research assistant at Rothamsted Research, where she is part of an EU Horizon 2020 project working with farmers in the South West. Ellen would like to pay particular thanks to Catherine Broomfield for initially introducing her to the topic of global and local food systems in her seminar “Livestock farmers: prophets or pariahs?”. Catherine’s extensive body of work on UK livestock systems can be found here.

Image credits: Ellen Fletcher

Food consumption is a topic under increasing scrutiny with almost all emerging publications on the climate crisis giving the topic time in the spotlight. The overarching consensus is that reducing individual consumption of animal products can pay dividends in fighting the global climate emergency. I want to be clear: I’m not here to dispute this. Are, however, the problems and solutions as simple as we are currently being told by environmental literature, the popular media and, increasingly, the public narrative? 

The Oxford University collaborative project Our World in Data (OWID) has published the striking article: You want to reduce the carbon footprint of your food? Focus on what you eat, not whether your food is local (Ritchie, 2020). The message behind the publication is seemingly simple: eating locally produced food makes very little difference to your carbon footprint as the relative contribution of transport in livestock’s total carbon footprint is very small. A worked example (accessible via the drop-down box in the article) shows the steps taken to reach this conclusion. In brief, the method is as follows:

  • The average carbon footprint for a range of food types were taken from Poore & Nemecek, 2018. For beef, this is 60 kg of CO2 eq./ kg, on average across a range of production types and locations. This figure is considerably larger than the average figures for other food types – for example, peas average at 1 kg CO2 eq./ kg.
  • The emissions from transport (national versus international) are calculated as a proportion of these global average footprints. Imported beef is assumed to travel 9000 km by boat.
  • For beef, where the global average footprint is so large, the contribution of long-haul transport is equivalent to just 0.35% of the total footprint and is essentially negligible. Eating locally therefore makes little difference for total emissions, according to the article.   

The article was rapidly shared online via various social media platforms. On Facebook, the webpage URL racked up nearly 280k engagements (shares, comments and reactions) in the first month of publication with 36.3k of these being shares of the article. (These figures are from SharedCount, a tool that tracks engagement with URLs. The site used the Our World in Data article URL and engagements were counted from publication (24/01/20) to 25/02/20.)

So, why should you think twice about their recommendations? It all comes down to the need to distinguish between local and global food systems and I argue that statistics from one cannot be sweepingly applied to the other. In using global average emissions to study local systems, the article assumes that the environmental benefits of eating locally are determined solely by transport emissions savings – an assumption that does not hold up.

Local versus global scales

Not all food systems are the same and how food is produced varies globally. Organic, pasture-fed, local UK beef is evidently very different from grain-fed cattle or beef ranched on former rainforest land in South America. For clarity: these systems are not necessarily better or worse than each other, but different (a topic already discussed at length by the FCRN in the report Grazed and Confused? (Garnett et al., 2017)). It seems obvious when extremes such as this are cited, but the failure of scientific publications and media alike to recognise these differences is driving the misrepresentation of agriculture and fuelling increasingly vocal, siloed views. 

The environmental impact of livestock is considerably influenced by its production processes and varies globally. The Food and Agriculture Organisation (FAO) has modelled these regional differences using its Global Livestock Environmental Assessment Model (GLEAM). The findings are shown in Figure 1 which has been reproduced from their ‘Tackling Climate Change Through Livestock’ report (Gerber et al., 2013).

Figure 1: Regional variation in beef production and GHG emission intensities. Data source: GLEAM. Figure taken from the Tackling Climate Change Through Livestock report (Gerber et al., 2013).

Figure 1 reiterates that simple global averages do not convey the huge variation in emissions from livestock globally. In fact, the study from which Ritchie (2020) took the global average figures also makes this point. The authors argue that the environmental impacts of products are highly variable and skewed by producers with particularly high impacts – for beef originating from beef herds, the 25% of producers that have the highest impacts produce 56% of beef’s GHG emissions (Poore & Nemecek, 2018).

GLEAM-interactive can be used to show the regional differences between UK-reared and South American-reared imported beef cattle. In the example below, eating local UK produce reduces the environmental impact of beef consumption by 43% relative to eating beef imported from Brazil. This saving does not come from reduced transport emissions, but from differences in production before the farm gate. Brazilian beef was chosen to represent an import with similar transportation distances to those cited in the OWID example. While only 1.2% of the UK’s beef imports come from Brazil, Brazil is the UK’s largest non-EU supplier of beef and the UK’s beef imports from Brazil increased by 45% in the last financial year (AHDB, 2019). While this one example may demonstrate the environmental benefits of local produce, it is extremely production-specific and this must be recognised.

Table 1: Values taken from the FAO GLEAMi calculator. Emissions were calculated using relative contributions of beef from specialised beef herds and dairy herds (compared to the OWID article that only considered specialised beef herds). As the environmental burden of dairy herds are split between milk and beef production their GHG emissions are comparatively lower (globally: 18.2 kg of CO2 eq./ kg of CW compared to 67.6 for specialised beef herds) (Gerber et al., 2013). The bone-free-meat to carcass weight ratio (0.72) was used from Poore & Nemecek, 2018.

Applying global averages to local scales fails to recognise these differences in production and, therefore, footprints. By solely considering the transport emissions for local and international produce and ignoring the differences in the impacts of production we are doing local food a huge injustice in the cases that local food has lower environmental impacts than imported food. This is not to say we shouldn’t study food problems from a global perspective; rather, local recommendations should not be extrapolated from these global scale studies and, importantly, clear signposting should be used to indicate if a statistic applies to the global or local perspective.

It is also worth mentioning that the focus itself on Global Warming Potential (GWP) and CO2 eq. is, by nature, too narrow to truly assess the costs and benefits of eating local produce. Environmentally, CO2 eq. fails to capture wider externalities associated with food production (such as the interplay between farming and biodiversity or water quality). The costs and benefits of eating locally are also not restricted to the environment. Focusing only on the metric of CO2 eq. fails to consider societal and economic consequences of local versus global food systems.

Potayto-potahto: why does it really matter? 

We are at a crucial time in discussions on sustainable food and the debate is becoming increasingly polarised. This is understandable, with so many vested interests – whether they be social, ethical, economic or environmental – the future of food is an emotive topic. Sustainable food production, however, requires that farmers and consumers work collaboratively towards a better future and so we must find some middle ground if we are to move forward. This is perhaps where science can play a part. Research should be used as a tool to provide clarity on sustainable food systems and to inform decision-making for both producers and consumers. However, it must be robust and thorough so as to inform the debate without introducing bias. Interchanging between global and local statistics, if not done carefully, confuses the already messy narrative of sustainable food production.

When assessing local systems, instead of simplifying the debate we need to be asking more questions.  What is the agro-ecological context – how was this food produced and where? This part of the debate is often brushed under the carpet perhaps because the narrative is not as simple as “good food versus bad food”. Looking forward, the focus of the food debate must shift to look at the quality of food (and all food, not just animal-based products) in addition to the quantity consumed. Of course, this is difficult and complex, but it is necessary if the UK food system is to play its part in mitigating climate change, improving our local environments, reducing pollution and improving biodiversity and soil health.



Agriculture and Horticulture Development Board (AHDB)., (2019) The UK cattle yearbook 2019. AHDB. Available from: [Accessed 3 April 2020]

Garnett, T., Godde, C., Muller, A., Röös, E., Smith, P., de Boer, I., zu Ermgassen, E., Herrero, M., van Middelaar, C., Schader, C. and van Zanten, H., (2017). Grazed and confused. Food climate research network, 708.

Gerber, P.J., Steinfeld, H., Henderson, B., Mottet, A., Opio, C., Dijkman, J., Falcucci, A. and Tempio, G., (2013). Tackling climate change through livestock: a global assessment of emissions and mitigation opportunities. Food and Agriculture Organization of the United Nations (FAO).

Poore, J. and Nemecek, T., (2018). Reducing food’s environmental impacts through producers and consumers. Science, 360(6392), pp.987-992.

Ritchie, H., (2020). You want to reduce the carbon footprint of your food? Focus on what you eat, not whether your food is local. Our World In Data. Available from: [Accessed 27 February 2020]


william Waterfield's picture
Submitted by william Waterfield (not verified) on

This is a good summary of a comlicated issue, teh one point that is missed out in teh sumamry is the differing nutritional attributes from teh various food sources and the importance that some of these nutrient have in our diets. 

Simon Ward's picture
Submitted by Simon Ward (not verified) on

The difference is not so much origin as system and allocation of emissions even within a system. As the study points out there is a large difference between a suckler cow system (where the cows emissions have to be applied to the calf) and a dairy system where the emissions of the cow are applied to the other products. The basis is not arbitrary but can be done in more than one way. The nitrous oxide emissions from the manure are as complex - they are an inevitable consequence of keeping stock but can be applied to an arable crop. The big issue is the fattening period - intensive 12 month beef systems based on soya (where not replacing rainforest) will generally have lower emissions than many (but not all) two year systems. I suggest geography is less the issue than system. Ultimately, even if beef has not contributed to global warming reduction in consumption is still an important part of the solution.  

Dick Morris's picture
Submitted by Dick Morris (not verified) on

EXCELLENT and very timely posting. The simplistic use of global average data to compare  systems of production that are fundamentally different is disingenuous at best, and grossly misleading at worst.  As others have noted, the focus on GHG emissions is also potentially missing major aspects of the different systems that bear on issues such as biodiversity and nutrient production efficiency, both of which can indirectly affect climate change. 
Congratulations to the blogger!

Simon Fairlie's picture
Submitted by Simon Fairlie (not verified) on

 Thanks to Ellen Fletcher for making this important point. There are  several other weaknesses in Hannah Ritchie's article which were also present in Weber and Matthew's 2008 paper, which she relies on. Here, with apologies for the length of this post, is my critique of Weber and Matthews from my 2010 book Meat : A Benign Extravagance.

"More food miles means greater energy expenditure; but we should be wary of exaggerating this factor of the energy equation. In 2008, an influential study by Christopher Weber and H Scott Matthews concluded that transport accounted for 11 per cent of US food greenhouse gas emissions, while most of the rest was due to production processes, particularly of meat and dairy. Surprisingly this is rather lower than the figure provided by the British Cabinet Office who state that 15.3 per cent of UK food emissions are caused by transport. But Weber and Matthews’ figures are better supported, and they use them to draw a challenging conclusion, namely that consumers concerned about the carbon impact of their food would do better to reduce meat and dairy consumption than to turn to a local diet:

For the average American household, ‘buying local’ could achieve, at maximum, around a 4-5 per cent reduction in Greenhouse Gas (GHG) emissions, due to large sources of both CO2 and non-CO2 emissions in the production of food. Shifting less than one day per week (ie one seventh of total calories) consumption of red meat and/or dairy to other protein sources, or a vegetable based diet could have the same climate impact as buying all household food from local providers.

In an accompanying table, the authors show that the average US household only needs to divert 17 per cent of its red meat or 38 per cent of its dairy consumption to vegan alternatives to achieve the same greenhouse gas savings that it would achieve by buying locally. Weber and Matthews’ ‘one day a week’ recommendation provides the theoretical basis for the Paul McCartney’s Meat Free Monday campaign (which, in the spirit of multiculturalism, has replaced the Christian tradition that Friday is the day for declining meat).

Weber and Matthews are right, the transport impacts of food production are slight compared to the total environmental impact of the food industry, and reducing consumption of industrially produced meat is an easier way for North Americans to lower the climate impact of their diet than going local. Nonetheless, they exaggerate their case by underestimating the potential for localization. When they talk about ‘buying all household food from local providers’, they don’t mean buying food that is entirely locally produced; they mean food whose final delivery ‘from farm or production facility to the retail store’ is local. Upstream inputs such as deliveries of fertilizer or animal feed are still permissible within their definition of local; and in the case of composite products, such as a strawberry yoghurt, so is the transport of the strawberries, the milk and the sugar to the factory — only the delivery of the yoghurt from the factory to the retailer comes under the heading of ‘final delivery’. 

 In other words, the ‘total localization’ that they use in their comparison is not total at all but represents a reduction of just 39 per cent of all food transport GHG emissions. If it were total, the local consumer would come out looking a lot better in their final analysis: a meat and dairy consumer who went 100 per cent local and refused to buy foods with all these upstream transport impacts would make more or less the same greenhouse gas savings as the dairy consumer who gave up dairy and went vegan. Of course, going local and reducing meat consumption are not mutually exclusive — in many geographical situations they go hand in hand. Consumers who both reduced their animal protein intake to default levels and went 100 per cent local as well would be performing better still. 

Going 100 per cent local is a bit extreme, but there is nothing inherently unrealistic about being up in the eighties or nineties, at least for consumers living in the countryside or market towns. Organic farms with default levels of livestock rely on low inputs from outside, they are less dependent upon fossil fuels than conventional dairy farms, and so are almost by definition highly localized. Unfortunately this scenario is not one that Weber and Matthews have covered in their analysis. 

On top of that, there is a whole further section of the food transport system which is not even mentioned by Weber and Matthews — namely what happens to the food waste. Any nutrients which have not been absorbed by humans, whether or not they have passed through the body, have to find their way back to the land, and the further humans live from the land that produces their food, the further these nutrients have to travel. The recycling of fertility is the aspect of food transport that exercises the FAO most: 

The cost of transporting nutrients to cropland is often prohibitive . . . Longer food chains, driven by the concentration of consumers in urban centres, mean that production systems have to bridge long geographical distances between the site of feed production and the consumer. Decreasing transport costs have allowed the relocation of production and processing activities to minimize production costs. Globally this process has helped to overcome local resource constraints and allowed people in food-deficit areas to be fed. However it also involves large-scale extraction and transfers of nutrients and virtual water embedded in feed and animal products, with detrimental long term consequences for ecosystems and soil fertility. 

Admittedly the FAO are here focussing on animal products, and their concern is to ruralize factory farms by relocating them in the rural hinterland rather than on the periphery of ‘food deficit areas’ (ie. cities), closer to the land than the consumer. But the same problem confronts an urbanized vegan economy: how do you get all the waste from centralized food processing facilities, retail outlets and peoples’ dustbins to cascade back to the land from which it came? The most efficient way discovered so far is by redistributing it in sacks of animal feed that are sold to small farmers who feed it to animals and put their manure back on the land. The least efficient way is to drive it out to a field and dump it, but this is wasteful and the economies of scale involved usually means that there is a problem with nitrogen leaching. For a vegan society, the most promising possibility is anaerobic digestion, but again this takes food out of the food chain, and there is still a transport toll involved in redistributing the heavy digestate that remains after the energy has been removed. None of this crucial part of the food distribution cycle is taken into account by Weber and Matthews.

The alternative is a local food economy whose citizens live where most of the food can be grown, rather than in food deficit areas, and where waste, at whatever stage it may be generated in the food cycle, has a only a short journey back to the land from which it came — either via the route of an animals’ gut and with the assistance of its mobility, or not, as the case may be. This is what ‘total localization’ involves and Weber and Matthews, perhaps because they live in food deficit areas themselves, haven’t grasped this. 

That is not to say that total localization is either desirable or achievable. Grain has to be traded to some extent to alleviate crop failures; and the environmental cost of shipping non-perishable goods to regions where they cannot be grown is slight. There are those who argue that Kenyans should be lifted out of poverty through the otherwise fatuous business of flying fresh flowers and out of season vegetables to Europe. On the other hand, trucking goods across continents to regions where they can be grown satisfactorily causes environmental damage and results in unsustainable concentrations of biomass and nutrients where they are not needed, for the sake of an economic advantage which is pocketed by supermarkets, not by farmers. As for countries like New Zealand that have a surplus of land and food, perhaps they should open their doors to immigrants and parcel out some of their farms to landless peasants. If we are going to globalize everything else, we should globalize land reform as well. 

At the end of their paper, Weber and Matthews point out that GHG emissions are ‘only one dimension of the environmental impacts of food production’. Food miles may not be over-extravagant in their energy use, but they are thickly implicated in a centralized distribution system which multiplies our energy expenditure at every opportunity and whose impacts include excessive packaging and refrigeration, waste, traffic congestion, road-building, noise, accidents, loss of local distinctiveness, exploitation and displacement of peasants, excessive immigration, urban slums, deforestation and habitat destruction, removal of biomass from third world countries, the undermining of local communities in the UK, the collapse of UK farming, and the blood which is spilt over oil fields. It is idle to imagine that we can create local food systems in isolation from this globalized economy. To reinvent a truly local food economy we will have to make the whole system more decentralized, and relocalize the delivery not just of food, but of a panoply of goods such as energy, fibre, building materials, waste disposal and water."

Add new comment

Member input

Plain text

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.