Agriculture is just one of the sectors in which carbon labelling — the labelling of a product to show how much carbon (and other greenhouse gases) have been emitted during its ‘lifecycle’ — is being used to show how individual products contribute to climate change. The logic behind applying carbon labels to agriculture seems sound enough: agriculture accounts for 10 to 12 per cent of global greenhouse gas emissions and produces much of the food we eat and the products we buy. Finding a way to tell consumers how much individual agricultural products contribute to this should encourage them to choose those products with the lowest carbon footprint and help make agriculture more sustainable. But the truth is that it is very difficult to provide accurate carbon labels for agricultural products. And carbon labelling can impact farmers in the developing world in ways that don’t support development.
Tesco leading the charge
Tesco — the third largest grocery retailer in the world — has recognised the role that agriculture plays in climate change. Aided by the Carbon Trust, it has adopted carbon labelling based on carbon footprinting that aims to show “the total carbon dioxide and other greenhouse gases emitted during [the product’s] life, including production, use and disposal.” Tesco hope this will make it easier for its customers to make more ‘sustainable’ choices in their shopping — for example, by choosing its non-biological washing liquid instead of powder or tablets because the liquid has a smaller carbon footprint.
So far, Tesco is using carbon labels on about 120 of its own brand products, from milk, orange juice and potatoes to washing detergents, light bulbs and toilet roll.
But other UK retailers — including the Co-op and Sainsbury’s — are reluctant to follow suit because it “remains unclear how carbon might be measured, how the information will be used by industry, and how the customer might usefully use the information to make informed choices”.
These retailers are right to be cautious — carbon footprinting, and labelling, is bedevilled by uncertainty, in both its methods and the data it uses.
What to measure and how to measure?
There are some crucial factors that make carbon labelling problematic:
1) A lack of detailed, context-specific data that shows the differences at the farm level — and therefore between production systems — in emissions of greenhouse gases (GHGs). Data is particularly lacking for the developing world. Small-scale farms in the developing world are labour intensive rather than capital intensive (that is, they primarily use manpower rather than machinery). This means their contributions to GHG emissions are often far lower than that of their developed counterparts. But the data (and therefore the labels) don’t highlight these complexities.
2) Disagreement over where to draw the boundary around production systems. Take footprinting apples as an example: you might choose to include the energy used to cool the apple during storage, but should you also include the energy used to build the cold store itself? Current footprinting methods would suggest not, which again favours developed over developing world production.
3) Disagreement over whether to include land use change in calculations. Land use change often makes up a significant chunk of GHG emissions. Some footprinting methodologies include a calculation of the contribution of land use change to GHG emissions because converting grasslands and forests to make way for agriculture leads to a release of carbon, but others don’t. PAS 2050 — the methodology that forms the basis of the Carbon Trust Label — includes land use change, but only change that happened after 1990. Developing countries have typically converted land to agriculture much more recently than the developed world. This somewhat arbitrary choice of date has important implications for equity — the total emissions of a product sourced from the developing world is very likely to be higher than the developed world equivalent because land use change will be included in calculations of the former, but not the latter. To make matters worse, detailed data on land use change is lacking.
4) An inability to account for ‘positive’ factors in production systems, such as the presence of trees that can store carbon and are common in many developing country farming systems (for example, cocoa and coffee). Carbon labelling focuses more on metrics than addressing how agricultural practices might contribute to low carbon production.
5) The sheer number of methodologies used for carbon labelling — some sources say 16 different methodologies are in use or have been developed since 2007. This is confusing for all involved.
6) Very few, if any, developing country representatives are involved in the design of carbon footprinting schemes due to a lack of expertise and financial resources. This is a key barrier to addressing the current inequity of carbon footprinting.
7) It’s unclear if consumers actually understand carbon labels and if it will drive behaviour change that helps to address the impact of agriculture on climate change.
It’s not fair, so don’t do it
Carbon labels in their current form confuse consumers, do not paint a true picture of what is happening on the ground and disadvantage developing country producers. What’s more, they might even distract us from other environmental issues — such as water and biodiversity — that also require urgent attention.
Measuring a product’s emissions at every stage from farm to plate undoubtedly has its uses. In particular, it can help identify emission ‘hotspots’ where efficiency improvements can be made within a company’s supply chain. But carbon labelling as we currently know it is both premature and badly thought out. Data and methodological issues and, above all, inequity need to be urgently resolved. Promoting particular agricultural practices may be a more effective way to address the contribution of agriculture to climate change.