Those interested in the environmental labeling of food products will have noted the availability of the eco-score in early May in the Yuka application. These are now 23 million consumers who have the eco-score to guide their choices on tens of thousands of food products based on their environmental impact.
All this would be very good and very beautiful if this eco-score was not one of the worst possible environmental scores, widely criticized by associations and interprofessions, favoring the least sustainable agricultural practices, and inciting the food industry to reduce their actions in favor of agriculture.
Sustainable products penalized
Calculating an eco-score is a childish simplicity. All you have to do is take the list of ingredients constituting a food product, add up the environmental impacts from Ademe’s Agribalyse impact database, and add some bonuses for packaging, labels and supply. local. And that’s all. The eco-score has the merit of simplicity and the major flaw of miss the most important topic: agriculture.
Most of the environmental impact of a food product comes from the agricultural production of its ingredients (60 to 90% of the impact depending on the sector). However, the Agribalyse 3.0 database of Ademe does not takes into account that the average values of conventional farming practices today. The calculated eco-score is therefore identical for milk produced using sustainable agro-ecological practices or milk from cows fed with soybeans from Brazil. Ditto for a strawberry, a banana, a tomato or a grain of wheat: the environmental impacts are considered in “average value”, regardless of their mode of production, sustainable or not.
Result: the eco-score penalizes sustainable products in favor of polluting products within their category. Worse, it encourages manufacturers to focus on actions favorable to their eco-score, such as packaging or local procurement, and to look away from agriculture, the main factor of impact. Yuka’s initiative will therefore at best have the effect of promoting the least durable products and discouraging the most committed manufacturers, which is exactly the opposite of the desired goal.
An approach to review
Ademe indicates that the Agribalyse 3.0 database should be enriched over time and integrate the impacts of sustainable agricultural practices. But it is the general approach to calculation (based on LCA, life cycle analysis) which needs to be reviewed.
Agricultural production is not a production of bolts. Its environmental impact is linked to a multiplicity of factors: its “mean value” does not make much sense. To estimate, for example, the CO2 impact of a dairy farm, you need 6 to 10 parameters, including the cow’s nutrition mode, their productivity, the feed production on the farm or the herd management mode.
The “average value” of CO2 emissions per liter of milk does not exist, neither for “normal” milk, nor for organic milk. What exists is a multiplicity of situations and practices which, combined, make it possible to calculate the environmental impacts per farm, per production area and therefore ultimately per food product.
This approach of calculation by models is more complex than a calculation by “average values”, but it seems to us much more suited to the living world. Finally, it would make it possible to move away from “averaging”, and to promote, by agricultural production basin, the efforts of industrialists for sustainable and climate-friendly agriculture. At a time when the government is looking for ways to encourage agricultural and agrifood players to take their environmental responsibility, it is urgent not to discourage them by a simplistic and counterproductive eco-score.
Jeremiah Wainstain is founder and leader of The Green Data.