How the data we unmake can make us: a brief reflection from my internship at GRAIN. By Stefano Menegat.

As in a classic novel by Agatha Christie, during the past weeks, I’ve been investigating, collecting clues, following trails, making hypotheses. Like Hercule Poirot, I activated my “little gray cells” to collect evidence and find out why key information was unavailable, why figures were substituted with other figures and who was responsible for this. No, this was neither a Cluedo game nor the theme of an escape room. This was my internship at GRAIN.



GRAIN’s logo:

As reported on its website, GRAIN is a small international non-profit organization that works to support small farmers and social movements in their struggles for community-controlled and biodiversity-based food systems. The word “small” shouldn’t be interpreted as diminutive. GRAIN is perhaps small in terms of the number of employees, but its network and its goals are vast, including local dimensions of social and environmental sustainability and global concerns for climate change and international governance.



Devlin Kuyek, who works out of Montreal, introduced me to the crime scene. Since the beginning of the 2000s, a figure has been circulating in both academic and non-academic circles, a figure everybody cites but nobody really knows where it comes from and how it was calculated in the first place: it is the strange case of the 1-2%. The figure refers to the estimated impact on anthropogenic emissions of greenhouse gases (GHG) due to the production and use of synthetic, nitrogen-based, fertilizers (N-fertilizers). Back in 2015, GRAIN researchers already started to look at this estimate with increasing mistrust because of its unclear origins and new research pointing to a different calculation.

Exxons Agriculture

GRAIN’s 2015 article “The Exxons of agriculture”:

N-fertilizer production is a very energy-intensive process, which relies on the production of ammonia, a compound obtained from the combustion of fossil fuels as feedstock to obtain hydrogen. All this entails a lot of direct GHG emissions, but also relevant indirect impacts, such as the emissions due to the production of the feedstock (i.e. natural gas, coal or oil) and losses from ammonia plants. On the other side, the application of N-fertilizers on fields requires energy for moving around tractors and sprayers. In addition, after the application on fields, N-fertilizers release great amounts of powerful GHGs, such as nitrous oxide. In other terms, the 1-2% estimate seems too conservative, but to find out how the figure was calculated and how a new, more accurate, estimate could be produced required the intervention of a specialist.

This might be the right job for an ecological economist, and as an ecological economist (still in the making), I’ve been asked to take on the case by combing through the scientific literature, the reports from both governmental non-governmental institutions, the figures published by the industry and alternative estimates, possibly more recent and more accurate. “Easy task,” I thought at first, assuming that the literature on the subject would have been well developed and articulated. But I was wrong, the 1-2% figure was omnipresent, widely cited and reported, sometimes interpreted as the total energy “cost” and sometimes as the amount of GHG emissions due to N-fertilizers production and use. Interestingly, the sources I identified, including the IPCC, the FAO, the work of several scholars and the studies commissioned by the industry all shared a sort of blind faith in such estimate without a full understanding of the methods and assumptions upon which the data relied. So, it took me time and effort to discover that the IPCC and others actually took the figure from an industry report based on a fifteen-year-old assessment which in turn was partially based on methods suggested by the IPCC and data already provided by a couple of experts now working as consultants for the industry. I won’t enter here in the details of the investigation; it suffices to say that by using recent activity data and regional emission factors, I found a very different figure: 3.5-4 times higher than 1-2%.

 I think the moral of the story here is quite important for ecological economists: when we think about climate change we consider that science should inform our political choices, but perhaps we neglect that politics is already determining some of our scientific “claims.” What the 1-2% story tells us is that we don’t know and maybe we cannot know what the “real” impact of N-fertilizers is on climate change. There is a huge difference between 1-2% and 7-8% of total GHG emissions – one which could determine the viability of bold political choices like imposing new standards on the fertilizer industry or redefining the position of important actors (i.e. China and India) within the international negotiation system. The limits of evidence-based policy are mostly exposed by the global character of climate change. Probably the real impact of N-fertilizers on GHG emissions is far from the 1-2% and 7-8% figures; surely the impact depends on the global economy and its regional characteristics related to demographic structures, production systems, distribution channels. As Bruce Gardner aptly warned in an article published in the American Journal of Agricultural Economics three decades ago, “the data we make can unmake us,” in the sense that the data we produce can reproduce dangerous and misleading stories. But perhaps the opposite is also true: that the data we unmake can make us, in the sense that our critical role within society depends on our capacity to unveil the complex story of figures and their related narratives (i.e. the 1-2% one)! I think this is true for GRAIN and as an evolving ecological economist, I think this is an important lesson I learned during my internship. This is also the reason why I think scholars should always stay in touch with the world outside of academia, with NGOs like GRAIN for instance, in order to maintain and expand their ability to think critically, to come back to the main assumptions supporting their ideas and to put forward new narratives. The other lesson I learned is that every source we use, no matter how recognized and quoted, can and should be critically examined. It is not by quoting the most cited sources that our theses or our articles will be rigorous (although they may be taken more seriously). On the contrary, the emotional and often uncritical attachment we have to fundamental sources like the IPCC reports should be turned into a renewed interest in and questions about the authors, the stories they present, the figures they use the underlying strengths and the weaknesses of the data. Reading an IPCC report with a critical eye doesn’t lead to negating the existence of climate change, it can lead to a useful starting point to push the current research agenda beyond its limits, towards new directions whose meaning will gradually emerge from the opening of new debates across academic and non-academic worlds.


Gardner, B. L. (1992). How the data we make can unmake us: annals of factology. American Journal of Agricultural Economics74(5), 1066-1075.


Categories: blog

Leave a Reply