Wednesday, November 15, 2017

Digesting a study on meat and climate change

One benefit of writing a blog that no one reads is to be able to write at will. During my hiatus since the end of summer, I have been chewing on a lot of things. Interestingly, of all the things I bit, food seems the most nourishing for this post is about an article titled “Nutritional and greenhouse gas impacts of removing animals from US agriculture”. The full article from the Proceedings of the National Academies of Sciences is here.

The basic point it makes is that shifting to plant-only diet can meet our calorie needs with less environmental impact but will need supplementation of essential nutrients that are deficient in plant diets. This overall conclusion is completely intuitive and reasonable and not particulary striking. Also there are a number of limitations that the authors acknowledge themselves. Models are a simplification of reality and as a modeler  I don't know what it means to be one if I cannot see the limitation in any given model. So I am going to vicariously satisfy the pleasure I might have derived if I had been a reviewer of this manuscript.

But before I get to the limitations, let me say, there was one thought provoking insight buried inside. It is that while deriving calories and protein from meat is more land intensive, livestock production happens on relatively poor quality land and there is only a fixed quantity of good quality land suitable for crops. And so if we are to increase crop production to make up for the shortfall in meat, we have the tough task of growing crops on land not readily suited to farming and we therefore have to use a additional inputs - chemicals, water and technology to augment land quality, which will have it’s own attendant environmental implications. How big will this rebound in pollution be and how much will it offset the benefits of going vegetarian is an empirical question. This is a fascinating extension to this study worth delving into. Now, the concerns I have are the following.

First, I find the title of the paper misleading and inaccurate. The study is a descriptive analysis of the current environmental footprint of rearing animals for human uses. This is completely different from a predictive let alone prescriptive assessment of what would happen if we took any action to change the status quo. For small changes to the status quo, one can extrapolate from a descriptive analysis in a relatively straightforward fashion. However, small changes will have marginal impacts on the environment. LCA studies which are at the core of the claims  about GHG emissions in this PNAS paper cannot be extrapolated to derive the inferences that are implicit in this work. I am going reserve the more dry (i.e. methodological) details for another post. Clearly, the motivation of the authors is about informing actions that can bring about non-marginal or big changes. So object to the word impact in the title, a more apt title would be “Nutritional and greenhouse gas footprint of animals in US agriculture”

Second, agriculture no doubt has a big environmental footprint overall on the earth. However, greenhouse gas emissions are but one of the ways it negatively impacts the environment and agriculture is a much smaller contributer relative to fossil fuel use for non-agricultural uses. But agriculture is perhaps largest contributor to water pollution, soil erosion, flooding, human health impacts for farm workers, loss of biodiversity, etc. Indeed making agriculture more sustainable for these other reasons might have greater benefits, and of course, have attendant co-benefits for climate change albeit a purely climate focus might suggest different changes. In the current political and social climate, a strategy focused on tangible, immediate benefits might encourage people to incur costs that also have long term climate benefits. So it would be an opportunity lost if the wealth data this study has synthesized were not to be exploited to derive implications of a reduction in meat consumption for these other impacts and who knows we might find some trade-offs.

A third thing which strikes me is that there is no mention of ethics or moral implications of not eating meat or better less meat. It is fine for a scientific study, and by that I mean one that simply aims to explain and describe the state of a system to ignore inconvenient topics like ethics and morality. But these cannot ignored in a normative sense. However much the authors might try to be clear about the narrow scope and long list of limitations of their work, one can be certain that the vested interests, of which there is no shortage of, will exploit it to draw the wrong conclusions and obfuscate matters further. That is why I feel a focus on just GHG emissions is insufficient and could even prove counterproductive in the current climate. For instance, I am vegetarian and was raised so for religious reasons. But given my beliefs, if I were not a vegetarian, I am more likely to become one out of concern for treatment of farm animals and aquatic life rather than for the sake of reducing GHG emissions.

There is a lot of fascinating research now trying to ask people how they value the environment and what types of framing and messaging might nudge people in the direction of more environment friendly choices. That is why the awarding of the Nobel Memorial Prize in Economic Sciencest to Professor Richard Thaler for his work on Behavioral Economics is particularly timely. My take of behavioral economics is that misbehaviour is the norm and not the exception. That is people are people and people are not pre-programmed mechanical calculators. Back to the point, unfortunately, almost all this research in a sustainability context is focussed on only one set of activities such as electricity use, transportation choice, food or consumer goods. But I think we need to find out how people trade-off between making choices across these domains. Food, I think is incredibly complex and needs a truly interdisciplinary analysis because it is cultural, social, personal and less given to hedonic analysis, i.e. breaking down into attributes of calories, proteins and vitamins.  On the other hand choices for housing, cars and appliances is amenable to break-down of the value into attributes of cost, convenience, location, environment, etc. and we might have a easier task attacking each of these piece meal. Maybe I ought to drop cars from this list, something tells me taking cars away can be as touchy as denying meat, especially for those of us in America.