It’s infeasible or condemns half the population to starvation.

Degrowth supporters call for the adoption of a more frugal lifestyle, based on local production and food self-sufficiency, and of short food chains (urban agriculture perfectly represents this model) (Latouche, 1993, 2008; D'Alisia et al., 2015). Some people envisage a somewhat autarchic society (Latouche, 1993; Pallante, 2005, 2011 - Maurizio Pallante is a leading figure of the Italian degrowth movement). Latouche (1993, p.164e166), distinguishes between “auto-subsistence”, where the farmers of a country would produce only for their survival, turning their back to the market (and to the urban population), and “self-sufficiency”, where farmers would produce a surplus for the market, able to feed the urban population of the country. Nevertheless, this policy, according to the author, should have the global market as a benchmark. However, this relies on the assumption that a country has enough agricultural land to produce all the food it needs, and sufficient internal energy and natural resources (i.e. fuel, machinery, labour) to run agricultural activities. This may not be the case for European countries, for example.

Some authors (e.g., Latouche, 1993, 2008; Pallante, 2011; Kallis et al., 2015) argue that a more rural society relying on renewable energies would be able to quickly create hundreds of thousands of new jobs, thereby helping to reduce unemployment. At the same time, a working week of four days should also be possible, and allow people to dedicate themselves to cultural activities and social relations (Kallis et al., 2013). While such ideas can offer food for thought, actual research on agriculture and degrowth is much needed in order to start providing models and scenarios for public debate. Feasibility (the compatibility of the effort with the external constraints imposed by the environment), and viability (the compatibility with internal constraints) analyses are needed to provide insights on the potentials and constraints of agricultural practices (Giampietro et al., 2013, 2014). Giampietro points out that the concept of “desirability” (compatibility with human expectations), is also relevant when addressing sustainability, as there can potentially be viable solutions that people might not like or accept (Giampietro, 2004; Giampietro et al., 2013, 2014). Large-scale scenario analyses will allow scaling up local production systems, and check whether, and how much, local food self-sufficiency, short food chains, and low-input farming practices are compatible with a country's food security, and how new farming practices would impact on its societal metabolism. At present, this is a key step forward that needs to be taken. Alternative agricultural practices have been suggested that are surely of interest. Nevertheless, a sound analysis of the role they might play in sustaining society, or of how society should change to rely on such agricultural practices, is still missing. For example, permaculture has been referred to as a sound model for food production (e.g., Latouche, 2012). Nevertheless, to date, there are no scientific publications, and, to my knowledge, no other publications either, presenting sound data concerning the performance of this type of farming system (see also Gomiero et al., 2011a). Lately, Ferguson and Lovell (2014, 2015) published two reviews confirming that data concerning the performance of this practice are still missing. Therefore, at the moment, it is not known how much food permaculture can supply, nor at what cost, both financial and environmental. Concerning organic agriculture, quite a few works are available on its performance (Gomiero et al., 2011b; de Ponti et al., 2012; Seufert et al., 2012). Nevertheless, the overall sustainability of organic agriculture is still debated, and there is a lack of scenario analysis in relation to the impact of a large-scale transition to low-input agriculture on the food system and society. That is to say, linking the agricultural system to societal metabolism. Georgescu-Roegen advocated the adoption of organic agriculture, but he was aware that a reduction in yield was to be expected. Therefore, he argued that population should be reduced accordingly, “mankind should gradually lower its population to a level that could be adequately fed only by organic agriculture.” (Georgescu-Roegen, 1975, p. 378). Kallis et al. (2015, p. 7) also recognises that “A solar civilization can only support smaller economies, given the low EROI of renewable energies compared to fossil fuels. A transition to renewables will inevitably be a degrowth transition.” It is important, then, to carry out sound research on what the transition implies, e.g., how many people can we expect to be living in a new solar-powered society? With what standard of life?

A process of deintensification of agriculture is certainly needed, in particular in regions such as Europe and the USA. Nevertheless, different realities face different problems. In some densely populated areas of Asia, de-intensification may not be possible due to the high demographic pressure and the lack of economic resources to import more food (Smil, 2000; Conway, 2012; Seufert et al., 2012). Concerning deintensification, the issue of how much a return to a more rural, self-sufficient society is biophysically feasible and viable in the long-run, and how this can be desirable for people, should be properly addressed. For example, turning European society into self-sufficient, no-inputs family farms may not even be feasible, because of a simple biophysical constraint: the lack of land to meet the food demand of its large population. Let us consider Germany as an example. Below, Table 1 illustrates the feasibility of Germany becoming self-sufficient under traditional/ organic farming, using the currently available data. Germany has a population of 82 million people, and its arable land amounts to 11,846,000 ha (“arable land” is the term used to describe the most productive land). To maximise the production of calories for Germany to feed its people, the following rather optimistic assumptions are made: all soil is of the best quality, best practices are used and best yield obtained. Full mechanisation is available, irrigation is available when needed, pest and weed control is provided and fully effective. There are no losses in the production-food chain (losses may reach 20%). To maximise the production of calories and protein, while providing a natural fertilisation of fields, wheat-pea rotation is assumed as the farming system. It is also assumed that all peas are used as human food; no animals are fed with crops from arable land. Yield for conventional wheat in Germany is 7t per ha, and 3t per ha for organic wheat (EC, 2013), a good yield of organic peas is about 4t per ha. In the field: In some years, yield has to be expected to be lower due to pests and climate extremes; Phosphate fertilizer should be added, or alternatively composted manure has to be applied every four years (that would require livestock to be fed, reducing the amount of crops for humans); The rotation cannot be sustainable, as pest outbreaks would emerge, a one-year break should be introduced after a rotation, or other crops used. In the food system: A conservative 10% losses in the food chain (field, storage, etc.) should be accounted for. If that is done, the production system fails to cover the cost of the basal metabolism; The diet is too poor in fats and other key nutritional elements, that required an important fraction of the arable land be allocated to different crops and to feed animals; If fossil fuels are not used, a lot of animal power is required for the farm works. That means that some land should be allocated to feed the animals. From Table 1 we see that, even using all the country's arable land to crop wheat and peas (to have a mix of energy and proteins) and assuming maximum productivity, no animals to be fed and zero loss in the production chain, the productivity (7.5 1013 kcal yr1 ) would barely suffice to sustain the population's basal metabolism (6 1013 kcal yr1 assuming an average energy intake of 2000 kcal cap. day1 ), and fail to meet the energy demand of an agrarian society whose work is physically demanding (9 1013 kcal yr1 assuming an average energy intake of 3000 kcal cap. day1 ). If we consider that the real productivity may be lower than the highest possible, as assumed in the model: 10e20% of the production may be lost along the food chain, an animal population is to be fed, and other crops, with lower energy and protein content, have to be cropped in order to have a more balanced diet, we may conclude that under old traditional farming practices (“traditional organic agriculture” Tello et al., 2012), or current organic agriculture, German agriculture may sustain maybe 50%, or less, than the present population, and provide a much lower quality diet (perhaps facing hunger in winter time). If intensive agriculture were to be allowed, but not food imports, the German population would have to put up with a diet based mostly on the direct use of cereals (no beer allowed!). Estimates suggest that to provide for a vegetarian diet, a minimum 0.1 ha per capita are necessary, while 0.15 ha per capita are needed if some animal products are added (Smil, 2000). Of course, the example provided in Table 1 is completely hypothetical and far from reality. We should also account for the demand for labour (if fossil fuels were to be excluded, maybe an 80e90% of the population should be working in the fields), the final cost of the food, as well as the overall effect on the organization of society (on the latter issue see Sorman and Giampietro, 2013). To conclude, if the urban population (75% of the German population, WB, 2016b), were to be become rural, with most families relying on their own farm's production (as some proponents of degrowth seem to suggest, e. g, Pallante, 2005, 2011), there would be little agricultural land left to crop. The environmental impact of this ruralisation process would be devastating and the socioeconomic cost enormous: it would be impossible to deliver to people even the most basic services. This is not to say that it is not possible to have a more rural society and reduce the environmental impact of agriculture (which we actually urgently need - Foley et al., 2011; Gomiero et al., 2011a; Stehle and Schulz, 2015), but that we have to analyse the possible impacts that such changes may have on the functioning of society, and the feasibility, viability and desirability of the potential alternatives.