1. Introduction
Meeting humanity's needs while not transgressing planetary boundaries (Rockström et al., Reference Rockström, Steffen, Noone, Persson, Chapin, Lambin, Lenton, Scheffer, Folke, Schellnhuber, Nykvist, de Wit, Hughes, van der Leeuw, Rodhe, Sörlin, Snyder, Costanza, Svedin and Foley2009) is increasingly challenging in an age of exponential growth in resource consumption. While progress in human development has accompanied this growth, the economic and development gap between, and within, nations is still widening (Estes, Reference Estes, Brulé and Suter2019), and unequal consumption of resources by wealthier nations and individuals fuels climate change and other forms of environmental degradation (Fuchs et al., Reference Fuchs, Sahakian, Gumbert, Di Giulio, Maniates, Lorek and Graf2021; Wiedmann et al., Reference Wiedmann, Lenzen, Keyßer and Steinberger2020). The impacts of this environmental degradation, particularly climate change, are felt most in those countries that are least responsible for them (Callahan & Mankin, Reference Callahan and Mankin2022), highlighting the need for more sustainable and equitable consumption (Fuchs et al., Reference Fuchs, Sahakian, Gumbert, Di Giulio, Maniates, Lorek and Graf2021; Sahakian et al., Reference Sahakian, Fuchs, Lorek and Di Giulio2021).
The doughnut economy addresses this by presenting a socio-ecological framework which states that sustainability can only be achieved if human activities and consumption stay between the social foundation (SF) and the ecological ceiling (EC) (Raworth, Reference Raworth2017). The SF is the minimum living standards necessary to ensure well-being (Raworth, Reference Raworth2017), while the EC comes from the planetary boundaries, which set environmental maxima for key earth system processes (Rockström et al., Reference Rockström, Steffen, Noone, Persson, Chapin, Lambin, Lenton, Scheffer, Folke, Schellnhuber, Nykvist, de Wit, Hughes, van der Leeuw, Rodhe, Sörlin, Snyder, Costanza, Svedin and Foley2009). These two boundaries form a doughnut around a safe and just space, where people are able to meet their basic needs without outstripping the planet's ecological capacity (Raworth, Reference Raworth2017). No country is currently operating within this defined safe and just space (O'Neill et al., Reference O'Neill, Fanning, Lamb and Steinberger2018; Vogel et al., Reference Vogel, Steinberger, Neill, Lamb and Krishnakumar2021), and 6 of the 9 planetary boundaries have been crossed (Richardson et al., Reference Richardson, Steffen, Lucht, Bendtsen, Cornell, Donges, Drüke, Fetzer, Bala, von Bloh, Feulner, Fiedler, Gerten, Gleeson, Hofmann, Huiskamp, Kummu, Mohan, Nogués-Bravo and Rockström2023). To facilitate a good life within planetary boundaries, an increase in the socio-ecological efficiency of provisioning systems is needed (Fanning et al., Reference Fanning, O'Neill and Büchs2020).
The doughnut economy connects human health and well-being to ecological health (Raworth, Reference Raworth2017), and food provisioning systems are crucial in this context as food has been recognized a key pillar of well-being, which cannot be achieved in the absence of a healthy diet (Holder, Reference Holder2019; Lamb & Steinberger, Reference Lamb and Steinberger2017). The planet has the capacity to support food production for all its inhabitants, but food insecurity persists today, and improper diets have serious consequences for human development and overall health (FAO et al., 2022; Willett et al., Reference Willett, Rockström, Loken, Springmann, Lang, Vermeulen, Garnett, Tilman, DeClerck, Wood, Jonell, Clark, Gordon, Fanzo, Hawkes, Zurayk, Rivera, De Vries, Majele Sibanda and Murray2019). Facilitating well-being via food provisioning will only become more challenging as global food production is expected to increase by 35–56% from 2010 to 2050 (van Dijk et al., Reference Van Dijk, Morley, Rau and Saghai2021) to support an estimated global population of 9.664 billion by 2050 (UN Department of Economic and Social Affairs: Population Division, 2024). High greenhouse gas (GHG) emissions from food (Crippa et al., Reference Crippa, Solazzo, Guizzardi, Monforti-Ferrario, Tubiello and Leip2021) further make sustainable food production a requirement for addressing climate change (Filho et al., Reference Filho, Setti, Azeiteiro, Lokupitiya, Donkor, Etim, Matandirotya, Olooto, Sharifi, Nagy and Djekic2022; Willett et al., Reference Willett, Rockström, Loken, Springmann, Lang, Vermeulen, Garnett, Tilman, DeClerck, Wood, Jonell, Clark, Gordon, Fanzo, Hawkes, Zurayk, Rivera, De Vries, Majele Sibanda and Murray2019).
Research (Gerten et al., Reference Gerten, Heck, Jägermeyr, Bodirsky, Fetzer, Jalava, Kummu, Lucht, Rockström, Schaphoff and Schellnhuber2020) has shown that a radical transformation of the food sector would enable a population of up to 10.2 billion to be fed within 4 key terrestrial planetary boundaries. This transformation would enable the estimated global population of 9.664 billion expected by 2050 (UN Department of Economic and Social Affairs: Population Division, 2024) to be fed within sustainable limits. The many challenges facing the global food system make it difficult to achieve a wide-scale transformation while ramping up food production. These challenges are particularly evident post-COVID-19, with 345 million people globally experiencing high levels of food insecurity, a rise of 200 million since the pandemic (WFP, 2023). This can be attributed to a combination of COVID-19 and its subsequent disruption of global supply chains, the impacts of climate change on agriculture, and global conflicts such as the war in Ukraine (WFP, 2023).
The social challenges associated with food provisioning are compounded by its environmental impacts: over half of habitable land globally is used for agriculture (Ritchie, Reference Ritchie2019), agriculture is the world's largest consumer of freshwater resources (UN Water, 2021); 30% of global energy consumption results from the food production and supply chain (Sims, Reference Sims2011); and food systems are responsible for 34% of global anthropogenic GHG emissions (Crippa et al., Reference Crippa, Solazzo, Guizzardi, Monforti-Ferrario, Tubiello and Leip2021). Decreasing these emissions will involve changes in how we consume food, in particular, the consumption of animal products, as 57% of food's contribution to GHG emissions comes from animal-based food (Xu et al., Reference Xu, Sharma, Shu, Lin, Ciais, Tubiello, Smith, Campbell and Jain2021), which only provide 18% of the global calories and 37% of total protein (Poore & Nemecek, Reference Poore and Nemecek2018).
The EAT-Lancet Commission's (Willett et al., Reference Willett, Rockström, Loken, Springmann, Lang, Vermeulen, Garnett, Tilman, DeClerck, Wood, Jonell, Clark, Gordon, Fanzo, Hawkes, Zurayk, Rivera, De Vries, Majele Sibanda and Murray2019) report on healthy diets and sustainable food systems found that a planetarily healthy diet, along with an ambitious intensification of healthy food production and reductions in food waste, can affect a food systems transformation that allows for human health needs with regards to food to be met within the planetary boundaries (Willett et al., Reference Willett, Rockström, Loken, Springmann, Lang, Vermeulen, Garnett, Tilman, DeClerck, Wood, Jonell, Clark, Gordon, Fanzo, Hawkes, Zurayk, Rivera, De Vries, Majele Sibanda and Murray2019). It further describes a planetarily healthy diet that allows for food production and consumption to remain within scientific targets that it defines for 6 key planetary boundaries. The report also examines environmental impacts of the global food system but does not integrate social issues into its framework. This absence forms the basis for some of the report's critiques, namely that the universality of its dietary recommendations makes it difficult to apply to specific populations (Kaiser, Reference Kaiser2021). The EAT-Lancet 2.0 is further advancing this work through a forthcoming report which will emphasize the need for equity and justice and will feature the results 12-month global consultation process to increase the legitimacy of the original report's recommendations by facilitating greater local buy-in. (EAT-Lancet, 2024; Rockström et al., Reference Rockström, Thilsted, Willett, Gordon, Herrero, Agustina, Covic, Forouhi, Hicks and Fanzo2023).
The sustainable consumption corridor (SCC) concept can aid in addressing the dual social and ecological sustainability gap in food systems research and highlight where inequalities exist within food systems. An SCC is defined as a space where consumption is sufficient to ensure that needs are met, while not crossing a maximum level which would harm the planet and hinder the ability of others to meet their needs (Fuchs et al., Reference Fuchs, Sahakian, Gumbert, Di Giulio, Maniates, Lorek and Graf2021; Sahakian et al., Reference Sahakian, Fuchs, Lorek and Di Giulio2021). This definition expands upon the doughnut economy's pairing of an EC and SF to consider how sustainable consumption can provide everyone the means to live a good life by reducing environmental overshoot and social undershoot simultaneously. It also corrects a critique of the SDGs – that their calls for GDP growth and nature protection are contradictory (Hickel, Reference Hickel2019) – by stating that ecological limits to consumption are needed to achieve sustainability (Fuchs et al., Reference Fuchs, Sahakian, Gumbert, Di Giulio, Maniates, Lorek and Graf2021; Sahakian et al., Reference Sahakian, Fuchs, Lorek and Di Giulio2021).
The need to reduce environmental impacts raises the question of how to best operationalize an SCC as a step toward achieving more sustainable consumption (Sahakian et al., Reference Sahakian, Fuchs, Lorek and Di Giulio2021). Various publications have outlined the theoretical underpinnings of the SCC concept, but the development of sector-relevant SCCs has been limited (Dillman et al., Reference Dillman, Czepkiewicz, Heinonen and Davíðsdóttir2021). The literature suggests using threshold-tied indicators to measure progress toward strong-sustainability, but planetary boundaries studies have had the tendency to propose either indicators or thresholds, rarely both (Dillman et al., Reference Dillman, Czepkiewicz, Heinonen and Davíðsdóttir2021; Fang et al., Reference Fang, Heijungs and De Snoo2015). This is seen in the Food Systems Countdown Initiative (Schneider et al., Reference Schneider, Fanzo, Haddad, Herrero, Moncayo, Herforth, Remans, Guarin, Resnick, Covic, Béné, Cattaneo, Aburto, Ambikapathi, Aytekin, Barquera, Battersby, Beal, Molina and Wiebe2023), which tracks food systems performance across a suite of sustainability indicators, but without thresholds.
The development of a targeted SCC indicator framework for food provisioning systems could help facilitate the transition toward sustainable food consumption. This paper builds upon other evaluations of food sustainability and considers the interconnected socio-ecological domains of food provisioning systems to investigate how the SCC framework can be used to consider the socio-ecological efficiency of food provisioning systems. To that end, this work will address the following research questions: (1) What constitutes an SCC for food provisioning systems? (2) What indicators and thresholds can be used to characterize and measure performance in relation to the SCC?
Taken together, addressing these two questions will result in the development of a comprehensive SCC framework for food provisioning systems. In the methodology section, a food-focused SF and EC are established to delineate the upper and lower boundaries of the consumption corridor, along with a theoretical description what indicators are best suited to measure them, toward the goal of developing a comprehensive indicator set which can be used to gauge progress toward a food provisioning SCC. The methodology section outlines the criteria used to select the most relevant indicators for the SF and EC, and how their thresholds were defined. These indicators and thresholds are presented in the results section as a part of a food-provisioning SCC framework. The article concludes with a consideration of the possibilities this framework presents for evaluations of food sustainability globally and on smaller scales, with suggestions as to how the framework might be adapted for different applications.
This work is needed as existing indicator sets do not connect food security and environmental sustainability within a single framework, demonstrating a research gap. By linking ECs and SFs with suggested indicators and threshold values in its SCC framework, this paper helps fill this gap by producing a benchmark that can be used by policy makers or other officials who are attempting to adjust consumption, which must decrease and/or change to meet global sustainability and climate goals (Dubois et al., Reference Dubois, Sovacool, Aall, Nilsson, Barbier, Herrmann, Bruyère, Andersson, Skold, Nadaud, Dorner, Moberg, Ceron, Fischer, Amelung, Baltruszewicz, Fischer, Benevise, Louis and Sauerborn2019; Ivanova et al., Reference Ivanova, Barrett, Wiedenhofer, Macura, Callaghan and Creutzig2020; Wiedenhofer et al., Reference Wiedenhofer, Smetschka, Akenji, Jalas and Haberl2018). However, the need for reduced consumption is not uniform; many developing countries need to increase consumption to meet their development needs, while wealthier developed countries need to dramatically reduce theirs to stay within reasonable environmental limits (Fuchs et al., Reference Fuchs, Sahakian, Gumbert, Di Giulio, Maniates, Lorek and Graf2021). The food-provisioning SCC presented here could expose where ecological overshoot and social undershoot are occurring, and in doing so, help identify which features of food systems might be adjusted to improve socio-ecological efficiency and promote greater sustainability.
2. Methodology: developing a sustainable consumption corridor (SCC) for food
The following sections describe this work's approach to the SCC concept and how it connects to related concepts such as the doughnut economy. It also describes the domains for the EC and SF and why they were chosen. The final section details the development of the indicator selection criteria and how they were used to select the EC and SF indicators.
2.1 Operationalizing the SCC concept
An SCC is defined as the space between the minimum levels of consumption needed to have a good life and maximum levels of consumption that should not be crossed, lest others not be able to do the same (Di Giulio & Fuchs, Reference Di Giulio and Fuchs2014; Fuchs, Reference Fuchs2017; Fuchs et al., Reference Fuchs, Sahakian, Gumbert, Di Giulio, Maniates, Lorek and Graf2021; Sahakian et al., Reference Sahakian, Fuchs, Lorek and Di Giulio2021). In doing so, the SCC framework uses a sufficietarian perspective based on ‘two types of enough’ to define the ‘safe’ and ‘just’ space where sustainability can take place (Spengler, Reference Spengler2016).
In an SCC, minimum consumption is defined by establishing the needs that are essential to well-being, and translating that into minimum consumption levels, which can vary depending on the theory of needs being used (Fuchs, Reference Fuchs2017). Sahakian et al. (Reference Sahakian, Fuchs, Lorek and Di Giulio2021) address this issue and describe the difference between Maslow's (Reference Maslow1943) hierarchy of needs and the work of Max-Neef (Reference Max-Neef1991) who distinguishes between needs and needs satisfiers. They emphasize that all the approaches share a common interest in the good life as something that goes beyond mere survival and set that as a guiding principle for how minimum consumption should be defined. The minimum consumption that must be met for well-being to be achieved can be conceived of in terms of different categories of resource use or sectors, as a world of sustainable consumption would be defined by many such sectors, which would function together to provide various components underpinning well-being, such as food (Fuchs, Reference Fuchs2017).
Establishing minimum needs is further challenging due to the tendency for the resources required to meet those needs escalating (Brand-Correa & Steinberger, Reference Brand-Correa and Steinberger2017). This escalation connects to the upper end of corridor as described by Fuchs et al. (Reference Fuchs, Sahakian, Gumbert, Di Giulio, Maniates, Lorek and Graf2021, p. 33) who say that ‘maximum consumption standards, in turn, are needed to ensure that consumption by some individuals does not threaten the opportunity for a good life for others’, strongly implying that limits to consumption/consumption maxima be expressed in environmental terms. The SCC shares an overarching vision with the doughnut economy and safe operating space concepts, as its key goal ‘is to support a societal transformation that ensures respect for socio-ecological balance while protecting individual freedom, achieving wellbeing, and promoting social justice worldwide’ (Sahakian et al., Reference Sahakian, Fuchs, Lorek and Di Giulio2021, p. 308). However, a key difference between the SCC and the other two is its emphasis on consumption. While the other two concepts imply and indeed state in their core publications that human activity patterns will have to shift to achieve sustainability, the SCC is much more explicit about the central importance of overconsumption as a barrier to sustainability.
The key SCC publications are not prescriptive in terms of the specific way in which the upper and lower consumption maxima should be defined. In this paper, the upper and lower limits of an SCC for food are defined in terms of the SF and EC. Defining the corridor in this way expands upon the doughnut economy's pairing of an EC and SF to consider how sustainable food consumption specifically can help give people the means to live a good life by reducing environmental overshoot and social undershoot in order to sustainably provide one of the cornerstones of well-being, food. We justify this approach by focusing on the similarities of these ‘double negative’ approaches (Feitelson & Stern, Reference Feitelson and Stern2023), while recognizing the differences between them. This plurality allows for the sectoral operationalization and is compatible with other characterizations of SCCs or doughnut economy approaches in the literature (e.g. Dillman et al., Reference Dillman, Czepkiewicz, Heinonen and Davíðsdóttir2021; Willberg et al., Reference Willberg, Tenkanen, Miller, Pereira and Toivonen2024).
2.2 Characterizing a social foundation for food
This work examines food provisioning systems from a eudemonic perspective which defines well-being as a state that can be achieved when basic needs are met, including food among other needs (Brand-Correa et al., Reference Brand-Correa, Mattioli, Lamb and Steinberger2020; Brand-Correa & Steinberger, Reference Brand-Correa and Steinberger2017; Lamb & Steinberger, Reference Lamb and Steinberger2017). Food security is the framework that global organizations such as the FAO and the World Bank (FAO, 1996, 2003, 2009; FAO et al., 2022; The World Bank, 2022) use to consider whether well-being needs are met with regards to food; it was also included in Raworth's (Reference Raworth2012) definition of the SF. The FAO has defined food security as ‘when all people, at all times, have physical, social and economic access to sufficient, safe and nutritious food to meet their dietary needs and food preferences for an active and healthy life’ – characterized by 4 domains, namely, availability, access, utilization, and stability – of food security (FAO, 1996, 2009; Peng & Berry, Reference Peng, Berry, Ferranti, Berry and Anderson2018). This definition goes beyond subsistence and makes it clear that well-being with regards to food entails more than the prevention of starvation; people must also be able to eat according to their preferences and in doing so, meet their needs.
These 4 domains serve as the foundation for the social floor. We have kept these domains as the foundation of the social floor but changed the 4th domain from stability to resilience, as is detailed below. Figure 1 shows the interconnected domains of the SF and which level of the food system they operate as well as their relationship to the EC, which impacts all domains globally. Resilience is shown separately from the other SF domains (availability, accessibility, and utilization and health impacts) as it operates across all levels (national, household, individual) of the food system, and similarly to the EC, it impacts the performance of the other SF domains. The domains are presented along with the level of the food system they are most associated with, but they are all interconnected and their impacts can cut across these various levels. The specific importance of each domain to the SF is further described in the results section.
Social Foundation for Food Provisioning Systems by domain and scale of the relevance for the domain.
2.3 Characterizing an ecological ceiling for food
This paper outlines an EC for food that presents the various global ecological systems (domains) jeopardized by the environmental impacts associated with food provisioning systems; it further describes the maximum sustainable impact of food provisioning on these domains by defining thresholds and indicators for them. These domains and thresholds were drawn from the EAT-Lancet report, which defined a planetarily healthy level of food system impacts on the most relevant planetary boundaries (Willett et al., Reference Willett, Rockström, Loken, Springmann, Lang, Vermeulen, Garnett, Tilman, DeClerck, Wood, Jonell, Clark, Gordon, Fanzo, Hawkes, Zurayk, Rivera, De Vries, Majele Sibanda and Murray2019). This approach was selected here as the planetary boundaries concept defines environmental maxima beyond which the earth system is no longer able to operate within a safe and just space (Rockström et al., Reference Rockström, Steffen, Noone, Persson, Chapin, Lambin, Lenton, Scheffer, Folke, Schellnhuber, Nykvist, de Wit, Hughes, van der Leeuw, Rodhe, Sörlin, Snyder, Costanza, Svedin and Foley2009), making it ideal for outlining the ecological boundaries of the SCC. Food was found to cross the boundaries for climate change, land-system change, freshwater cycling, phosphorous cycling and biodiversity loss, making it the single largest driver of boundary transgression (Willett et al., Reference Willett, Rockström, Loken, Springmann, Lang, Vermeulen, Garnett, Tilman, DeClerck, Wood, Jonell, Clark, Gordon, Fanzo, Hawkes, Zurayk, Rivera, De Vries, Majele Sibanda and Murray2019).
These domains were selected here for the EC, and Figure 2 presents the scale of each domain and if they are top-down or bottom-up in character in terms of how their impacts are calculated and aggregated. Figure 2 shows the different domains of the EC and the geographic level their boundaries are calculated (green). The method of aggregation for the boundaries is also reflected, whether they are calculated at a global level and then divided to calculate per capita boundaries or thresholds (top down, shown in blue) or calculated at a sub-global level and then aggregated together for the global boundary (bottom-up, grey). The specific importance of each domain is further described in the results section.
Ecological Ceiling for Food-Provisioning Systems by domain and planetary boundary operational scales (inspired by Fang et al., Reference Fang, Heijungs and De Snoo2015).
2.4 Indicator selection for the social foundation and ecological ceiling
The indicator selection process followed a similar three-step approach as presented by Dillman et al. (Reference Dillman, Czepkiewicz, Heinonen and Davíðsdóttir2021), in which, first, the material topics relevant to the sector in question were reviewed, in this case, the food provisioning sector. After reviewing the material topics, how these topics were typically measured were reviewed. Lastly, using the indicator selection criteria described in the previous section, appropriate threshold indicators were then selected. Each step in this process followed an iterative approach whereby the lead author reviewed the literature and collected indicators, which would then be reviewed by the rest of the research team, in which indicators were assessed according to the selection criteria and sources expanded using the snowball method of authors knowledge and related resources.
2.4.1 Indicator selection criteria
The indicator selection criteria were developed in line with other approaches in the food provisioning systems literature; for example, Aurino's (Reference Aurino2014) work outlining the indicator selection process used for FAO et al.'s (2013) suite of food security indicators. Aurino (Reference Aurino2014) emphasized the importance of linking indicator selection to the goals of the assessment in question to avoid ‘laundry lists’ of indicators. Principle 4 of the BellagioSTAMP recommendations for sustainability assessment – which highlights the need to pair indicators with key domains and compare indicator values with defined targets – also informed the selection criteria (Pintér et al., Reference Pintér, Hardi, Martinuzzi and Hall2012).
As per Lancker and Nijkamp (Reference Lancker and Nijkamp2000) (Dillman et al., Reference Dillman, Czepkiewicz, Heinonen and Davíðsdóttir2021; Fang et al., Reference Fang, Heijungs and De Snoo2015) indicators do not provide any information about absolute sustainability, unless associated with a threshold value. Indicators without thresholds can evaluate the progress of a system toward a state of greater sustainability, but do not specify when sustainability is reached (Lawrence, Reference Lawrence1998). Both types of indicators are useful but given the intended purpose of this paper – to evaluate the sustainability via a food provisioning systems SCC framework – indicator selection for this paper included a preference for indicators with associated thresholds or targets. Indicators without thresholds were only included if they captured a component of the domains which could not be captured by a threshold linked indicator. The indicator selection criteria presented below represent a synthesis of these approaches to developing indicator sets:
1. Indicators must be based on scientific evidence.
2. Indicators should:
a. connect to food consumption (or associated consumption of resources), show an outcome which is dependent upon consumption, or describe factors that impact the ability of a society to support an adequate level of consumption with regards to their associated domain;
b. directly relate to SF achievement, that is, meat consumption ratios are not relevant as it is possible to achieve the SF for food with no meat consumption at all;
c. be scale appropriate for the domain they fall under;
d. be presented with thresholds, if they are state rather than directional indicators.
3. Indicators must be suitable for comparison between geographic locations which may or may not have very different food systems.
4. Indicators should not be age or gender specific except in cases where an indicator is more typically reported that way (e.g. stunting in children vs. underweight for adults).
Following the initial selection of indicators using the above criteria, the following two criteria were applied to develop the final indicator set:
5. Indicators should be considered holistically as a set to avoid unnecessary overlap between indicators while still covering the essential components of the SF domains.
6. Indicators that are already directly reported and globally available should be chosen over ones which can be calculated through the analysis of publicly available data, to ease data availability constraints.
2.4.2 Selecting social foundation indicators
Following the process outlined in section 2.4, first, a desk review of academic literature and food policy sources to gather background information regarding the role of food provisioning systems in promoting well-being and SF achievement was performed. This initial review led to the development of the SF domains connected to the definition of food security (FAO, 2022; FAO et al., 2013). With the SF domains selected, a second review of relevant indicators was performed to gather indicators that could measure performance and progress for each of the domains. The relevant academic literature for both the background and the indicator review processes was obtained via searches in Google Scholar and Scopus, and most of the non-academic sources were accessed by combing through relevant governmental sources, for example, UN, FAO, World Bank, etc. This initial collection resulted in a bank of 75 indicators from 6 sources. After refining for relevancy and the selection criteria, the final 18 indicators presented in Table 1 were selected.
Indicators of social foundation domains of food provisioning systems, by domain
2.4.3 Selecting ecological ceiling indicators
The indicator selection criteria outlined above were also used to develop the EC indicators and thresholds. The process of selecting indicators for the EC was simpler than for the SF as the EAT-Lancet report further connected the planetary boundaries to food provisioning systems by defining 6 planetary boundaries that food systems have the greatest impact on and outlining thresholds for the indicators associated with each of them. These thresholds represent the maximum contribution to each boundary that will allow the food system to stay sustainable. Given the widespread use of the planetary boundaries framework, and its centrality to the SCC concept, those same indicators and thresholds set out in the EAT-Lancet report were largely adopted here.
While the planetary boundaries concept was developed as a global framework, it has since been operationalized on smaller scales, though not without debate regarding the appropriateness of these methods (Ferretto et al., Reference Ferretto, Matthews, Brooker and Smith2022). To facilitate wider applicability of SCC framework, the threshold figures for each EC domain were downsized to an acceptable per capita annual contribution from the food system to each domain. Using the per capita number, it is possible to calculate the relative impact on the boundaries that a given area or region should have, given its population size. This is in line with the approach of O'Neill et al.'s (Reference O'Neill, Fanning, Lamb and Steinberger2018) work, which downscaled the planetary boundaries to a per capita level to compare the impacts of individual nations on the planetary boundaries with their SF performance. Moberg et al. (Reference Moberg, Karlsson Potter, Wood, Hansson and Röös2020) and Hallström et al. (Reference Hallström, Davis, Håkansson, Ahlgren, Åkesson, Wolk and Sonesson2022) also downscaled the boundaries presented in EAT-Lancet (Willett et al., Reference Willett, Rockström, Loken, Springmann, Lang, Vermeulen, Garnett, Tilman, DeClerck, Wood, Jonell, Clark, Gordon, Fanzo, Hawkes, Zurayk, Rivera, De Vries, Majele Sibanda and Murray2019) to evaluate the sustainability of diets in Sweden. While acknowledging that downscaling raises issues concerning how to justly allocate ecological impacts while taking historical patterns into consideration (e.g. Hickel, Reference Hickel2020), the authors adopted a similar framework here, given the scope of the SCC framework. Taking a top-down approach and downscaling each threshold to a per capita value – thereby assigning responsibility the ecological impacts captured by the domains equally – allows for the SCC to be used in international comparisons of food provisioning systems in various countries over time.
3. Results: threshold linked indicator system for the characterization of the food provisioning systems SCC
The results section details the domains and indicators used to develop a combined food-provisioning SCC. The relationship between the SF and EC domains, and how they function to delineate sustainable vs un-sustainable consumption is outlined below in Figure 3. Figure 3 shows the inner and outer boundaries of an SCC for food provisioning (inspired by Raworth [Reference Raworth2017] and Fuchs et al. [Reference Fuchs, Sahakian, Gumbert, Di Giulio, Maniates, Lorek and Graf2021]). The EC domains (green) outline the boundary beyond which consumption is unsustainable, while the SF domains (orange) show the social needs that must be met via food provisioning in order for consumption to be sustainable. The indicators collected to measure performance across the SF and EC domains are then presented in Tables 1 and 2.
A food provisioning SCC, depicting the ecological ceiling and social foundation, with identified domains, used in this work to define the ‘safe’ and ‘just’ space for food provisioning.
Threshold-linked indicators for ecological ceiling domains of food-provisioning systems
Note. Per-capita threshold numbers were calculated by the authors using the global threshold numbers and population numbers from the UN Department of Economic and Social Affairs (2024) except for the CO2eq (all GHG) threshold, where the global threshold was derived from the per-capita one.
3.1 Social foundation
3.1.1 Domains
As described in section 2.1, the SF domains of availability, accessibility, utilization and health impacts, and resilience were drawn from the food security framework, and are further defined here before their associated indicators are presented below. Availability refers to a sufficient quantity of nutritious food to support human health and well-being at the national level, yet is dependent on factors on the global level as well, given the importance of imported food (FAO, 1996, 2006, 2009; Peng & Berry, Reference Peng, Berry, Ferranti, Berry and Anderson2018).
Accessibility operates at the household level and is the ability to access the resources needed for a nutritionally sufficient diet. These resources can be physical (e.g. transportation infrastructure) as well as economic (food affordability). Access also requires a socio-cultural support system to assist those households who do not have the financial or physical means to access food (FAO, 1996, 2006, 2009; Peng & Berry, Reference Peng, Berry, Ferranti, Berry and Anderson2018). Developing countries often struggle with the physical components of accessibility, as infrastructural difficulties often prevent access to a steady food supply (Peng & Berry, Reference Peng, Berry, Ferranti, Berry and Anderson2018). In wealthy countries, members of deprived socio-economic groups are more likely to struggle with food availability and accessibility due to unequal distribution of resources within those countries that result in food deserts and other barriers to food acquisition (Long et al., Reference Long, Gonçalves, Stretesky and Defeyter2020; Pollard & Booth, Reference Pollard and Booth2019), despite their greater purchasing power.
Utilization describes the extent to which individuals can obtain a nutritious diet. This requires clean water and hygiene facilities, the knowledge needed to prepare nutritious meals, as well as being physically healthy enough to digest and cook food (FAO, 1996, 2006, 2009; Peng & Berry, Reference Peng, Berry, Ferranti, Berry and Anderson2018). Negative health impacts resulting from food provisioning systems are linked to utilization and are key to overall SF achievement, so those health impacts are also included here.
Resilience operates across all three levels and addresses the ability of the food system to deal with shocks (FAO, 2009; Peng & Berry, Reference Peng, Berry, Ferranti, Berry and Anderson2018). These shocks include natural phenomena such as climate change and disease, as well as man-made ones, like wars and economic crises. Stability relates to the reaction of food systems to shocks but also refers to the stability of the other three domains over time (Anderson, Reference Anderson, Ferranti, Berry and Anderson2019). Resilience was chosen instead of stability for this domain to better capture how provisioning systems respond to disturbances, as instability has become the norm in a global system that is increasingly prone to shocks, making resilience a requirement for sustainable food provisioning systems moving forward.
3.1.2 Indicators
The SF indicators displayed below in Table 1 were chosen via the process outlined in section 2.3.1 and taken together, they provide an overview of SF performance. The indicators are structured according to the SF domains, and the smallest number of indicators which would fully capture the primary components of each domain were chosen. While this set of indicators is intended to be comprehensive and accurately convey performance at a country wide level with regards to the SF, all components of each domain were not necessarily directly captured by the indicator set. For example, physical access or proximity to a stable food supply is a key component of the SF, but Table 1 does not include an indicator which directly measures physical access to healthy food. This is primarily because none of the indicators collected through the indicator search were suitable for use across a variety of different food systems. For example, distance from a grocery store, an indicator which makes sense in an urban environment, is far less informative in a rural context where people might be growing a large portion of their food supply locally. However, this does not mean that physical access is not captured by the framework, as thresholds for more general performance indicators would not be met in the absence of physical access to food.
The prevalence of undernourishment indicator is useful to consider in this context. It is situated within the accessibility domain of this framework, in part because doing so reflects how it is reported by the FAO (2022), but also because it serves as an indicator of overall accessibility. The accessibility domain of food provisioning systems is considered to operate at the household level; accordingly, some of the other indicators included in this section like diversity of household diet are collected and processed at the household level. Prevalence of undernourishment is an individually collected indicator which is then aggregated to produce a national-level result. However, this does not preclude its utility in evaluations of household-level food security. While accessibility is a household-level domain, given that most food access is mediated through household-level food purchasing, food insecurity is typically measured on the individual level. Given that high rates of food insecurity would indicate a lack of food access, prevalence of undernourishment is acceptable as a proxy state indicator for food accessibility.
3.2 Ecological ceiling
3.2.1 Domains
As described in section 2.2, the EC domains of climate change, nitrogen and phosphorus cycling, freshwater use, land-system change, and biodiversity loss are drawn from the planetary boundaries framework as employed in the EAT-Lancet report. The impacts of food provisioning on each domain are briefly described here, ahead of the presentation of their associated indicators. Agriculture is a leading contributor to climate change via processes including methane emissions (IPCC, 2019b), nitrous oxide emissions from fertilizer (Gao & Cabrera Serrenho, Reference Gao and Cabrera Serrenho2023), deforestation and other land clearing techniques, and food transport, storage, and waste (IPCC, 2019a; Willett et al., Reference Willett, Rockström, Loken, Springmann, Lang, Vermeulen, Garnett, Tilman, DeClerck, Wood, Jonell, Clark, Gordon, Fanzo, Hawkes, Zurayk, Rivera, De Vries, Majele Sibanda and Murray2019). Production, trade, and usage of fertilizer in agriculture also disrupts nitrogen and phosphorus cycles and drives runoff into freshwater systems, causing eutrophication (Alewell et al., Reference Alewell, Ringeval, Ballabio, Robinson, Panagos and Borrelli2020; Anas et al., Reference Anas, Liao, Verma, Sarwar, Mahmood, Chen, Li, Zeng, Liu and Li2020; Willett et al., Reference Willett, Rockström, Loken, Springmann, Lang, Vermeulen, Garnett, Tilman, DeClerck, Wood, Jonell, Clark, Gordon, Fanzo, Hawkes, Zurayk, Rivera, De Vries, Majele Sibanda and Murray2019). This impact compounds food's massive freshwater footprint, where food production is world's largest water-consuming sector (UN Water, 2021).
Food provisioning leads to land-system change, with half of all habitable land given over to agriculture (Ritchie, Reference Ritchie2019), with animal agriculture being especially land intensive (Sun et al., Reference Sun, Scherer, Tukker, Spawn-Lee, Bruckner, Gibbs and Behrens2022). Land/sea use change is also the largest driver of recent biodiversity loss globally (Jaureguiberry et al., Reference Jaureguiberry, Titeux, Wiemers, Bowler, Coscieme, Golden, Guerra, Jacob, Takahashi, Settele, Díaz, Molnár and Purvis2023). This impact is significant as biodiversity provides supporting and regulating ecosystem services (Willett et al., Reference Willett, Rockström, Loken, Springmann, Lang, Vermeulen, Garnett, Tilman, DeClerck, Wood, Jonell, Clark, Gordon, Fanzo, Hawkes, Zurayk, Rivera, De Vries, Majele Sibanda and Murray2019), as well as influencing and often improving ecosystem stability (Loreau & de Mazancourt, Reference Loreau and de Mazancourt2013; Pennekamp et al., Reference Pennekamp, Pontarp, Tabi, Altermatt, Alther, Choffat, Fronhofer, Ganesanandamoorthy, Garnier, Griffiths, Greene, Horgan, Massie, Mächler, Palamara, Seymour and Petchey2018). Without these services, agricultural output is threatened, and food systems are more vulnerable to external shocks such as climate change (FAO, Reference FAO2019).
3.2.2 Indicators
The EC indicators which are outlined in Table 2 are compliant with the indicator selection criteria outlined in section 2.3.1, but they are drawn from the planetary boundaries presented in the EAT-Lancet report. These thresholds define the outer limit of acceptable ecological impacts annually from the entire food provisioning system – including fisheries (Troell et al., Reference Troell, Jonell and Crona2019) – across 6 domains that represent the key planetary boundaries most impacted by the food system.
Global and per-capita thresholds are given for all indicators, except for the biodiversity domain, where the extinction rate indicator is only presented with a global threshold. The global threshold was selected using a conservative rationale that the rate of extinction from agriculture should be no greater than the background extinction rate (Willett et al., Reference Willett, Rockström, Loken, Springmann, Lang, Vermeulen, Garnett, Tilman, DeClerck, Wood, Jonell, Clark, Gordon, Fanzo, Hawkes, Zurayk, Rivera, De Vries, Majele Sibanda and Murray2019). We did not reduce this threshold to a per-capita level as it is difficult to ‘assign’ an acceptable number of extinctions to individuals. However, a per capita threshold is provided for cropland use; given land system change's role in driving extinction globally (Jaureguiberry et al., Reference Jaureguiberry, Titeux, Wiemers, Bowler, Coscieme, Golden, Guerra, Jacob, Takahashi, Settele, Díaz, Molnár and Purvis2023; Tilman et al., Reference Tilman, Clark, Williams, Kimmel, Polasky and Packer2017), this also provides information about biodiversity loss.
The only climate change indicators presented in the EAT-Lancet report were CH4 and NO2 emissions, which were reported in CO2eq. The rationale given for not including CO2 in the EAT-Lancet report was that the climate change threshold value was meant to represent the maximum GHG emissions which would be acceptable, assuming the Paris Agreement's 2050 targets were met which would eliminate CO2 emissions from energy production and transition to sustainable agriculture systems, shifting land use from a carbon source to a carbon sink. Given the extensive current CO2 emissions resulting from land-use change, transportation, farm and food processing equipment, biogenic emissions, and other impacts of food production and consumption in our current food provisioning systems, an additional CO2eq (all GHG) indicator was added to the climate change domain. The CO2eq (all GHG) figures, which show the 2030 and 2050 per-capita targets needed in order to not cross the global threshold, were calculated using data from Akenji et al. (Reference Akenji, Bengtsson, Toivio, Lettenmeier, Fawcett, Parag, Saheb, Coote, Spangenberg and Capstick2021). As is shown in Table 2, the per-capita thresholds fall year by year in line with expected population increases, further demonstrating that the ecological efficiency of food provisioning will have to increase.
4. Discussion
This study aimed to provide a framework for evaluating the socio-ecological efficiency and adequacy of food provisioning systems via the production of an indicator set which details the ecological maxima and social minima of food provisioning systems. These indicators were selected using a qualitative review process that was used to develop indicator selection criteria, resulting in the indicators presented in Tables 1 and 2. The implications of those indicators are discussed here, including their contributions to the existing food sustainability literature, their policy relevance, how they might be tailored for use in different contexts, and an overview of limitations of the indicator system presented.
4.1 Connecting ecological performance with human well-being in food provisioning systems
This indicator set helps bridge the research gap concerning environmental sustainability in food provisioning systems and SF achievement. Food security research often ignores the environmental component of sustainability, even though sustainable development as a paradigm and a policy goal is intimately linked with the organizations which hold institutional responsibility for food security monitoring. For example, the FAO identifies itself as the custodial organization for many of the SDGs related to food security, including goals 2, 4, and 6. However, in the FAO's food-security indicator set, none of the indicators included directly address the environmental impacts of food production and consumption. This is understandable given their focus on food security through the prism of human development but presenting these social indicators alone ignores the human–environment connection in food provisioning systems. Their ability to promote human well-being is missed, as many of the actions which could mitigate the environmental footprint of food provisioning systems, could also promote well-being (Willett et al., Reference Willett, Rockström, Loken, Springmann, Lang, Vermeulen, Garnett, Tilman, DeClerck, Wood, Jonell, Clark, Gordon, Fanzo, Hawkes, Zurayk, Rivera, De Vries, Majele Sibanda and Murray2019).
Similarly, considerations of food system sustainability often fail to fully address the role of food provisioning systems in promoting well-being. This is true of Willett et al.'s (Reference Willett, Rockström, Loken, Springmann, Lang, Vermeulen, Garnett, Tilman, DeClerck, Wood, Jonell, Clark, Gordon, Fanzo, Hawkes, Zurayk, Rivera, De Vries, Majele Sibanda and Murray2019) EAT-Lancet report, which describes a planetarily healthy diet which would meet human dietary needs in a sustainable fashion but does not consider the social factors which underly the ability of individuals to meet their nutritional needs. The sectoral approach to SCC analysis utilized here links the impacts of human consumption to their well-being and sustainability impacts in a concrete way (Dillman et al., Reference Dillman, Czepkiewicz, Heinonen and Davíðsdóttir2021). Given the size and complexity of food provisioning systems, a set of ecological and socially linked performance indicators (that measure the extent to which the needs of a given population within sustainable limits) serve as a useful tool for identifying where ecological undershoot or social overshoot is occurring in food provisioning systems. The potential for this set of indicators to be used as a way of tracking ecological overshoot and social undershoot is highly relevant given the identified need to improve socio-ecological efficiencies in provisioning systems (Fanning et al., Reference Fanning, O'Neill and Büchs2020). This efficiency needs to be analyzed at multiple levels, and tailoring the global framework presented here for use in smaller scale studies would further enhance its utility.
4.2 Indicator and threshold challenges
While the indicator set provides a comprehensive snapshot of social and ecological sustainability of food provisioning globally, it was challenging to fully capture aspects related to the social and cultural importance of food within the existing domain structure. The current SF domains articulate the structures that must be in place for individuals to meet their well-being needs with regards to food in accordance with their personal preferences, as required by the accepted definition of food-security. Those preferences are socially and culturally derived and connect to other aspects of well-being such as affection and creation. Due to challenges associated with deriving sustainability thresholds for such domains, these aspects were not included here, but considering the cultural importance of food while recognizing the need to enter an intergenerationally sustainable state represents an interesting avenue for future research.
This was not the only aspect of the domains that was difficult to capture. For example, of the SF domains, the characteristics of resilience made it particularly challenging to measure, as many of the factors impacting resilience operate at the international level yet are experienced differentially at the national level. Resilience indicators would ideally capture the exposure of food provisioning systems to various shocks impacting food provisioning systems, as well as how equipped countries are to respond to them. Key among these underlying shocks is climate change, which has been identified as a major threat to food systems globally. Notre Dame's climate Global Adaptation Index measures countries vulnerability to climate change as well as their readiness to implement adaptation. The vulnerability component of the index includes indicators of the food system impacts of climate change, making it a suitable choice based on the indicator selection criteria (Chen et al., Reference Chen, Noble, Hellmann, Coffee, Murillo and Chawla2015). Import dependency ratio and value of food imports over total merchandise exports help illustrate how vulnerable a food system might be to these underlying shocks, as a high level of variability or adequate financial resources could indicate higher vulnerability resulting from an inconsistent food supply, but they do not measure the underlying shocks which could lead to variability or the impacts of those shocks on food systems. This is a fundamental limitation of all indicators included here (and indeed indicators more broadly) and incorporating the indicators into dynamic systems models more capable of capturing these underlying shocks is an avenue for future research.
In addition to indicator availability, another issue that emerged during the indicator review was that of threshold consistency. While thresholds were available or could be selected by the authors for most indicators, due to differences between indicators and the circumstances they are meant to benchmark, different thresholds represented different levels of achievement depending on the indicator in question. For example, some of the indicators, such as people using safely managed drinking water service (which has a threshold of 100%) were associated with thresholds that represent an idealized scenario where a need is essentially being perfectly met. Others, such as the threshold of 100 for the average dietary energy supply adequacy (ADESA) indicator, represent a minimum standard that must be met in order for consumption needs related to the domain whose performance the indicator is measuring be fulfilled.
ADESA measures whether a given country has an adequate supply of dietary energy to meet its population's needs. In countries with low levels of undernourishment, ADESA values span a wide range of values, from 113 for Japan to 149 for the USA, indicating significant variation in the amount of energy supply countries use to meet their citizens' needs. However, many countries with higher levels of undernourishment have comparable ADESA values, for example, Bangladesh, with an ADESA of 112, and 11.4% prevalence of undernourishment. Given this variation, an ADESA value of 100 was chosen as a minimum threshold, as countries with a value of less than 100 definitely do not have sufficient resources to feed their populations, even if in practice, a value of greater than 100 is required. This illustrates the challenges associated with implementing thresholds as a mandatory part of an indicator system, and the practical advantage of approaches such as that of the Food Systems Countdown Initiative (FSCI) which do not require thresholds for their indicators. The FSCI could therefore be a useful indicator system if the goal is simply tracking food system performance over time, but for evaluating progress toward sustainability, the SCC provides additional clarity.
4.3 Future work and policy relevance
As the wide range of impacts articulated in Table 1 shows, changing food provisioning systems such that they do not prevent the planet from meeting targets outlined in the SDGs and the Paris Agreement will require a substantial systems transformation. The EAT-Lancet report highlights the need to sustainably increase food production and reorient agricultural policies from emphasizing the quantity of food over quality and coordinate strong governance of land and oceans as a way of doing so (Willett et al., Reference Willett, Rockström, Loken, Springmann, Lang, Vermeulen, Garnett, Tilman, DeClerck, Wood, Jonell, Clark, Gordon, Fanzo, Hawkes, Zurayk, Rivera, De Vries, Majele Sibanda and Murray2019). The SCC framework for food provisioning systems, by providing a baseline assessment of socio-ecological performance that highlights where ecological or social improvement is most needed, could further analyze which specific policy interventions might be most useful to facilitate this shift.
Adjusting the framework so that it could be used in a variety of contexts, including more localized scale could be particularly helpful if it is being used as a policy benchmark. Works such as ‘A good life for all within planetary boundaries’ (O'Neill et al., Reference O'Neill, Fanning, Lamb and Steinberger2018) and the EAT-Lancet report (Willett et al., Reference Willett, Rockström, Loken, Springmann, Lang, Vermeulen, Garnett, Tilman, DeClerck, Wood, Jonell, Clark, Gordon, Fanzo, Hawkes, Zurayk, Rivera, De Vries, Majele Sibanda and Murray2019), which respectively examined the relationship between ecological impacts and social achievement and the ecological impacts of food systems, acknowledged the difficulties inherent in scaling down global thresholds to per capita ones so that they could be compared on a national scale. These works and others that similarly scale down the planetary boundaries have been critiqued for doing so (Ferretto et al., Reference Ferretto, Matthews, Brooker and Smith2022). However, given the intended purpose of the food provisioning SCC – to provide a snapshot of food-related consumption and its socio-ecological efficiency in a globally comparable way – we believe the per capita approach was suitable here. Furthermore, work has been done showing that GHG emissions in many countries can increase in the short term and medium term, even if historical emissions are not taken into account (Ala-Mantila et al., Reference Ala-Mantila, Heinonen, Clarke and Ottelin2022), showing that a per-capita approach can still reflect the need for increased consumption in some countries.
While the framework in its current form is suitable for global comparisons, it would be possible to downscale it to evaluate the sustainability of food provisioning systems within a single country. When scaling the framework down to a national level, the indicators selected for each domain could be weighted through participatory approaches through stakeholder engagement such that they can increase relevancy to the given context surrounding food provisioning systems in that country, as well as being feasible with regards to any local data availability limitations. This is in line with the methodology seen in Dillman et al. (Reference Dillman, Heinonen and Davíðsdóttir2023) which measured Iceland's performance toward an SCC for mobility provisioning systems, and a similar approach could be taken for a food provisioning SCC. Taking this approach could help illustrate where ecological overshoot or social undershoot are occurring within countries, and aid policy makers who are considering how to change consumption patterns.
5. Conclusion
Consumption has yet to be fully mainstreamed into considerations of socio-ecological sustainability in food provisioning systems, demonstrating a need for further research. This paper has shown how a food provisioning SCC could be used as tool for furthering this kind of research and assist policy makers in viewing sustainability issues through a consumption-oriented lens. It does so by defining domains and thresholds to outline a food provisioning SCC which can be used to indicate where ecological overshoot or social undershoot are occurring, therefore highlighting where consumption must shift to facilitate movement into a safe and just space for food provisioning. The global thresholds define the boundary of an SCC for food provisioning systems and by providing per-capita thresholds for each domain, the paper illustrates how the framework presented here also could be used to develop national-level SCCs which could be used to perform food sustainability evaluations within a given country. These could be useful in international or national sustainability policy making and research, as the SCCs could serve as benchmarking tool for that helps identify where inequality or socio-ecological inefficiencies exist, thereby guiding the direction of future decision-making.
Author contributions
Conceptual development, S. S. K, K. J. D., B. D., and J. H.; methodology, S. S. K., K. J. D., B. D., and J. H.; writing – original draft, S. S. K.; writing – review and editing, S. S. K, K.J. D., B. D., and J. H.; project administration, K. J. D., B. D., and J. H.; supervision, K. J. D., B. D., and J. H.; funding acquisition, K. J. D., B. D., and J. H.
Funding statement
This work was supported by the Icelandic Research Council (RANNÍS) [228717-051].
Competing interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Research transparency and reproducibility
The authors declare that they have complied with standards for research transparency and reproducibility.
Open access
