A social–ecological analysis of the global agrifood system

Elisa Oteros-Rozas, Adriana Ruiz-Almeida,Mateo Aguado, José A. González, and Marta G. Rivera-Ferre

PNAS December 26, 2019 116 (52) 26465-26473; first published December 16, 2019 https://doi.org/10.1073/pnas.1912710116


Edited by B. L. Turner II, Arizona State University, Tempe, AZ, and approved November 7, 2019 (received for review July 23, 2019)

Significance

The failure to end hunger and the environmental deterioration underpinning food systems have prompted a paradigm shift around food security. We propose a social–ecological approach and carry out a quantitative analysis of 43 indicators of food sovereignty and 28 indicators of the sociodemographic, social wellbeing, and environmental sustainability situation in 150 countries. The results highlight the existence of an agrifood debt among countries (i.e., disequilibria in the natural resources consumed, the environmental impacts, and the social wellbeing in regions that play different roles within the globalized agrifood system). Three spotlights underpin this debt: 1) interregional contrasts in food security, 2) a concern about the role of agrifood trade, and 3) a mismatch between regional biocapacity and food security.

Abstract

Eradicating world hunger—the aim of Sustainable Development Goal 2 (SDG2)—requires a social–ecological approach to agrifood systems. However, previous work has mostly focused on one or the other. Here, we apply such a holistic approach to depicting the global food panorama through a quantitative multivariate assessment of 43 indicators of food sovereignty and 28 indicators of sociodemographics, social being, and environmental sustainability in 150 countries. The results identify 5 world regions and indicate the existence of an agrifood debt (i.e., disequilibria between regions in the natural resources consumed, the environmental impacts produced, and the social wellbeing attained by populations that play different roles within the globalized agrifood system). Three spotlights underpin this debt: 1) a severe contrast in diets and food security between regions, 2) a concern about the role that international agrifood trade is playing in regional food security, and 3) a mismatch between regional biocapacity and food security. Our results contribute to broadening the debate beyond food security from a social–ecological perspective, incorporating environmental and social dimensions.
agrifood system
food sovereignty
hunger
social wellbeing
sustainability development goals

Agrifood systems, given their place among the most vulnerable coupled nature–human systems (1, 2), have multiple interactions with global environmental change and play a major role in the present and future of humanity (3, 4). They contemporarily sustain and challenge social wellbeing and human life on the planet (4) by providing food while contributing to global greenhouse gas emissions, land degradation, eutrophication, and water quality depletion (5⇓–7). Five of the 7 planetary boundaries are directly linked to agrifood systems (8, 9). However, in a context in which enough calories are produced to feed the entire human population (10), chronic hunger still affects 1 in 9 people in the world (11). The evident failure of policies to end hunger and the environmental deterioration underpinning agrifood systems have prompted a paradigm shift in the way that food security is approached both scientifically and in policy terms: from being mostly focused on the technical and agrarian aspects of food production (1, 5, 12, 13) to adopting a social–ecological systems approach (14) and more precisely, an agrifood system approach, including both environmental sustainability and social wellbeing (4, 6, 14⇓⇓⇓–18). This systemic perspective allows emphasis on the use of natural resources for primary production as well as food transformation, commercialization, and consumption, therefore connecting pieces within agrifood systems that previous analyses had studied separately.

Likewise, the multiple social and ecological dimensions of food security are transversal to 14 of the 17 United Nations’ 2030 Sustainable Development Goals (SDGs) (19). The SDGs aim to eradicate poverty, establish socioeconomic inclusion, and protect the environment (20), and therefore, the need for an integrative approach to address them has already been discussed (21). However, critical studies have elicited an incompatibility between the environmental and the social and economic aspirations of the SDGs (22).

While food security is commonly defined as the physical, social, and economic ability to access sufficient, safe, and nutritious food (23), SDG2 implicitly recognizes that a broader approach to food security is needed to end hunger. Indeed, for the first time, SDG2 links the objectives of zero hunger, food security, and improved nutrition (subgoal 1) with the need to promote sustainable agriculture (subgoal 2) (19). SDG2 explicitly refers to the global moral imperative to eradicate hunger while respecting environmental sustainability. However, this entails controversy about the tradeoffs between achieving food security mainly through increasing food production (24, 25), addressing and minimizing the environmental impacts of agriculture and food (6), and adapting to climate change (26). Therefore, actions toward the transformation of agrifood systems need to account for the synergies and tradeoffs that exist between SDG2 and other SDGs (27), which need to be identified through social–ecological systems approaches.

The international debate around the social–ecological sustainability of food systems is prolific, and different authors have recently suggested a change from the land sparing/sharing debate to a focus on human wellbeing (4); proposed “leverage points” for improving global food security and environmental sustainability (6); argued for sustainable healthy diets to keep food systems within the planetary boundaries (28); developed assessments of the environmental impacts of food systems (29, 30); applied metrics for the assessment of the sustainable nutrition outcomes of food systems (31); and provided valuable systemic analyses of global food systems (18). However, although it is widely acknowledged that the social and environmental costs and benefits associated with environmental change are not distributed equally among actors and regions (32), how this phenomenon occurs is still poorly understood, because less than 6% of food security publications in the past 25 y included equity or justice as part of their analysis (14).

A tipping point in these international debates was the 2012 32nd Regional Conference for Latin America and the Caribbean (33), where the United Nations Organization for Food and Agriculture (FAO) agreed to initiate discussions about alternative approaches to address hunger and the unsustainability of agrifood systems. One such approach was food sovereignty (FSv) (33). FSv emerged in the late 1990s, arguing that hunger is not merely a matter of food availability and quality but that equally or more critical issues are political aspects harnessing equity and justice within food systems. Some of those political aspects may include agricultural trade liberalization, power relations between different actors (from small producers to large transnational corporations and consumers, therefore within and between countries), a lack of wide social participation along the whole food chain, or access to the means of production (34, 35). FSv, as a political concept, was coined by La Via Campesina, an international movement of farmers, peasants, and landless workers, and has been developed and discussed at large by civil society organizations, farmers’ trade unions, academia, governments, and international institutions (36) to become a well-rooted concept (37). Within this approach, food is framed as a human right, including its environmental and sociocultural aspects, which are viewed as both drivers and outcomes of food security and where small farmers are considered to play a central role (34).

However, progress toward ending hunger needs to be measurable through indicators that aid societies in assessing their performance (20). Gustafson et al. (31) and Ruiz-Almeida and Rivera-Ferre (38) expressed the need to address food security from a systems perspective that incorporates social–ecological sustainability, and to do so, they provided comprehensive sets of indicators at the national scale. Ruiz-Almeida and Rivera-Ferre (38) proposed a panel of 97 indicators, but they did not perform any analyses on data. Gustafson et al. (31) suggested 23 indicators but applied them only to 9 countries. Chaudhary et al. (18), based on Gustafson et al. (31) and adding 2 additional indicators on biodiversity impacts and health-sensitive nutrient intake, applied descriptive analyses to quantify 156 nations performance.

However, achieving SDG2 requires not only measuring performance but also, understanding the social–ecological relations among different countries beyond national patterns. Therefore, based on data for the set of indicators proposed by Ruiz-Almeida and Rivera-Ferre (38), we analyze here the global agrifood system from a social–ecological perspective.

While previous works advanced in the same direction, we believe that our contribution constitutes a step farther in 2 ways. First, with the multivariate analysis of 43 indicators of FSv, 150 countries are grouped according to their relative scores. Hence, a picture of the dynamics of the global agrifood system at a larger scale is provided. Second, we enlarge the social–ecological approach by incorporating the description of world regions according to 7 demographic and economic indicators, 4 social wellbeing indicators, and 17 environmental sustainability indicators.

In particular, we carried out a principal component analysis (PCA) of indicators (variables) describing the agrifood system (SI Appendix, Tables S3–S7) in countries (observations). Some indicators were originally published as indexes or proportions, while others needed to be transformed in relative terms with respect to country area or population size in order to adjust magnitudes and allow comparison between countries (details of the definitions, sources, and data transformations are given in SI Appendix, Tables S1 and S2). Using the standardized coordinates of the most significant PCA factors, we performed a hierarchical cluster analysis (HCA) based on the Euclidean distance and Ward’s agglomerative method. Finally, we used Kruskal–Wallis and χ2 tests to characterize each cluster of countries according to its performance in terms of FSv, social wellbeing, and environmental sustainability (a more detailed description of statistical methods and results can be found in SI Appendix). This approach aims to contribute to other ongoing efforts to provide indicators for monitoring SDG progress (http://indicators.report).

Through the proposed analyses, we 1) identify world regions formed by countries under similar conditions of FSv, 2) relate the state of FSv of the different regions with their state of social wellbeing and environmental sustainability, and 3) critically reflect on the implications for SDG2 of an agrifood debt between world regions that has been so far poorly addressed. Within-countries agrifood debt and inequalities can be also relevant to provide a complete picture, but their analysis is out of the scope of this paper.

Results

The State of Food Sovereignty: Who Is Who in the Global Agrifood System.

The 150 countries analyzed were statistically clustered into 5 groups (Fig. 1 and SI Appendix, Table S8) according to their performance in the 43 indicators across the 6 pillars of FSv (Figs. 1 and 2). By also characterizing them according to their bioregional context (SI Appendix, Table S9), socioeconomic characteristics (SI Appendix, Table S10), FSv (SI Appendix, Table S11), environmental sustainability, and social wellbeing (SI Appendix, Table S12), we grasp who is in the global food system and what relation with the ecological and sociopolitical dimensions of food the different groups of countries have.

Download figure
Open in new tab
Download powerpoint
Fig. 1.

World map with the countries colored according to the groups that emerged from the PCA and the HCA: purple is group 1, blue is group 2, green is group 3, yellow is group 4, red is group 5, and gray indicates the countries excluded from the analysis due to a lack of data.


Download figure
Open in new tab
Download powerpoint
Fig. 2.

Box plots of FSv indicators in the 5 clusters identified. Only indicators with statistically significant differences between the groups are represented. The 6 pillars of FSv are separated into (A) access to resources, (B) productive models, (C) commercialization, (D) food security and food consumption, (E) agrarian policies and civil society organizations, and (F) gender. The letters at the top of each box plot (A–D and combinations) correspond to the statistically significant differences in the pair comparisons (more details are in SI Appendix).


Landgrabbed and Undernourished: Agricultural Exporters but Food Importers.

The first group includes 45 countries (SI Appendix, Fig. S4 and Table S8) mostly from eastern, middle, and western Africa (Fig. 1 and SI Appendix, Table S9). The countries in this group show the lowest gross domestic product (GDP) per capita and very low income (SI Appendix, Table S10). The food systems in the countries of this group are characterized by a productive model based on the largest rural and agricultural population of the sample, the smallest cultivated area per farmer, the largest total economically active population in agriculture, a limited use of fertilizers, and a low production of meat (Fig. 2 and SI Appendix, Table S11). Agriculture is responsible for a high share of the GDP of these countries, which are, on the one hand, the largest exporters of agricultural products (significantly different from all other groups) and on the other hand, the largest importers of food (although only significantly different from groups 4 and 5), showing net reception of official development assistance for food and agriculture (SI Appendix, Table S11). This group of countries ranks first in suffered landgrabbing (SI Appendix, Table S10). The population in the countries of this group shows the lowest levels of access to resources, such as electricity, sanitation, and drinking water; the most severe food deficits; and significant vulnerability, which is consistent with the lowest protein supply and adequacy of the dietary energy supply among the groups (Fig. 2 and SI Appendix, Table S11). That is, the value added of the agricultural products produced is not retained, exporting huge amounts of agricultural products while failing to feed large shares of their population. These countries also show the lowest degree of economic, social, and political globalization, despite some of them being among the biggest exporters of luxury commodities, like coffee and cocoa. The indicators of social wellbeing are coherent with the former, as this group of countries has the worst scores for all of the indicators analyzed, including significantly shortest life expectancy and worse life satisfaction (Fig. 3 and SI Appendix, Table S12). In terms of environmental sustainability, these countries are associated with the lowest ecological footprint and low CO2 emissions and water withdrawal from agriculture (Fig. 3 and SI Appendix, Table S12).

Download figure
Open in new tab
Download powerpoint
Fig. 3.

Box plots of environmental sustainability (A) and social wellbeing (B) indicators in the 5 clusters identified. The letters at the top of each box plot (A–D and combinations) correspond to the statistically significant differences in the pair comparisons (more details are in SI Appendix).


Diverse Intensive Producers of Crops.

The second class is the largest and most heterogeneous group, clustering together 49 countries, mostly from Asia and the Americas (Fig. 1 and SI Appendix, Tables S8 and S9). In comparison with the other groups, these countries have the largest population densities and low–medium income (SI Appendix, Table S10). These countries show medium levels of rural and agricultural populations, however, with the smallest agricultural area and production of mammals (Fig. 2 and SI Appendix, Table S11). What they have in common is that they are characterized by intensive production models based on a large use of fertilizers and agricultural water withdrawal (also used for wheat, maize, and/or rice production, for which some of the countries in this group, such as China and India, are among the largest producers in the world) and the vastest surfaces of permanent crops (as percentage of agricultural area) that are dedicated partly to fruit production (mostly in Southeast Asia and Central and South America). They are large food exporters (Figs. 2 and 3 and SI Appendix, Table S11), and also, they are characterized by the second highest food deficit, low energy and protein intake (Fig. 2 and SI Appendix, Table S11), and overall intermediate levels of social wellbeing (Fig. 3 and SI Appendix, Table S12).

Least Ecologically Wealthy and Landgrabbers.

The 18 countries clustered in this class are not geographically or ecoregionally grouped (Fig. 1 and SI Appendix, Tables S8 and S9). They show medium population densities and GDP per capita (SI Appendix, Table S10). Countries in this group show a limited proportion of agricultural area and forests as well as overall very little cereal, meat, and fruit production but the greatest use of fertilizers per hectare (Fig. 2 and SI Appendix, Table S11). Little of the population lives in rural areas or is dedicated to agriculture, and overall, the population in these countries seems to have good access to all resources except renewable water (the group includes several island states) (Fig. 2 and SI Appendix, Table S11). These countries have limited exports of agricultural products, and they are net food importers, with limited value added to agriculture (Fig. 2 and SI Appendix, Table S11). Most of these countries coincide with those in which the importance of food imports has increased in recent years. They show an intermediate level of food security and consumption in comparison with the other groups, despite limited subsidies invested in supporting agriculture (SI Appendix, Table S11). Overall, they seem to have a good situation in terms of social wellbeing (Fig. 3 and SI Appendix, Table S12). Two significant characteristics of this group are the largest biocapacity deficit in comparison with the other 4 groups (Fig. 3 and SI Appendix, Table S12) and the largest area of land grabbed abroad as a percentage of its total area (SI Appendix, Table S10).

Intensive Food Producers and Exporters.

The fourth group clusters 8 vast countries from Oceania and the Americas with the largest intragroup variance (Fig. 1 and SI Appendix, Tables S8 and S9). This group exhibits high income and the largest GDP per capita alongside the smallest population densities (SI Appendix, Table S10). Access to resources is high in these countries, and the agrifood system is focused largely on a model of intensive production of cereals, fruit, meat, and biofuels dependent on large inputs of pesticides and with very low rural and agricultural populations (Fig. 2 and SI Appendix, Table S11). However, organic agriculture (as a percentage of total agricultural area) scores high in this group due to Uruguay and the United States. Countries in this group seem to be the “breadbasket of the world”: they show a large share of food and agricultural exports while indicating limited food imports (Fig. 2 and SI Appendix, Table S11). In fact, some of them, such as Australia, Argentina, Canada, the United States, and most recently, Brazil dominate global food exports. This group also shows the largest financial support for agriculture (Fig. 2 and SI Appendix, Table S11). Food deficit is quite low, and the protein supply is high, consistent with the high level of energy adequacy (Fig. 2 and SI Appendix, Table S11). Another common feature of countries in this group is their large ecological footprints and agricultural CO2 emissions as well as a large biocapacity that sustains their large biocapacity reserves (Fig. 3 and SI Appendix, Table S12). Furthermore, they have the best records for all social wellbeing indexes (Fig. 3 and SI Appendix, Table S12).

Overnourished Agricultural Importers.

The fifth group is the most homogeneous group and includes 30 countries (SI Appendix, Fig. S4 and Table S8), mostly in Europe (Fig. 1 and SI Appendix, Table S9), and it features the largest population densities, the second largest GDP per capita, and high income (SI Appendix, Table S10). Access to resources is satisfactory in these countries (Fig. 2 and SI Appendix, Table S11). Agriculture is quite intensive, with little rural and agricultural populations, little presence of women who are economically active in agriculture, and a large use of fertilizers (Fig. 2 and SI Appendix, Table S11). These are the largest importers of agricultural products, however, with little imports and exports of food (Fig. 2 and SI Appendix, Table S11). People in these countries are overall food secure but have a diet mostly based on a large consumption of proteins (Fig. 2 and SI Appendix, Table S11). These countries are in a biocapacity deficit because of the ecological footprint of the built-up land and the croplands, instead showing the lowest grazing footprint and agricultural water withdrawal but large CO2 agricultural emissions (Fig. 3 and SI Appendix, Table S12). These are the countries with the largest degree of globalization and net contribution of official development assistance for food and agriculture as well as overall high levels of social wellbeing (Figs. 2 and 3 and SI Appendix, Tables S11 and S12).

Discussion

The FSv indicator framework (38) used here to analyze agrifood systems from a social–ecological systems perspective (14) contributes to systematically and quantitatively assessing environmental, social, and economic relationships between countries within a globalized world. It allows us to measure progress through periodical monitoring: ideally, countries should not be so easily clustered, and if they are, the groups should not show significant differences in terms of their sociodemographic characteristics, environmental sustainability, and social wellbeing. The current cluster might help countries to 1) be conscious of the impacts of their national agrifood policies in other countries and on their social–ecological sustainability; 2) evaluate the dependence of their food security, social wellbeing, and environmental sustainability on other countries; and therefore, 3) allow governments to make sensible changes to agrifood policies to contribute to SDG2. In addition, these results might also provide guidance to development banks and other financial institutions.

Winners and Losers: Agrifood Debt.

The International Council for Science critically pointed toward the internal inconsistency between ecological sustainability and socioeconomic progression in the SDG framework (39), and there is scientific quantitative evidence about the nature and extent of this incompatibility of sustainability and development (39⇓–41). Our results provide evidence of the challenges to end hunger in a globalized agrifood system, which is far from equitable from both socioeconomic and environmental perspectives. Certain countries, such as Australia, Brazil, Argentina, and those in Europe and North America, hold a critical stake and should reduce overconsumption, while other countries, such as most of Africa, would benefit from improving their self-sufficiency. Consistent with previous research, our results suggest that intertwined and nested material flows in global agrifood systems result in interregional social inequities in the distribution of both costs and benefits of producing, trading, and consuming food, hence affecting social wellbeing, unevenly distributing environmental impacts, and challenging environmental sustainability.

In line with the concept of “ecological debt,” which was coined by academics in 1992 and further adopted and developed by civil society organizations and governments (42), the results presented here illustrate an agrifood debt (i.e., the interregional social–ecological disequilibria in the natural resources consumed, the environmental impacts produced, and the social wellbeing attained by populations in regions that play different roles within the globalized agrifood system). Given that a substantial proportion of the world's 815 million people who are unable to meet daily food needs are food producers, such as small-scale farmers and fishers (11), our results are consistent with and strengthen the conclusions of extensive previous research showing that food security is largely a matter of redistribution, entitlements, food access, and access to services and means of production (22, 43). Globalization poses complex tradeoffs for food system resilience across scales due to high social, economic, and ecological interconnectedness, tradeoffs, and, hence, vulnerability (14). We suggest that this agrifood debt, which has been poorly addressed to date, should be recognized, assessed, and monitored through 3 spotlights: 1) a severe contrast in diets and food security between regions, 2) a concern about the role that international agrifood trade is playing in regional food security, and 3) a mismatch between regional biocapacity and food security.

Nutritional and Environmental Contrasts in Diets and Food Security between Regions.

An unbalanced food system features a contrast between high rates of undernutrition (group 1) vs. overnutrition (in groups 4 and 5), leading to increasing overweight and obesity (11), which highlights the need to promote dietary changes in many countries of the Global North (18, 28, 29, 44). Divergences in diets are reflected by the differences in carbon footprints between the groups: mean dietary carbon footprints vary from ca. 0.7 kg CO2 eq. per capita per day for certain African countries (all in group 1) to 4 kg CO2 eq. per capita per day for New Zealand, Australia, the United States, France, Austria, Argentina, and Brazil (all in groups 4 and 5) (18). If current crop production used for animal feed and other nonfood uses, such as biofuels (particularly in the United States, China, Western Europe, and Brazil), were used for direct human consumption, ca. 70% more calories would be available, potentially satisfying the basic needs of 4 billion people (26). Our results confirm that most countries with high nutritional quality show high ecological footprints, and therefore, changes in the diets in North America (group 4) and Europe (group 5) would entail the largest reductions in environmental impacts of the global agrifood system (18, 45, 46). The nutrition transition, however, is also affecting the Global South (47, 48). Hence, concerns about nutrient density and public health must be incorporated into considerations around the environmental impacts of food and consequently, integrated into agricultural policies (14, 18, 49). The point at which the higher carbon footprint of some nutrient-dense foods is offset by their higher nutritional value is a priority area for additional research.

Therefore, “doubling the agricultural productivity of small-scale food producers,” as stated by SDG2, is per se not the way to eradicate hunger in a context where the world is already producing food to feed 12 billion people (50, 51). On the contrary, only increasing agricultural productivity has led to reduced prices (ultimately harming farmers) and created health problems and the associated costs (52). It might be even ecologically counterproductive unless other agrifood policies are adopted, such as the reduction of the demand side and particularly, animal-sourced food in the diets of countries in groups 4 and 5 and of some countries in group 3, the diminution of impacts of the supply chain, and the incentive of low-impact and crop-diversifying farming systems (14, 18, 29, 45, 46).

International Agrifood Trade, Food Security, and Environmental Sustainability.

International trade plays an important role in food security (50), and the promotion and maintenance of certain lifestyles and diets rich in calories have been possible thanks to the global food trade (53); SDG2 states that “access to financial services, markets and opportunities to value addition” is needed. However, unless regulated and complemented with other policy instruments, the global agrifood trade entails tradeoffs in terms of social equity and environmental sustainability (54, 55). In fact, securing the food supply through imports occurs only in strong-enough economies (46, 53). In a globalized and financialized system, food tends to flow toward money and power, not toward hunger, and therefore, international agrifood trade can contribute to increase social inequality in the form of food insecurity by facilitating food being exported/traded away from the hungry (51). The results presented here show how certain groups of countries lose (groups 1 and 2) with respect to others (4, 5). In particular, the first group of countries (mostly in Africa) is a clear example of this: despite their large exports of agricultural products and the largest imports of food, undernutrition remains a critical constraint.

More than one-fifth of global calorie production is exported, mostly from countries in group 4 (the United States, Canada, Brazil, and Argentina) (56). Industrialized countries with high GDP per capita tend to be major net importers of biodiversity, while tropical countries, such as Argentina and Brazil, suffer habitat degradation and biodiversity loss as a result of producing crops for exports (57). Land use for export production is responsible for 25% of the projected global extinctions and related biodiversity loss (57, 58), ∼20% of global harvested cropland area is devoted to export production (56), and most of the new cropland expansion is globally attributed to the production of crops for export (59). International food trade has been related to a virtual transfer of water (60), carbon (61), nitrogen (62), and phosphorus (63), while the environmental impacts of agricultural production tend to remain in the producing countries (64).

Mismatch between Regional Biocapacity and Food Security.

There has been a strong decoupling between regional biocapacity and food consumption. The global telecoupling (65) and increasing interdependence among countries in regard to availability and access to food sources and the genetic resources supporting their production (66) result in the increasing reliance of some regions on social and natural resources from other regions of the world, which also increases agrifood debt. A clear example of this phenomenon is land grabbing, which is largely exerted by companies mostly from countries in groups 3 (the Middle East), 2 (e.g., China), and 5 (e.g., Europe) on countries from group 1 (e.g., Africa) (67, 68). For example, restrictions on agricultural production and changes in bioenergy demand have nurtured the dependence of the European Union (group 5) on the appropriation of biological productivity outside its boundaries, with increasing reliance on Latin America as the main supplier (59). Additional evidence in that direction is related to food loss and waste: industrialized Asia (China, India, and North Korea), Europe, North America, and Oceania have the highest per capita carbon footprint of food loss and waste, while sub-Saharan Africa has the lowest (69).

Limitations.

The intricacy and complexity of the currently globalized food system are impossible to fully disentangle with a purely statistical exercise, like the one presented here. For instance, certain dimensions of FSv, such as power relations between different actors within the agrifood system, cannot be easily assessed at a national scale. Nonetheless, new data sources that might contribute to the monitoring of FSv are continuously appearing, such as the Land Matrix project (https://landmatrix.org) for large-scale land acquisitions. A participatory, qualitative assessment with stakeholder collaboration could improve both the selection of indicators and the interpretation of results, for example, to be tailored at the national or regional levels, therefore also improving the usability of the knowledge generated (70). Clustering 150 countries into 5 groups entails that intragroup variance remains large, therefore providing a relevant picture at a global scale but not illustrating regional differences. Moreover, using the country scale as a unit of analysis implies that environmental, social, and nutritional inequities cannot be accounted for. For example, data on ethnic minorities, regional groups, indigenous populations, slum dwellers, and women aged 50 and over are rarely collected (20). A further constraint is the limited quality and/or availability of data: while data on economic indicators are widely available for most countries, data on environmental and social indicators of contested phenomena (e.g., land grabbing) are incomplete and of poor quality (20). However, given the international legitimacy of the data sources (e.g., Food and Agriculture Organization’s Statistical Database [FAOSTAT], World Development Indicators [WDI]) and their broad use in previous scientific research, we are confident that data quality/availability issues do not undermine the main conclusions of this research.

Conclusions

In the last 4 decades, there has been an intense debate about the best policies needed to achieve what is currently stated in SDG2. Between the mid-1960s and the early 2000s, food availability improved globally, and global per capita exports of agricultural products almost doubled, but food self-sufficiency did not change significantly (53). The global population increased 2.5 times between 1961 and 2016, while calorie production increased by more than 4 times by 2013 (71). While the need to increase food production has been repeated like a mantra in many instances (6, 72), old policies focused on productivity, favoring agricultural industrialization, trade liberalization, privatization, and deregulation, have failed to end hunger (73). As this study corroborates, undernutrition is not only a matter of food availability and access (14, 53). Instead, the eradication of hunger would be facilitated by a redistribution of current consumption levels (74, 75). Furthermore, the challenges and responsibility for achieving SDG2 as well as other SDGs, still within planetary boundaries, are not evenly distributed across the globe (6, 20). Providing quantitative agrifood system metrics at the global scale using FSv indicators together with other sociodemographic, social wellbeing, and environmental indicators allows the identification of tradeoffs in environmental and social justice within the global agrifood system and is currently largely ignored (76) but needed in order to comprehensively address SDGs. On a global scale, our results are consistent with previous extensive research and with what we would expect to see if wealthy countries are exporting environmental degradation to import cheap food that is largely wasted and overconsumed: groups 4 and 5 (European and North American countries) should play a prominent role in the transformation of the global agrifood system toward one that is more environmentally sustainable and socially equitable.

There is no fundamental tradeoff between eradicating hunger, achieving environmental sustainability (76), and social equity. However, for the achievement of SDG2 through transformational, socially fair, environmentally sustainable, and resilient food systems, our results point to the following key wedges: biodiversity conservation through environmentally friendly agrifood practices (77), the reduction of agrifood waste (14), the regionalization of food distribution (14), and the adoption of healthy and sustainable diets (26). We need to make these steps fast enough to advance SDG2 while preserving environmental sustainability and social wellbeing.

Materials and Methods

Data Collection.

Ruiz-Almeida and Rivera-Ferre (38) compiled data on 97 indicators for 223 regions (countries or officially recognized territories) between 1961 and 2012 and followed a 6-step methodology to select the most suitable indicators according to data availability, quality, and representativeness of the 6 categories of FSv (more detailed information is provided in SI Appendix, Fig. S1). Indicators were collected from open source databases compiled by international organizations with recognized legitimacy, such as public institutions, agencies, and programs related to the United Nations organization (FAO, United Nations Development Program [UNDP], and United Nations Environmental Programme [UNEP]), international financial institutions (World Bank), and other international organizations (e.g., Organization for Economic Co-operation and Development [OECD] and World Trade Organization [WTO]). Nine indicators were selected to describe the pillar of “access to resources,” 16 were selected for “productive models,” 8 were selected for “commercialization,” 5 were selected for “food security and consumption,” 3 were selected for “agrarian policies and civil society organization,” and 2 were selected for “gender.” For the objectives of this research, we selected indicators following 4 main steps (SI Appendix, Fig. S3): 1) we retrieved the historical data available from all indicators; 2) we selected the indicators for which the last data available referred to the timeframe between 2008 and 2012—for 223 countries or territories, there were 92 indicators that fulfilled these criteria; 3) we debugged the database to reduce the extrapolation of data as much as possible and to avoid outliers due to country or population size; and finally, 4) we incorporated extra data characterizing the bioregional location of the country, the social wellbeing, and the environmental sustainability of each country. The debugging process, which included 4 steps, reduced the initial database to 150 countries and 43 FSv indicators (SI Appendix, Fig. S3). In the final database, 200 missing values (3.1% of the full database) were estimated through the nonlinear iterative partial least squares algorithm. To relate the state of FSv of the groups of countries with their state of social wellbeing and environmental sustainability, further indicator selection was conducted. We scanned more than 200 indicators of social wellbeing and environmental sustainability and selected 17 indicators for environmental sustainability (SI Appendix, Table S3) and 4 indicators for social wellbeing (SI Appendix, Table S4) based on the following criteria: 1) capacity to express the required information, 2) availability of the information for a sufficient number of countries, 3) availability of the information for the selected timeframe (2008 to 2012), and 4) veracity of the data according to internationally legitimized sources. For the description of the demographic and economic contexts of each country, another 7 indicators were selected (SI Appendix, Table S6). For the bioregional characterization of the groups, we used a synthesis of the classification of the world ecoregions (i.e., the 26 categories included 14 terrestrial, 7 inland water, and 5 marine ecoregions) (70) (SI Appendix, Table S7).

Data Analysis.

Tests for the normality of the distribution of all 43 indicators of FSv were performed. Only 1 indicator proved to follow a normal distribution and was standardized (n − 1). The indicators that did not follow a normal distribution or show negative values were transformed through ln(x) or ln(x + 1) when the indicator (x) showed a value equal to 0. The indicators that did not follow a normal distribution and showed negative values were first rescaled (0 to 100) and then transformed through ln(x + 1).

For the identification of groups of countries, we carried out a PCA [covariance (n − 1)] based on the matrix of countries (observations) and indicators (variables). We followed the Kaiser criterion (eigenvalue >1) to determine the significant number of components. To identify possible groups of countries with similar values of FSv indicators, we carried out an HCA based on the Euclidean distance (percentage of distance similarity at a 95% level of confidence) and Ward’s agglomerative method (71) using the standardized coordinates of the most significant factors of the PCA. Finally, to characterize each group resulting from the HCA in terms of ecoregions, socioeconomic characteristics, FSv, social wellbeing, and environmental sustainability indicators, we performed Kruskal–Wallis and χ2 tests. All statistical analyses were run with XLSTAT69.

The characterization of the groups of countries clustered together in the HCA was represented in a world map (Fig. 1) and 2 sets of box plots (Figs. 2 and 3).

Data Availability.

The full database used, with all indicators, is available in Dataset S1.

Footnotes


↵1E.O.-R. and A.R.-A. contributed equally to this work.
↵2To whom correspondence may be addressed. Email: elisa.oterosrozas@gmail.com.


Author contributions: E.O.-R., J.A.G., and M.G.R.-F. designed research; E.O.-R., A.R.-A., M.A., J.A.G., and M.G.R.-F. performed research; E.O.-R. analyzed data; and E.O.-R., A.R.-A., M.A., J.A.G., and M.G.R.-F. wrote the paper.


The authors declare no competing interest.


This article is a PNAS Direct Submission.


This article contains supporting information online at https://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1912710116/-/DCSupplemental.


Published under the PNAS license.

References
↵

J. R. Porter et al., “Food security and food production systems” in Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (Cambridge University Press, Cambridge, UK, 2014), pp. 485–533.
Google Scholar
↵

M. G. Rivera-Ferre,
M. Ortega-Cerdà,
J. Baumgärtner, Rethinking study and management of agricultural systems for policy design. Sustainability 5, 3858–3875 (2013).
Google Scholar
↵

C. Raudsepp-Hearne et al., Untangling the environmentalist’s paradox: Why is human well-being increasing as ecosystem services degrade? Bioscience 60, 576–589 (2010).
CrossRefGoogle Scholar
↵

E. M. Bennett, Changing the agriculture and environment conversation. Nat. Ecol. Evol. 1, 18 (2017).
Google Scholar
↵

J. A. Foley et al., Solutions for a cultivated planet. Nature 478, 337–342 (2011).
CrossRefPubMedGoogle Scholar
↵

P. C. West et al., Leverage points for improving global food security and the environment. Science 345, 325–328 (2014).
Abstract/FREE Full TextGoogle Scholar
↵

P. Smith et al., “Agriculture, forestry and other land use (AFOLU)” in Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, O. Edenhofer et al., Eds. (Cambridge University Press, Cambridge, UK, 2014), pp. 811–922.
Google Scholar
↵

J. Rockström et al., Planetary boundaries: Exploring the safe operating space for humanity. Ecol. Soc. 14, 32 (2009).
Google Scholar
↵

W. Steffen et al., Sustainability. Planetary boundaries: Guiding human development on a changing planet. Science 347, 1259855 (2015).
Abstract/FREE Full TextGoogle Scholar
↵

M. J. Chappell,
L. A. LaValle, Food security and biodiversity: Can we have both? An agroecological analysis. Agric. Human Values 28, 3–26 (2011).
CrossRefGoogle Scholar
↵

FAO, The state of food and agriculture—leveraging food systems for inclusive rural transformation (2017). http://www.fao.org/3/a-i7658e.pdf. Accessed 26 November 2019.
Google Scholar
↵

M. Tester,
P. Langridge, Breeding technologies to increase crop production in a changing world. Science 327, 818–822 (2010).
Abstract/FREE Full TextGoogle Scholar
↵

H. C. J. Godfray et al., Food security: The challenge of feeding 9 billion people. Science 327, 812–818 (2010).
Abstract/FREE Full TextGoogle Scholar
↵

M. E. Schipanski et al., Realizing resilient food systems. Bioscience 66, 600–610 (2016).
CrossRefGoogle Scholar
↵

P. J. Ericksen, Conceptualizing food systems for global environmental change research. Glob. Environ. Change 18, 234–245 (2008).
Google Scholar
↵

V. Vallejo-Rojas,
F. Ravera,
M. G. Rivera-Ferre, Developing an integrated framework to assess agri-food systems and its application in the Ecuadorian Andes. Reg. Environ. Change 16, 2171–2185 (2016).
Google Scholar
↵

J. Loos et al., Putting meaning back into “sustainable intensification.” Front. Ecol. Environ. 12, 356–361 (2014).
CrossRefGoogle Scholar
↵

A. Chaudhary,
D. Gustafson,
A. Mathys, Multi-indicator sustainability assessment of global food systems. Nat. Commun. 9, 848 (2018).
Google Scholar
↵

UN, Sustainable development goals (2016). https://www.un.org/sustainabledevelopment/sustainable-development-goals/. Accessed 8 April 2016.
Google Scholar
↵

R. Swain, “A critical analysis of the sustainable development goals” in Handbook of Sustainability Science and Research, W. L. Filho, Ed. (Springer, 2018), pp. 341–355.
Google Scholar
↵

M. M. L. Lim,
P. Søgaard Jørgensen,
C. A. Wyborn, Reframing the sustainable development goals to achieve sustainable development in the anthropocene—A systems approach. Ecol. Soc. 23, 22 (2018).
Google Scholar
↵

V. Spaiser,
S. Ranganathan,
R. B. Swain,
D. J. T. Sumpter, The sustainable development oxymoron: Quantifying and modelling the incompatibility of sustainable development goals. Int. J. Sustain. Dev. World Ecol. 24, 457–470 (2017).
Google Scholar
↵

FAO, Rome Declaration on world food security and world food summit plan of action (Rome, Italy) (1996). http://www.fao.org/3/w3613e/w3613e00.htm. Accessed 26 November 2019.
Google Scholar
↵

D. Tilman,
C. Balzer,
J. Hill,
B. L. Befort, Global food demand and the sustainable intensification of agriculture. Proc. Natl. Acad. Sci. U.S.A. 108, 20260–20264 (2011).
Abstract/FREE Full TextGoogle Scholar
↵

M. C. Hunter,
R. G. Smith,
M. E. Schipanski,
L. W. Atwood,
D. A. Mortensen, Agriculture in 2050: Recalibrating targets for sustainable intensification. Bioscience 67, 386–391 (2017).
Google Scholar
↵

S. J. Vermeulen,
B. M. Campbell,
J. S. I. Ingram, Climate change and food systems. Annu. Rev. Environ. Resour. 37, 195–222 (2012).
CrossRefGoogle Scholar
↵

B. M. Campbell et al., Urgent action to combat climate change and its impacts (SDG 13): Transforming agriculture and food systems. Curr. Opin. Environ. Sustain 34, 13–20 (2018).
Google Scholar
↵

W. Willett et al., Food in the anthropocene: The EAT-lancet commission on healthy diets from sustainable food systems. Lancet 393, 447–492 (2019).
CrossRefPubMedGoogle Scholar
↵

J. Poore,
T. Nemecek, Reducing food’s environmental impacts through producers and consumers. Science 360, 987–992 (2018).
Abstract/FREE Full TextGoogle Scholar
↵

M. Springmann,
H. C. J. Godfray,
M. Rayner,
P. Scarborough, Analysis and valuation of the health and climate change cobenefits of dietary change. Proc. Natl. Acad. Sci. U.S.A. 113, 4146–4151 (2016).
Abstract/FREE Full TextGoogle Scholar
↵

D. Gustafson et al., Seven food system metrics of sustainable nutrition security. Sustainability 8, 196 (2016).
Google Scholar
↵

M. Nilsson et al., Mapping interactions between the sustainable development goals: Lessons learned and ways forward. Sustain. Sci. 13, 1489–1503 (2018).
PubMedGoogle Scholar
↵

FAO, Thirty-second FAO Regional Conference for Latin America and the Caribbean (Buenos Aires, Argentina) (2012). http://www.fao.org/3/mj764e/mj764e.pdf. Accessed 26 November 2019.
Google Scholar
↵

P. Claeys, Human Rights and the Food Sovereignty Movement: Reclaiming Control (Routledge, London, UK, 2015).
Google Scholar
↵

L. Levidow,
M. Pimbert,
G. Vanloqueren, Agroecological research: Conforming - or transforming the dominant agro-food regime? Agroecol. Sustain. Food Syst. 38, 1127–1155 (2014).
Google Scholar
↵

O. De Schutter, Report of the special rapporteur on the right to food (A/HRC/25/57, New York) (2014). http://www.ohchr.org/EN/HRBodies/HRC/RegularSessions/Session25/Documents/A_HRC_25_57_ENG.DOC. Accessed 26 November 2019.
Google Scholar
↵

R. Patel, Food sovereignty. J. Peasant Stud. 36, 663–706 (2009).
CrossRefGoogle Scholar
↵

A. Ruiz-Almeida,
M. G. Rivera-Ferre, Internationally-based indicators to measure agri-food systems sustainability using food sovereignty as a conceptual framework. Food Syst, doi:10.1007/s12571-019-00964-5 (2019).
CrossRefGoogle Scholar
↵

ICSU ISSC, Review of Targets for the Sustainable Development Goals: The Science Perspective (International Science Council, Paris, France, 2015).
Google Scholar
↵

D. I. Stern,
M. S. Common,
E. B. Barbier, Econonmic growth and environmental degredation: A critique of the environmental Kuznets Curve. Univ York 44, 24 (1994).
Google Scholar
↵

C. J. A. Bradshaw,
X. Giam,
N. S. Sodhi, Evaluating the relative environmental impact of countries. PLoS One 5, e10440 (2010).
CrossRefPubMedGoogle Scholar
↵

G. Goeminne,
E. Paredis, The concept of ecological debt: Some steps towards an enriched sustainability paradigm. Environ. Dev. Sustain. 12, 691–712 (2010).
Google Scholar
↵

A. Sen, Poverty and Famines: An Essay on Entitlement and Deprivation (Clarendon Press, Oxford, UK, 1981).
Google Scholar
↵

J. Dixon et al., The health equity dimensions of urban food systems. J. Urban Health 84 (suppl. 3), i118–i129 (2007).
CrossRefPubMedGoogle Scholar
↵

B. Bajželj et al., Importance of food-demand management for climate mitigation. Nat. Clim. Chang. 4, 924–929 (2014).
CrossRefGoogle Scholar
↵

M. Berners-Lee,
C. Kennelly,
R. Watson,
C. N. Hewitt, Current global food production is sufficient to meet human nutritional needs in 2050 provided there is radical societal adaptation. Elem. Sci. Anth. 6, 52 (2018).
Google Scholar
↵

P. Amuna,
F. B. Zotor, Epidemiological and nutrition transition in developing countries: Impact on human health and development. Proc. Nutr. Soc. 67, 82–90 (2008).
CrossRefPubMedGoogle Scholar
↵

N. P. Steyn,
Z. J. McHiza, Obesity and the nutrition transition in Sub-Saharan Africa. Ann. N. Y. Acad. Sci. 1311, 88–101 (2014).
CrossRefPubMedGoogle Scholar
↵

A. Drewnowski et al., Energy and nutrient density of foods in relation to their carbon footprint. Am. J. Clin. Nutr. 101, 184–191 (2015).
Abstract/FREE Full TextGoogle Scholar
↵

P. D’Odorico,
J. A. Carr,
F. Laio,
L. Ridolfi,
S. Vandoni, Feeding humanity through global food trade Earth’s Future. Earths Futur. 2, 458–469 (2014).
Google Scholar
↵

J. Ziegler, Betting on Femine: Why the World Still Goes Hungry (New Press, New York, NY, 2013).
Google Scholar
↵

T. G. Benton,
R. Bailey, The paradox of productivity: Agricultural productivity promotes food system inefficiency. Glob. Sustain. 2, e6 (2019).
Google Scholar
↵

M. Porkka,
M. Kummu,
S. Siebert,
O. Varis, From food insufficiency towards trade dependency: A historical analysis of global food availability. PLoS One 8, e82714 (2013).
CrossRefPubMedGoogle Scholar
↵

J. Sun et al., Importing food damages domestic environment: Evidence from global soybean trade. Proc. Natl. Acad. Sci. U.S.A. 115, 5415–5419 (2018).
Abstract/FREE Full TextGoogle Scholar
↵

M. Davis, Late Victorian holocausts: El Niño famines and the making of the Third World (Verso) (2001). https://books.google.es/books?id=2LjYUY3LHkoC&printsec=rontcover&hl=es&source=gbs_ge_summary_r&cad=0#v=onepage&q&f=false. Accessed 22 October 2019.
Google Scholar
↵

G. K. MacDonald et al., Rethinking agricultural trade relationships in an era of globalization. Bioscience 65, 275–289 (2015).
CrossRefGoogle Scholar
↵

A. Chaudhary,
T. Kastner, Land use biodiversity impacts embodied in international food trade. Glob. Environ. Change 38, 195–204 (2016).
Google Scholar
↵

A. Chaudhary,
T. M. Brooks, National consumption and global trade impacts on biodiversity. World Dev., doi:10.1016/j.worlddev.2017.10.012 (2017).
CrossRefGoogle Scholar
↵

T. Kastner et al., Cropland area embodied in international trade: Contradictory results from different approaches. Ecol. Econ. 104, 140–144 (2014).
Google Scholar
↵

J. A. Allan, Virtual water: A strategic resource global solutions to regional deficits. Gr. Water 36, 545–546 (1998).
Google Scholar
↵

M. E. Schipanski,
E. M. Bennett, The influence of agricultural trade and livestock production on the global phosphorus cycle. Ecosystems (N. Y.) 15, 256–268 (2012).
Google Scholar
↵

J. N. Galloway et al., International trade in meat: The tip of the pork chop. Ambio 36, 622–629 (2007).
CrossRefPubMedGoogle Scholar
↵

T. Kastner,
K.-H. Erb,
S. Nonhebel, International wood trade and forest change: A global analysis. Glob. Environ. Change 21, 947–956 (2011).
Google Scholar
↵

P. Meyfroidt,
E. F. Lambin,
K. H. Erb,
T. W. Hertel, Globalization of land use: Distant drivers of land change and geographic displacement of land use. Curr. Opin. Environ. Sustain. 5, 438–444 (2013).
CrossRefGoogle Scholar
↵

J. Liu,
W. Yang,
S. Li, Framing ecosystem services in the telecoupled Anthropocene. Front. Ecol. Environ. 14, 27–36 (2016).
Google Scholar
↵

C. K. Khoury et al., Increasing homogeneity in global food supplies and the implications for food security. Proc. Natl. Acad. Sci. U.S.A. 111, 4001–4006 (2014).
Abstract/FREE Full TextGoogle Scholar
↵

W. Anseeuw, The rush for land in Africa: Resource grabbing or green revolution? South African J Int Aff 20, 159–177 (2013).
Google Scholar
↵

S. Batterbury,
F. Ndi, “Land-grabbing in Africa” in The Routledge Handbook of African Development, J. A. Binns, K. Lynch, E. Nel, Eds. (Routledge, London, UK, 2018), pp. 573–582.
Google Scholar
↵

FAO, The State of Food Insecurity in the World 2013. http://www.fao.org/publications/sofi/2013/en/. Accessed 26 November 2019.
Google Scholar
↵

W. C. Clark,
L. van Kerkhoff,
L. Lebel,
G. C. Gallopin, Crafting usable knowledge for sustainable development. Proc. Natl. Acad. Sci. U.S.A. 113, 4570–4578 (2016).
Abstract/FREE Full TextGoogle Scholar
↵

N. McKeon, Food Security Governance: Empowering Communities, Regulating Corporations (Routledge, Abingdon, UK, 2015).
Google Scholar
↵

FAO, Global Agriculture towards 2050. High Lev Expert Forum-How to Feed World 2050 (Food and Agriculture Organization, Rome, Italy, 2009), pp. 1–4.
Google Scholar
↵

R. Burdock,
P. Ampt, Food sovereignty: The case and the space for community led agricultural autonomy within the global strategic framework for food security and nutrition. J. Agric. Sci. 9, 1–18 (2017).
Google Scholar
↵

C. M. Pittelkow et al., Productivity limits and potentials of the principles of conservation agriculture. Nature 517, 365–368 (2015).
PubMedGoogle Scholar
↵

W. Easterly, The trouble with sustainable development goals. Current History 114, 322–324 (2015).
Google Scholar
↵

D. P. van Vuuren et al., Pathways to achieve a set of ambitious global sustainability objectives by 2050: Explorations using the IMAGE integrated assessment model. Technol. Forecast. Soc. Change 98, 303–323 (2015).
Google Scholar
↵

M. Altieri, Agroecology, small farms, and food sovereignty. Mon. Rev. 61, 102–113 (2009).
Google Scholar