The Global Food
System: An Analysis

March 2017 | Report

By Eva Gladek, Gerard Roemers, Oscar Sabag Muños, Erin Kennedy, Matthew Fraser, and Peter Hirsh

Commissioned by WWF Netherlands

Executive Summary

When the demand for water within a defined time period cannot be met with locally available resources, either due to lack of water availability or poor water quality, the region where this is occurring is said to be experiencing “water stress.”  Regional and temporal variability in water availability explain why many regions of the world are water stressed. It is estimated that, globally, more than 2.5 billion people live in water stressed areas (GrowingBlue, 2015). Generally such situations arise due to the increase in population and economic growth leading to an increased water usage by municipalities and different economic sectors. 

About 70% of the planet’s accessible fresh water withdrawals are currently used for agricultural activities, more than twice that used by industry (23%), and dwarfing municipal use (8%). Agriculture consumed over 8.300 billion m3 of water per year over the period 1996-2005, representing 92% of total global fresh water use (Hoekstra et al., 2012).

The current structure of the food system lies at the center of a nexus of global problems, stretching from poverty to environmental degradation. The increase in food production needed to meet the anticipated demands of the near future cannot be achieved by simply extrapolating current trends in production and consumption. A continuation of the recent historical trends of expansion and intensification will undermine the very resource base on which the food system itself depends.

The preservation of ecosystems and the future wellbeing of the human population are all centrally dependent on a structural transformation of the food system towards a sustainable and resilient state.

The current food system is the product of a historic development pathway

Global food and agricultural production have increased significantly since the end of WWII spurred by a combination of population and economic growth along with technological and cultural shifts in production practices. Due to increases in population, wealth, and urbanization, the world has seen an overall increase in food demand, coupled with a shift in dietary preferences towards more resource-intensive foods.

The Green Revolution played a significant role in establishing intensive agricultural production methods globally and shaping the reigning philosophies in mainstream agricultural practice. Global yields have steadily increased since the 1950s; there is more food produced today per person than ever recorded. Though widely credited with helping avert anticipated large-scale food shortages in the post-WWII era, the intensification practices brought on by the Green Revolution have also been critiqued for driving ecological degradation, unsustainable resource consumption, and entrenching dependency on non- renewable resources like fossil fuels.

Intensification, consolidation, and specialisation are some of the large scale behavioural trends inherent to the food system. Intensive practices dominate the system as a whole and a small number of actors in the fields of production, processing and retail control most of the food system and strongly influence policy making. Loopholes in trade agreements are widely abused by more powerful nations, resulting in unfair competition for developing countries, ultimately manufacturing dependence and eroding local food security.

Recent trends and policies towards growing non-food crops, like biofuels and biomaterials, are leading to re-assignment of land and other base resources, resulting in less availability of these resources for food production. Funding for agricultural research and development is mostly available in higher- income nations, leaving lower-income nations behind. Research and development efforts have been focused on enhancing conventional production methods, with very little funding allocated to the development of sustainable agricultural techniques.

The food system is the largest contributor to both environmental and humanitarian impacts

Agriculture now occupies roughly half of the plant-habitable surface of the planet, uses 69% of extracted fresh water and, together with the rest of the food system, is responsible for 25 – 30% of greenhouse gas emissions. The expansion of industrial fishing fleets and a higher demand for seafood globally have led to the collapse or total exploitation of over 90% of the world’s marine fisheries. A growing demand for land-based animal products is the primary driver of tropical deforestation. Through its direct and intermediate impacts, the food system is the largest contributor to the depletion of biodiversity.

The agri-food sector is the world’s largest economic sector and is therefore deeply entwined with poverty. Half the global workforce is employed in agriculture. A majority of the world’s poorest people are subsistence farmers and fishermen. Small farmers and fishers around the world are caught in cycles of poverty, without access to education, employment, economic and social infrastructure, and political representation. Many do not receive adequate compensation, work in unacceptable conditions, or do not have access to sufficient, affordable, or proper-quality food. Poverty is the largest threat to producers of food globally and the largest driver of food insecurity.

However, simply ensuring a sufficient level of food production will not address the more entrenched impacts and humanitarian imbalances within the food system. We currently produce more than enough food for the global population, yet over 795 million people remain undernourished.

Increased population and growing wealth suggest that a doubling of food production may be necessary by 2050

Though its environmental and humanitarian impacts are already severe, the food system is poised for further expansion. In 2012, the Food and Agriculture Organization of the United Nations (FAO) estimated that by 2050 we will need to increase food output by 60% based on a business-as-usual scenario. Since the FAO’s projections, population increases have been further revised upwards and food demand is likely to double. This represents a larger increase from today’s production than we have seen since the 1960s. Past concerns about the scalability of global food supply have historically been laid to rest by a continuous increase in output through intensification, but recent trends have renewed concerns about the continuity of global food supply in the coming decades. The genetic potential of major crops is being reached, land is being degraded, and there is a structural lack of investment in low-producing regions. These combined issues have led to a lower rate of growth in yields in recent decades; yield increases are not currently on track to meet projected increases in demand. This situation drives policy-makers and researchers to redouble their efforts on further advancing the intensive practices that led to dramatic increases in yields in recent decades.

The planetary boundaries and unsustainable resource extraction are hard limits to the food system’s further expansion based on past trends

The FAO’s 2012 global food projections study concluded that sufficient global land, water, and fertiliser resources exist to supply the 2050 projected global food demand, though with difficulty due to emerging scarcity. Even so, these conclusions are based primarily on the physical availability of basic resources and do not take into account the transgressions of planetary boundaries.

Four planetary boundaries have already been transgressed; biospheric integrity, the biogeochemical cycles of nitrogen and phosphorus, and climate change. Biospheric integrity is an apex boundary that is further breached when any of the other boundaries are impacted. The extraction of biological resources accounts for around 21% of the total material extraction by mass globally, but is responsible for a disproportionate majority of impacts that relate to planetary boundary transgressions. A majority of biological resource extraction can be attributed to the food system, making it the primary single contributor to the transgression of many planetary boundaries.

In addition to the planetary boundaries, a second set of limits to the expansion of the food system is the depletion of non-renewable or slowly renewable resources, such as fossil fuels and wild fish stocks. From our survey of impacts stemming from the global food system, we conclude that pursuing a growth and intensification trajectory is untenable as the main strategy for addressing the projected food demands of the 2050 population. Moreover, this pathway will only provide temporary solutions at the expense of long-term productive capacity due to, for example, the erosion and salinisation of soils.

Alternative pathways can provide for the needs of our growing population without compromising human or ecological health

The growth and intensification pathway is not the inevitable choice for addressing the 2050 food demands of the population. Over 30% of food is currently wasted; a larger percentage of the population is now overweight than undernourished; land resources are increasingly allocated towards non-food uses; nutritious diets can be provided with a fraction of the average resource demand that they currently require. All of these systemic failures present opportunities for transitioning the food system in a direction where it provides fully for the needs of people without infringing on key limits.

A counter-movement to intensive, conventional agricultural and extractive systems is slowly emerging. These practices still only make up a minority of the global agricultural production and are generally under- researched. New practices and food processing techniques present a small, but promising, new direction for innovations in the food system. We can produce sufficient food, even for a much larger population, if structural changes are made to how we approach both production and consumption.

To successfully move towards a sustainable and resilient food system, we must consider the systemic nature of the system’s behaviours and impacts. Severe, irreversible and non-linear impacts that may lead to the crossing of key systemic tipping points should be avoided at highest cost. These include impacts in areas of preservation of global biodiversity, mitigation of climate change, management of soils and essential non-renewable resources, the preservation of culture and heritage, and the preservation of human health. If we do not address and change the central root causes that lead to multiple impacts, impacts will continue to occur. To ensure that solutions are comprehensive and adaptive, we need to hard-wire systems thinking into the food policy. By accounting for systemic effects, we can come to understand feedback loops and adverse effects early on and adapt policy accordingly.

Making food policy decisions for the global food system requires stronger and more cooperative international governance. Many impacts in the food system today can be traced back to a structural limitation of governance and enforcement.

We need to address four main challenges simultaneously in order to transition to a sustainable and resilient food system

Challenge 1: Adaptive and Resilient Food System

An adaptive and resilient food system is one that will be able to respond to changing circumstances and new challenges as they emerge. This is one of the most important systemic criteria for a sustainable food system, since we cannot predict all of the conditions or changes that will emerge in the future. Adaptive capacity and resilience must be built into both biophysical aspects of the system (through the preservation of biodiversity, maintenance of healthy soil systems, maintenance of buffering capacity in water bodies, etc.) and socioeconomic aspects of the system (knowledge transfer, development or organizational capacity, elimination of poverty cycles, etc.).

Challenge 2: Nutritious Food For All

The most basic and fundamental challenge that the food system must address is to ensure the supply of adequate nutrition for the world’s population. Ideally, it should achieve the objective set out by the World Food Summit in Rome, which states that food security is addressed when, “all people, at all times, have physical and economic access to sufficient, safe, and nutritious food to meet their dietary needs and food preferences for an active and healthy life.”

Some of the priority objectives for addressing this challenge should, at minimum, include: reducing overall food demand (e.g., through reducing food waste); progressively shifting to lower-impact, less-resource- intensive food sources; ensuring that scarce resources (land, water) are allocated to food production as a priority over non-food uses; improving economic access to food; and improving farmer productivity in the developing world.

Challenge 3: Within Planetary Boundaries

A sustainable food system should remain within planetary boundaries in all of the key biophysical impact areas across the entire life cycle of food production, consumption, and disposal. Though we should continuously strive for full net zero impact within the food system, there are some areas, such as preservation of biodiversity, which should be prioritized over others. In general, severe and irreversible impacts to complex ecological and cultural systems, and the depletion of non-renewable natural resources caused by the food system, should be addressed with the highest urgency.

Many of the approaches that are necessary to address Challenges 1 and 2 are also essential for bringing the operations of the food system within the scope of the planetary boundaries. Notably, reducing food demand and shifting to lower-impact sources of food are critical prerequisites for bringing down the overall resource throughput of the system. In addition, this challenge requires at least the following measures: reducing the impact of existing agricultural and extractive practices (e.g., applying conservation measures, moving to lower-impact fishing techniques); Placing limits on system expansion and intensification, particularly when addressing the global yield gap (e.g., reducing arable land expansion, and if necessary directing it towards marginal lands); and investing in the development of new sustainable agricultural techniques (e.g., organic cultivars, agro-ecological practices).

Challenge 4: Supporting Livelihoods and Wellbeing

The food system should structurally support the livelihoods and well-being of people working within it. It should be possible to fully nourish and support oneself and earn a reasonable living wage in exchange for average work hours within the food system.

Ensuring that the food system supports livelihoods and wellbeing is more than an end in itself; it is also essential for addressing the other three challenges. Without secure livelihoods, smallholder farmers and fishermen will continue to struggle in building the necessary capacity and resource base to transition to sustainable models of production. A resilient system cannot be built upon an unstable foundation. Therefore, addressing the systemic structures that perpetuate poverty is critical to the success of achieving a sustainable and resilient food system.


table of contents

Further reading

Read our project summary for a brief overview of this report, or our article on the Four pillars of a sustainable food system for key takeaways.


Header image by Shubert Ciencia 


The global food system is in need of a dramatic  transformation. The pathway we are currently on is  leading to an impasse: the increases in food production  needed to meet the anticipated demands of a much larger  and wealthier human population cannot be achieved by  simply extrapolating current trends in production and  consumption.

Can we achieve a food system that works within the  planet’s biophysical boundaries, inclusively supports  human livelihoods, and ensures food security for a  growing and changing population? This has become one  of the central questions in humanity’s broader quest to  shape a sustainable future. 

The Dilemma

In the 8 – 10,000 years of practicing agriculture (Harlan  & MacNeish, 1994), only a small fraction of the 200,000  years that modern humans are estimated to have existed  (Harpending & Eswaran, 2005), food production has  altered our environment more dramatically than any  other socioeconomic activity. Agriculture now occupies  roughly half of the plant-habitable surface of the planet  (FAO, 2015b), uses 69% of extracted fresh water (Aquastat,  2014), and, together with the rest of the food chain, is  responsible for between 25 – 30% of global greenhouse  gas emissions (IPCC, 2014). The expansion of industrial fishing fleets and an increased global appetite for seafood  have led to the collapse or total exploitation of 90% of the  world’s marine fisheries (FAO, 2014b).

Likewise, a growing demand for land-based animal  products is the primary driver of tropical deforestation  (Convention on Biological Diversity, 2015). Through  its myriad direct and intermediate impacts, the food  system is the single largest contributor to the depletion  of our most precious non-renewable resource: global  biodiversity (see section 3.1).

Though its environmental impacts are already severe, the food system, which we define as the complete set of people, institutions, activities, processes, and infrastructure involved in producing and consuming food for a given population, is poised for a necessary expansion.

In 2012, the Food and Agriculture Organization of the United Nations estimated that by 2050 we will need to increase food output by 60% based on a business-as- usual scenario. Since the FAO’s projections, population increases have been further revised upwards and the food demand is likely to need to double (United Nations, 2015). This represents a larger increase from today’s production levels than we have achieved through advances of the Green Revolution since the 1960s (Searchinger et al., 2013).

The food system is the single largest contributor to the depletion of global biodiversity.

Structural challenges

Simply ensuring a sufficient level of food production, however, does not address some of the more entrenched impacts and humanitarian imbalances in the current food system. We currently produce more than enough food for the global population, yet despite this fact, over 795 million people remain food insecure.

On the other side of the spectrum, in 2014, the number of overweight people reached 1.9 billion, with over 600 million obese (World Health Organization (WHO), 2015). Due to a combination of poverty, lack of education, and evolving commercial practices in the food industry, there is an increasing emergence of “double burden” families that have members who are both overweight and malnourished (World Health Organization (WHO), 2015).

As the world’s largest economic sector, the agri-food system is also deeply entwined with the issue of global poverty. Half of the global workforce (1.3 billion people) are employed in agriculture, with an estimated 2.6 billion deriving their primary livelihoods from it (International Labour Organization (ILO), 2015). A majority of the world’s poorest people are subsistence farmers and fishermen, whose basic livelihoods continue to be threatened by structural poverty traps (Carter & Barrett, 2006).

It is clear that ensuring adequate food globally, though critical, is just one piece of a much more complex puzzle. The current structure of the global food system lies at the centre of a nexus of global problems stretching from poverty to environmental degradation.

Breaking the pattern

The dilemma of the global food system is a deeply existential one. On the one hand, we have a moral imperative to ensure an uninterrupted food supply, on the other, doing so based on the expansion of current practices will have devastating consequences for our natural environment, undermining the very basis of the food system’s functioning. Most of the solutions proposed to resolve this dilemma focus on the expansion of arable lands and the increase of yields per hectare through the intensification of agricultural production. There is good reason to question whether or not this approach, which in many ways represents a continuation of existing trends, will result in a food system that sufficiently resolves the nexus of problems we face: 

  • Universal food security has not been achieved despite the fact that food production levels are sufficient to feed everyone globally; 10.8% of the global population remains food insecure despite a global surplus in caloric production of over 20% (Marx, 2015; authors’ estimates based on FAOSTAT data).
  • The global nutrient cycles of nitrogen and phosphorous are broken, not only because of practices in agriculture, but to an equally large extent through the lack of collection of nutrients from municipal waste water systems (Vitousek et al., 1997).
  • Production practices are evaluated based primarily on short-term increases in yields, rather than on their ability to sustain long-term productive output based on care for soils, appropriate labour systems, and the need for adaptation to the effects of climate change (Phelps, Carrasco, Webb, Koh, & Pascual, 2013).
  • Despite clear indications that allocating arable land use to the production of first generation biofuels is not a good use of resources by almost any measure, policies remain in place to continue this trend (Bastos Lima & Gupta, 2014).
  • Around one third of food globally is wasted, indicating large potential gains for reducing impact and saving scarce resources (Gustavsson, Cederberg, & Sonesson, 2011).
  • The very structure of global food markets and trade continues to keep individuals trapped in poverty and threatens local food access in developing countries (Serpukhov, 2013).

As the food system has expanded over the past decades, many of these concerns have come into sharper focus rather than becoming resolved. This observation points to the fact that more effective and durable solutions to achieving a sustainable and resilient food future may lie in deeper parts of the system: in its very structure and the underlying incentives that lead to continued problematic outcomes.

Two women vendors in a Chinese street market, Creative Commons: thisnomad 

New pathways

Food is a daily necessity, a carrier of our cultural values, family traditions, and even personal ideologies. The very discussion of the challenge of the food system is often framed politically, as a battle between the needs of humans versus the needs of the environment. Discussions about organic agriculture or Genetically Modified Organisms (GMOs) are almost never merely about technological efficiency; they touch on several polarizing debates around people’s identities, ethics, and views of the world.

We need a multitude of strategies at different levels of the food systems functioning that go beyond individual convictions in order to address the urgent challenges at hand. To that end, it is essential to take an objective look at the data and look beyond the well-worn pathways of argumentation.

This report presents a baseline analysis of the global food system using methodologies taken from systems science. One of our primary objectives is to present a clear overview of the current performance of the global food system: its inputs, outputs, impacts, structure, and behaviour. With this factual basis, we hope to lay the foundation for further in-depth analysis, and inform a deeper and broader look at the potential systemic approaches for transitioning towards a truly sustainable, resilient food system.

The inevitability of an expansion of food production based on current business as usual models is far from a closed question; a coordinated effort between policy makers, knowledge institutes, producers, financial institutions, and consumers is needed to shape a new, coherent pathway forward.

Reader's Guide

This report has five main chapters, each focused on answering specific questions regarding the food system. The first four chapters of the report provide an overview of the current state of the food system, its behaviours and global trends, the impacts and challenges associated with it, and the structural causes underlying these features. In the fifth chapter, we present an outlook for a sustainable and resilient food system.


1. Current State

The first chapter of the report provides a first, broad look into the food system, following food as it moves ‘from farm to fork.’  The data presented in this chapter form the basis for the analyses that follow in subsequent chapters.  The chapter is structured to sequentially address all major phases of the food production chain. The  chapter begins with an overview of global production of food crops, livestock, and seafood; the resource demands of this production; and the techniques and practices implemented in the productive and extractive sectors.  In the following sections we present data on the food processing industry, global trade in food commodities, and food sales. Global consumption patterns and quantities, as well as food waste along the food chain are discussed. 


2. Behaviours and trends

Reflecting on the overview of the current state of the global food system, we present a high level look at some  of the main trends and emergent behaviours  that characterise the system. We further elaborate on how the food system is evolving and some of the broader implications for its future trajectory.


3. Impacts

The food system is associated with a range of biophysical and humanitarian impacts; these are discussed in more detail in this third chapter of the report. This chapter provides insight in the magnitude of these impacts as well as their key drivers. The discussion that follows examines the  impact-based limits to the further expansion of the food system under its historic model of development and suggests a systemic approach for considering how to holistically address these impacts in policy and strategic development.


4. Structural Causes

This chapter uses an analytical framework, Root Cause Analysis, to identify the structural causes that drive the system to its current negative impacts and behaviours. In this chapter, we provide a deeper layer of insight than in the impacts chapter, since we seek to identify not only the direct causes  of these impacts, but also the underlying structures (trade architecture) and self-reinforcing mechanisms  (the poverty trap) that keep these impacts in place. These underlying structures are the targets to address in order to tackle the abuses and problems that characterise the system in a lasting manner.


5. Towards a Sustainable and Resilient Food System

This chapter outlines an outlook for a truly sustainable food system. This outlook is sketched by outlining the changes necessary with regards to the biophysical and humanitarian impacts of the current food system identified in chapter 3. These performance areas are then grouped into four over-arching categories or “challenges” that a sustainable food system should address. 

Conceptual Framework

Figure 1. The conceptual framework used in this systems analysis. In this graphic, “emergent behaviour” is not intended to accurately depict the actual interactions between actors, nor how this behaviour affects actors outside of the food system. Click to expand

To change outcomes, identify behaviours and change structures

The behaviour or functioning of complex socio-ecological system, such as the food system, is difficult to predict. This is because the functioning of the system arises from the collective behaviour of a large number of actors (e.g. farmers, fishermen, multinational companies, and consumers), while in return the behaviour of each of these actors is influenced the structure of the food system and the behaviour of other actors. 

Farming practices are a case in point. When a farmer decides what crops to cultivate, and how to cultivate them, he or she will make this decision based on for example the local climate and soil conditions (which are part of the biophysical structure of the food system), or available subsidies (which are the result of another actor’s behaviour, in this case probably an (intra)national government. In turn the actions of the farmer have an influence on the biophysical structure of the food system: for example, when fossil fuels are used for agricultural machinery farming can, in the long run, influence local and global climate conditions. 

Systems theory proposes that the structures of a system give rise to behaviours, which are in turn the drivers behind system impacts. The figure above illustrates some of the most important system structures in the food system: biophysical elements, the people and organisations in the system, and economic, governance, and social structures. Specific actors, such as farmers, or consumers, interact with these structures; from the collective action of all these actors, a certain state of the system emerges. The systems state can be observed by looking at certain biophysical or humanitarian impacts, such as biodiversity loss. 

Ultimately, the state of the food system is the result of the behaviour of many different actors, who interact with many different parts of the systems structure. Therefore, researching the food system from the perspective of systems thinking, we focus precisely on these interrelationships.  Our approach takes a holistic lens that understands the system as a dynamic whole, rather than looking at certain parts of the system in isolation. This way we avoid one-dimensional solutions, which may solve one problem while triggering another, and instead come up with a set of holistic strategies for a truly sustainable food system.

Chapter 1

Current state

1.0 Introduction

The food system is both enormous and complex. The trend of globalization has intensified the level of interdependency between its actors and processes over the last half century, leading to an increasingly “global” system in the true sense of the word. The full scope of the food system stretches to include the vast majority of the human population (as either producers, traders, or consumers), the majority of all economic activities, and a large proportion of many categories of resource use.A wealth of data is collected annually on the performance of the global food system by intergovernmental organisations such as the Food and Agriculture Organization of the United Nations (FAO), national and local governments, non-governmental organisations (NGOs), and a variety of research and academic institutions. Statistics collected cover everything from agricultural yields and regional availability of tractors to trade balances and malnutrition rates. In this chapter, we explore the current state of the global food system through the lens of some of its core processes: production and extraction, processing, trade, retail, consumption, and waste. We present key statistics along each of the steps of this chain, which will serve as the basis for further interpretation and analysis in later parts of the report and in the follow up studies to this work. Understanding the basic nature of the resource flows and production practices in the food system is an essential prerequisite to gaining insight into the problems at hand.

Key messages:

  • Currently 30 major crops account for 90 to 95% of human food consumption (United Nations Environmental Programme, 2007). Cereal production occupies the largest percentage of cultivated land, accounting for almost half of total cultivated area, followed by oil crops, which occupy almost one fifth.
  • Of the 1.5 billion hectares of agricultural land worldwide, only a third is used for the production of food crops. The remainder is primarily dedicated to the production of livestock. Because 38% of global crops are used as feed for animals, only 20% of global agricultural land is utilized for the direct production of crops for human consumption (FAO, 2015b).
  • Fish provide 4.3 billion people with around 15 percent of their animal protein intake (FAO, 2014b). The global fisheries and aquaculture sector produced over 176 million tonnes of seafood in 2011 (FAO, 2015b). Although the production of fish, seafood, and algae is still dominated by extractive wild capture fisheries, global aquaculture (aquatic farming) has more than doubled since the start of the millennium, and is positioned to become the primary contributor to seafood production in the near future.
  • The production of food is dominated by East Asia, Latin America, and Europe; between them, these regions produce over half of the world’s food supply. »Contrary to popular expectations originating from topics like “food miles” and import dependencies, the amount of international trade is relatively insignificant compared to total volumes of production (14% of total annual production), though some commodities, like coffee, are outliers in this regard.
  • There is enormous variability in global agricultural production and wild extraction systems. The type of practice selected is one of the main determinants of resource demand and yield, and by extension, environmental impact

1.1 What is the food system?

The food system can be defined as the complete set of people, institutions, activities, processes, and infrastructure involved in producing and consuming food for a given population. Specifically, food-system-related activities include: growing, harvesting, processing, packaging, transporting, marketing, selling, cooking, consumption, and disposal of food and any food-related items. Also included are any inputs needed (land, agricultural chemicals, labour, water, machinery, knowledge, capital) and outputs generated apart from food (greenhouse gas emissions, agricultural wastes, municipal wastewater) at each step along this chain. The food system further encompasses the public officials, civic organisations, educators, researchers, and all other parties that influence it through policies, regulations, or programmes. On the highest, most abstract level, the food system includes the frameworks, belief systems, and paradigms that define its rules and invisibly control its functioning.

Geographical system boundaries

Though the world can be said to have a multitude of smaller-scale food systems that serve local communities or regional populations, the last century has seen the progressive emergence of a global food system that has effectively linked disparate geographic regions into an interdependent structure. Though different activities within the food system are highly dependent on local contextual factors and the severity of key impacts is likewise determined on different scales (for example, water scarcity), the central drivers of the system’s behaviour are more centrally dependent on the dynamics of the global system.

Functional system boundaries

The function of the food system can be defined as transferring energy and materials into organic components, which provide human beings with the bio-available energy and key physical nutrients they need in order to function. Despite the range of important secondary functions fulfilled by the food system, such as education, employment, and maintenance of cultural systems, minimally reduced, the primary function of the system remains the delivery of food to people.

In our research, we have specifically focused on products for food uses, and have only given attention to products for non-food uses (such as fibre, fuel, pharmaceuticals, and chemicals) insofar as they compete for the same systemic resources as required by food production (land, water, energy, labour). While we consider wild harvest of plants and non-fish seafood as part of the scope of the food system, the availability of data on these activities is scarce, and therefore is not covered explicitly in this report.

We have also delineated the boundary of the system to exclude the full impact of adjacent supply chains (e.g., petrochemicals, machinery, cooking fuel, etc.). In calculating the impacts of the food system, we have taken into account the impact of direct inputs (such as fuel and agricultural chemicals), but not the impacts of the broader supply chains that are responsible for producing those inputs.

Tractors and other farming machinery are among the metrics for evaluating intensification.
Creative Commons: Wikimedia

1.2 Global food production

The food we eat daily is the final product of the world’s largest production line: the global agri-food complex. In this section we provide a snapshot of the volume of food produced annually using the planet’s land and water resources (for the reference year 2011). As shown in Figure 2, about 1.5 billion hectares of land are used for crop production (arable land), while an additional 3.4 billion hectares of non-arable land are used to pasture animals (FAO, 2015b). The total area of agricultural land represents 38% of the earth’s terrestrial surface (and almost 50% of its vegetated area). The food system also uses 69% of fresh water resources and 26% of final energy consumption through the entire food life cycle (FAO, 2011; IEA, 2010). Plants capture around 65 billion tonnes of carbon from the atmosphere every year through photosynthesis; an estimated 24% of this annually captured mass is consumed by humans (Haberl et al., 2007).

This section provides a high-level overview of the system’s crop and animal production. We consider land use for food production in terms of tonnes produced. The nutritional and caloric density of food is covered in section 1.7. Our main objective in this section is to understand how land resources are currently used and what opportunities might exist for their reallocation. Figure 3 is a full page graphic that shows an overview of how our global appropriation of land and ocean resources is used for production and extraction activities, which ultimately result in products for food and other uses.

1.2.1 Crop production

Using data from the Food and Agriculture Organization (FAO), we examined the production of crops in terms of their demand for land area (FAO, 2015b). Some of the most important conclusions of this analysis are discussed in this section. In 2011, global crop production amounted to nearly 12 billion tonnes using just over 1.5 billion hectares of land. This resulted in a global average yield of around 7.9 tonnes per hectare, though a significant portion of this figure consists of inedible fractions and fodder (FAO, 2015b).

Food crops

Currently, 30 major crops account for 90 to 95% of human consumption (UNEP, 2007). Cereals occupy the largest extension of arable land area at 47%, followed by oil crops at 19%. Other important sources of carbohydrates, proteins, and fats, such as roots and tubers, pulses, and nuts, jointly cover 10% of cultivated land area, while fruits and vegetables use just under 8%. Only 4% of arable land area is dedicated to crops such as sugar, spices, and stimulants, which are used for human consumption but do not provide significant amounts of essential nutrients. 

Food vs. feed

Only 45% of our arable land is used to produce food that is directly consumed by humans; 33% is used to produce animal feed. Oil cakes, the protein remnant after oil is extracted from oil crops, are another important component of animal diets. Oil cake, a residual product from oil crop processing, represents 64% of the mass of oil crops. Due to its by-product status, it has not been accounted for in the land allocation for animal feed.

Non-food crops

Only 1.1% of global arable land is dedicated to the production of non-food crops like fibres, rubber, and tobacco.

Processed foods

20% of all crops go through major transformation processes prior to consumption. Of the total amount of crops and processed products, 39% are consumed by humans; 38% are used as animal feed, and the rest are used for industrial purposes including energy production and chemical manufacturing. A more in depth look into food processing can be found in section 1.3.

Non-food uses of food crops

Besides fibres, tobacco, and rubber, which are inedible and grown for industrial uses, a significant fraction of food crops is used for purposes other than human or animal consumption, occupying 12% of arable land globally. The majority of these are crops used for the production of biofuels. Other uses of these crops include the production of materials, like bioplastics, chemical substances with industrial uses, and medicines.

The largest sources of crop-derived raw materials for industrial processing are, sugar (47%) and cereal crops (36%). In terms of the total production of these crops, 15% of the sugar produced, 10% of cereal crops, and 36% of vegetable oils produced are destined for industrial processing.

Post-harvest losses

Just under five percent of crop output is lost before being consumed or processed, representing a total of 5% of arable land use. Roots and tubers suffer the highest percentage of losses (10%) followed by fruits (9%), vegetables (8%), and sugar crops (7%). Roots and tubers suffer most losses during the post-harvest and processing stages mainly since fresh roots and tubers are perishable and susceptible to damage or disease post-harvest, especially in places that lack proper storage facilities. In the case of fruits and vegetables, losses mostly result from damage due to handling or spoilage. In the case of sugar crops, most losses occur during distribution and industrial processing  (Gustavsson et al., 2011).

Figure 2: A breakdown of how global land is divided into basic functional  categories and how arable land is specifically divided into different functions.
(FAOSTAT, 2015).

1.2.2 Livestock production

We used data from the Food and Agriculture Organization (FAO) to examine the production of animal products and its associated land use (FAO, 2015b). A striking proportion of agricultural land, almost 80%, is directly or indirectly allocated to livestock production. This includes intensive and extensive pasture lands, as well as one third of the arable land area, which is used to produce fodder crops. 

There are over 31 billion animals kept as livestock in the world: 21 billion chickens, turkeys, ducks, geese, and other birds; 4.6 billion rabbits and guinea pigs; 2.1 billion sheep and goats; 1.6 billion cattle and buffalo, just under a billion pigs; 150 million horses, asses, camels, and llamas; and nearly 6 million deer, ostriches, antelopes, and other animals. In addition to this global stock of cultivated birds and mammals, there are over 78 million beehives.

A wide range of primary animal products is derived from the global livestock population: 1.1 billion tonnes of food in total. Milk constitutes the largest share of this mass (64%). Meat, on the other hand, accounts for 25%, most of it coming from pork (34%), poultry (32%), and beef (21%). With a share of 6% by mass, eggs are the third largest category of primary animal products. 

As can be seen in Figure 2, most animal products are consumed directly by humans (86%), with a particularly high percentage in the case of meat (97%). A significant portion of animal products (7%) is used as animal feed. This is the fate of 11% of milk, 1% of animal meat, and 7% of animal fats. 4% of all animal products are used for non-food purposes, such as the manufacturing of soap, clothing, and carpets. The proportion of non-food use in terms of animal products is highest for fats, of which 47% are destined for industrial uses. 

Catfish ponds in Louisiana
Creative Commons: US Department of Agriculture

1.2.3 Fisheries and aquaculture production

Fish provide 4.3 billion people with around 15 percent of their animal protein intake (FAO, 2014b). Fishing from wild populations is the remaining form of large-scale hunting within the food system. Aquaculture, by contrast, is a form of farming: the rearing of fish and other aquatic organisms within enclosures. As such, these sectors are highly distinct, though because they produce many common products and aquaculture relies in part on wild fish as feed, they are linked in economic terms. 

The global fisheries and aquaculture sectors produced over 176 million tonnes of seafood in 2011. Most of this consisted of finfish (67.8%) with a smaller fraction attributable to crustaceans and mollusks (19.8%), and algae (12.4%). Other forms of seafood constituted 13% by mass the total of animal products in 2011 (FAO, 2015a). It is important to note that the official figures from the FAO only reflect data on monitored fish stocks. Rough estimates indicate that unmonitored (IUU) fishing lands an additional 11 – 26 million tonnes of fish each year, representing 12 – 28.5% of global capture fisheries production (FAO 2014b). The global fisheries sector has and continues to be heavily influenced by subsidies that encourage overfishing, mostly in developed countries. This has led to the expansion of the global fishing fleet to a size 2 – 3 times larger than wild fisheries can sustainably support (Sumaila et. al, 2010, 2013; Nelleman et al, 2008). This continuous structural support of overfishing has led to the progressive decimation of global wild fish stocks since the 1950s (FAO, 2014b).

With 90% of wild fish stocks fully- or over-exploited (FAO, 2014b), the aquaculture sector has been expanding rapidly to keep pace with global seafood demand. Trends in aquaculture production continue showing growth while capture fisheries reached a peak in output in the 1990s and have since modestly declined. With an average annual growth rate of 6.2% between 2000-2012, global aquaculture has more than doubled since the start of the millennium, and is positioned to become the dominant form of seafood production in the near future (FAO, 2014b; Steffen et al., 2015).

Despite aquaculture’s rapid expansion, capture fisheries still dominated the sector in 2011, when over half of the total production of seafood took place via extractive production methods rather than aquaculture. This fraction remains particularly high for finfish, of which over two thirds are supplied by capture fisheries (FAO, 2015b).


With more than a third of global catches in seafood and algae, the Atlantic Ocean provides the largest share of seafood for wild capture fisheries (FAO, 2015a). The Pacific and Indian Oceans come next, each contributing 17% of the total mass of captured seafood. Inland fisheries provided 17% of global captures, but this number obscures the fact that inland fisheries are almost entirely dedicated to finfish capture (94%) with a relatively minor fraction of crustacean, mollusk, and algae production. In fact, almost one fourth of the total mass of all annual finfish production can be attributed to inland captures whereas for other seafood and algae only a minor share (2.7% and 1.1% of total capture respectively) takes place in inland waters. Globally, the most important species, by tonnage caught, is the anchoveta or Peruvian anchovy (which is used almost exclusively for the production of animal and fish feed rather than for human consumption), followed by Alaskan pollock (FAO, 2014b, 2015b). 


Aquaculture is the practice of farming fish or other aquatic organisms in enclosures in rivers, lakes, at sea, or in tanks. It can be done in fresh, brackish, or  saltwater. There are at least 567 species produced in aquaculture systems; besides finfish such as carp, other products include crustaceans, like shrimp and crab; mollusks like octopus, shellfish, and snails; other invertebrates, like sea cucumbers; amphibians and reptiles, like east Asian bullfrogs and crocodiles (FAO, 2014b). For some species, hatchery and nursery techniques have been developed, but many other production techniques still depend on wild seed and juveniles. 

Although not as commonly discussed as animal production, aquatic plants, like the water caltrop and edible lotus, and algae, like the Japanese kelp and the micro-algae Spirulina, are also produced in aquaculture systems. These are commonly used as fish feed (Hasan & Chakrabarti, 2009), or for the extraction of food additives (Maqsood, Benjakul, & Shahidi, 2013). Overall, the most important aquaculture species produced by tonnage is the grass carp, while the whiteleg shrimp is the most significant in terms of economic value (FAO, 2014b).

Crustaceans, mollusks, and algae are already primarily produced through aquaculture. The total area of water and land surface dedicated to aquaculture production systems is not globally documented. Most production takes place in marine waters (36%) or brackish waters (35%) such as coastal zones or estuaries. The remainder of aquaculture production is located in fresh water bodies such as lakes and rivers. Freshwater aquaculture is dominated by the farming of finfish (88%) whereas the majority of production in brackish waters (83%) and marine waters (54%) is used for the production of crustaceans and mollusks (FAO, 2015b).

Aquaculture’s rapid growth initially led to several adverse environmental impacts, but these effects have since been reduced; for example, by slowing conversion of mangroves to shrimp ponds and through reduced reliance on wild-caught fish as feed (Paul & Vogl, 2011). However, given the growth of the aquaculture sector, its associated impacts are at risk of increasing. In addition to the ongoing demand for wild caught fish for feed production, many problems have been associated with poor management, lack of capacity and access to technical knowledge, and irresponsible practices (FAO, 2013). 

Food vs. feed

Although the majority of primary production from fisheries and aquaculture (81%) is directly consumed as food by humans, a significant portion is used as animal feed in aquaculture or livestock production (13% of the global total). The share of production dedicated to feed is particularly high for finfish, of which 19% of the mass ends up as feed. A majority of Peruvian anchovy, the most-landed species by mass, is destined for use as feed. Around 7% of fisheries production is used for non-food-related purposes. For example, 40% of algae is used for industrial purposes, such as the extraction of chemical substances and energy generation (OECD-FAO, 2015).

Total global food production

Figure 3. This sankey diagram shows the allocation of land and oceanic resources into various types of crop, livestock, and seafood production for the year 2011. The second column of the diagram shows the mass of production of each crop, indicating the large variability in production per hectare for the different crop classes presented here. The diagram also shows  the production of fisheries products and livestock.
(FAO, 2015b; FAO, 2006). Click to expand

Regional distribution of crops, livestock, and seafood production in total tonnes per country

Figure 4. Regional distribution of crops, livestock, and seafood production in total tonnes per country. (FAO, 2015b). Click to expand

1.2.4 Regional division

Food production is not evenly distributed around the world; there are large differences between regions regarding both the quantity and the type of food that is domestically produced. The specialization of regions with regards to food production is one of the drivers behind both inter-regional trade as well as different consumption patterns across the globe. In this section, we discuss the geography of food production in more detail. The results presented here are based upon an analysis of data on production quantities as assembled by the FAO (FAO, 2015b).

The distribution of food production

As shown in Figure 4, the United States, China, India, Brazil, and Russia are the world’s most significant food-producing countries in terms of quantity; together they produce over half of the world’s food supply. On the other hand, countries in Africa, the Middle East, and Oceania are together responsible for a mere 10% of global production. East Asia is the world’s most productive region, accounting for 20% of global food production, followed by Latin America and Europe (including Russia and Turkey), which contribute 19% and 17%, respectively.

These numbers mean little on their own. It is more interesting to compare total domestic food production with the population of the regions in which that production is taking place. Although food availability is only a small piece of the puzzle when it comes to ensuring food security for a region’s population, it does provide a crude indication of the extent to which domestic production is sufficient to guarantee food availability (FAO, IFAD, & WFP, 2015).  Measuring food production on a per capita basis reveals a very different geography of production, as seen in Figure 5. While Oceania has an annual primary production of nearly 15 tonnes per person, which is over 8 times the world average, Sub-Saharan Africa’s production stands at barely 0.8 tonnes per person. The U.S. and Canada, Europe, and Latin America are all at above world average levels, while all five Asian and African regions are under the world average of 1.7 tonnes per person (FAO, 2015b).

Regional distribution of crops, livestock, and seafood production in total tonnes per country

Figure 4. Regional distribution of crops, livestock, and seafood production in total tonnes per country. (FAO, 2015b). Click to expand

Regional distribution of crops, livestock, and seafood production in total tonnes per country

Figure 4. Regional distribution of crops, livestock, and seafood production in total tonnes per country. (FAO, 2015b). Click to expand

Regional specialization

Countries and regions have specialized in the production of certain food types for a number of reasons varying from the regional climate and soil conditions to historically determined path-dependencies, cultural preferences, and economic factors. The data presented in this section are all based on an analysis of the FAO’s 2011 production statistics.

In the United States and Canada, fodder crops constitute almost 50% of primary production. Together with cereals and oil crops, these three food categories account for 80% of this region’s output. 

In Latin America and the Caribbean, sugar and oil crops dominate agricultural production. The region is also the leader in the production of stimulants and takes second place, after East Asia, in the production of fruits.

Europe produces a large share of the world’s primary animal products. The region accounts for nearly a third of the world’s milk output, more than any other region, and a fifth of global meat production, second only to East Asia. 

East Asia is a major and diversified food-producing region, leading in the production of vegetables (with more than half of the world’s output) as well as fruits, nuts, meat, eggs, honey, meat, fish, mollusks and crustaceans, and algae. Meat production is focused on swine and poultry and production of dairy is minimal.

The Middle East and North Africa, a region with low availability of arable land per capita, concentrates much of its production on vegetables; crops with high added value. However, it also has considerable production of cereal and fodder crops.

Sub-Saharan Africa is unique in relying on roots and tubers as its primary staple crops instead of cereals. This region produces only 5% of the world’s food supply, but it is the leader in production of roots and tubers; 30% of the global production of these crops occurs here. Secondly, Africa has very high production of pulses, nuts, and stimulants, taking second place in the production of all of these food categories. 

Southeast Asia is second in fish and seafood production. Together with East Asia, these two regions account for 63% of all fish and seafood production. Southeast Asia’s crop production is concentrated on cereals and sugar crops. This region also has the highest production proportion of both oil crops and stimulants. The region of Central and South Asia dominates the production of spices, accounting for over half of the global total. This is also where animal products form the highest share of regional production (14%) though in this region they consist almost entirely of dairy products. 

Oceania has the highest per capita food production in the world, though fodder crops represent over 70% of the region’s primary output. Its share in total primary animal production (3%) is more than twice its share in global food production, indicating a high degree of specialization. Cereals have the lowest proportion of regional production here, at just 9%. Surprisingly, considering its access to coastal waters and fisheries, the region accounts for only 1% of the world’s fish and seafood production.

The Efficiency of production

The efficiency with which crops and livestock are produced is one of the key factors in explaining regional differences in the quantity of food that is produced. In terms of production per arable land area, the most efficient region is Latin America, with a yield of almost 12 tonnes per hectare, or 1.5 times the world average. Sub-Saharan Africa, meanwhile, has a yield about three times smaller than the global average. Almost directly mirroring the pattern of global production per capita, the U.S. & Canada, Europe, and Oceania also all exhibit a greater yield than the world average, while Central and South Asia, Southeast Asia, the Middle East and North Africa, and Sub-Saharan Africa are all below it. The one major exception is East Asia, which has relatively limited per capita availability of arable land, but manages to achieve the second most efficient average yield in the world (FAO, 2015b).

1.2.5 Agricultural inputs

Today’s level of food production relies on vast, continuous supplies of agricultural inputs including water, land, fertiliser, pesticides, labour, and capital. Agriculture is particularly water-intensive relative to all other economic activities. The FAO estimates that agriculture was responsible for 69% of global fresh water withdrawals in 2007 (Aquastat, 2014). Contemporary production methods also require significant inputs of fertiliser and pesticides. The graphics on these pages depict the estimated annual demands of fresh water, fertilser, and pesticides by the agricultural sector.


Agricultural production uses 7.4 trillion cubic meters of water annually based on estimates of the Water Footprint Network (Mekonnen & Hoekstra, 2011). Oil crops, on average globally, consume more water per tonne than cereal crops. Similarly, meat and animal products are very water-intensive. Beef, in particular, consumes more water per tonne than any major category of food with a global average of about 15,000 m3 per  tonne (Mekonnen & Hoekstra, 2012). Spices and stimulants are also very water-intensive per tonne, but do not represent a large portion of agricultural water consumption due to their relatively low volume of production.

The outer ring of the large graph depicted in Figure 6, shows direct water consumption per food product category, while the inner ring shows indirect water consumption divided into two overarching categories: animal products and crops. One third of all crops produced are used as animal feed. The water used for the production of these crops is therefore allocated to the production of animals as embodied or indirect water consumption. Combining both the direct and indirect water consumption of animals, we see that animal products are responsible for almost 30% of agricultural water consumption, despite representing only 11% of global agricultural production in kilograms (FAO, 2015b). This demonstrates the variability in water resource intensity between crops and animal-based products.

Understanding the water demands of different crops reveals their relative input intensity. Gaining more insight into the origin of the water used for crop production is critical to understanding the potential impacts associated with specific crops. Date palms and cotton, for example, receive a low proportion of their water from rainfall relative to other crops, relying on irrigation instead. Areas of India, Pakistan, and Bangladesh near the Ganges and Indus rivers, eastern China, and the Mississippi river have particularly high water footprints (Mekonnen & Hoekstra, 2012). The impacts associated with agricultural water use are discussed further in section 3.1.3.

Water consumption is not limited to agricultural production, but is a vital resource throughout the life cycle of food products, especially in food processing. It is therefore important to note that this graphic only depicts water consumption through the production of raw commodities.


The global food system uses around 200 million tonnes of fertilisers annually, the vast majority of which are synthetic and derived from fossil fuels (FAO, 2015b). Figure 6 shows that, following the pattern of water consumption per crop,  cereals also dominate fertiliser consumption at 71% of the global total. Fodder consumes the second highest amount of fertiliser at 15%. Nuts, which represent only 0.2% of global production mass, consume nearly 3% of global fertiliser. Fertiliser is applied on a per-hectare basis, making total fertiliser consumption per mass of food output highly dependent on crop yields. Sugar crops use 2% of global land, representing 21% of global production mass, yet account for only 0.7% of global fertiliser use (though it is important to note that these figures are distorted due to the fact that sugar harvests are measured pre-processing, which includes all of the harvested inedible, cellulosic fractions). Finally it is important to note that fertiliser use varies greatly across different production systems for the same type of crop, demonstrating the high variability between different agricultural practices (as further discussed in section 1.2.7).


“Pesticide” is an umbrella term describing any form of chemical control of unwanted biological agents, including, but not limited to, rodents, insects, weeds, and pathogens. Pesticides, for the purposes of this report, refer to herbicides, fungicides, and insecticides. Herbicides control the growth of unwanted plants, often called weeds. Fungicides control the growth of fungal pathogens on plants. Insecticides are used to control the presence of insect pests, and are generally applied either as a seed dressing or topically in prevention or response to a pest incident (Eurostat, 2000). In our assessment, we do not include pesticides that are expressed in plant tissue, as is the case with certain Genetically Modified Organisms. Globally, the food system used an estimated 4.4 million tonnes of pesticides in 2011 (FAO, 2015b). Figure 6 shows that cereals and fruits consume the largest share of pesticides. Vegetables, stimulants, roots and tubers, and oil crops each consume around 9% of global pesticides. Although not evident from this graph,  total pesticide consumption is the product of both application rates (kg of pesticide per hectare) and total hectares of each crop type. Cereals’ large share of total pesticide consumption is due to their share of total land use, while the large portion of pesticides used in the production of fruits can be attributed to their high pesticide demand per hectare (Eurostat, 2000).

Pesticide, fertilizer, and water inputs per major food type

Figure 4. (FAO, 2015b; Mekonnen & Hoekstra, 2011). Click to expand

1.2.6 Labour

Though it is difficult to accurately measure, more than 2 billion people are estimated to work within the global food system by the International Labour Organization (ILO, 2007). Of these individuals, roughly 1.3 billion, or 50% of the global workforce, is thought to work in agriculture (UNCTAD, 2013b). Of all farms, the overwhelming majority (95%) are family farms managing fewer than 5 hectares of land (FAO, 2014a). However the definition of “small-scale farms” varies depending on the geographical location, ranging from less than 1 hectare to 10. In Africa and Asia small scale farms predominate with an average size of 1.7 hectares (UNCTAD, 2013b). Farms below 10 hectares managed by pastoralists, forest keepers, and small farmers represent 80% of the total farmland in Sub-Saharan Africa and Asia and IFAD estimates that they produce 80% of food consumed in these regions (IFAD, 2010).

Because most small-scale farmers live in poor, rural areas, children are often required to work on family farms to provide essential labour. According to the International Labour Organization, 60% of all child labourers globally work in agriculture, representing 0.5% of the world’s child population (ILO, 2015). It is important to note that not all participation of children in productive activities is considered child labour. There are appropriate activities that can benefit both children and their families that do not expose them to hazards or detract from their schooling. However, in most instances, child labour is directly correlated with a lack of access to, or poor quality of education as well as structural poverty within the family and region (ILO, 2015).

The incidence of poverty among small and medium scale farmers is very high. The largest segments of the world’s poor are women, children, and men who live in rural environments, most of whom fall in this category (UNCTAD, 2013a). Poverty among farmers is not a problem limited to the developing world; across all regions globally, farmers are the lowest income earners in the food system. For example, nearly 30% of all U.S. farm workers had family incomes that placed them below the national poverty line (National Farm Worker Ministry, 2015).

In addition to farming, it is estimated that 58.3 million people were worked in the fisheries and aquaculture sectors in 2012, which is approximately 2% of the global workforce (FAO, 2014b). Taken together, Asian countries make up 97% of global fisheries activities. For aquaculture specifically, East Asia, including India, accounts for 92% globally, of which China represents 61% (FAO, 2014b). Fishermen (those not working in aquaculture) are numbered at approximately 28 million. Roughly 84% of fishermen work in Asia, with China being the most dominant labour market. For these people, fisheries are a vital means to provide income and livelihoods. 

Just as subsistence farming is the dominant economic model for a majority of the world’s farmers (smallholders), subsistence fishing is common for most of the world’s fishermen. Forced and child labour is similarly prevalent in fishing and aquaculture as it is in farming, often for similar reasons, such as filling crucial labour gaps for families  (FAO & ILO, 2011).  While precise figures on child labour in fisheries and aquaculture are scarce, case specific evidence suggests that its rate of occurrence could be high (ILO, 2013). Forced labour in the fisheries and aquaculture sectors mostly involves migrants, temporary, or illegal crew members. 

Next only to farmers, fisheries workers are the lowest income earners compared with others employed in the food system. While it is difficult to account for different poverty thresholds in each country, it is clear that most people employed in food production (farmers and fishers) are in close proximity to, or below, the poverty threshold.

The number of labourers in the food and drink manufacturing industry is significantly lower than in primary production. The ILO estimates that there are over 22 million people are employed globally  in the food and drink manufacturing sector (ILO, 2007). In the U.S. alone the food processing industry provides 1.5 million people with employment (United States Bureau of Labour Statistics, 2011). Interestingly, individuals working in different steps of the food chain such as in transportation, wholesale, and processing tend to earn more than those in food production (for an example in the coffee chain, see Beshah, Kitaw, & Dejene, 2015).

1.2.7 Production practices

There is enormous variability in global agricultural production and wild extraction systems. The type of practice selected is one of the main determinants of resource demand and yield, and by extension, environmental impact. Getting fine-grained insight into why certain practices are more productive or less impactful than others, and how these features may vary across geographic regions, is essential to understanding what is happening in this critical part of the food system and informing appropriate policy decisions for how to steer it. 

This section presents a high-level overview of the different production practices that are commonly used in crop cultivation, livestock production, and fisheries and aquaculture production. This information, once connected with contextual geographic data and detailed studies on each type of practice, can inform the construction of scenarios for evaluating future pathways for the food system.

Classification of different production methods

Figure 7: A classification of different production methods. This is a non-exhaustive list,  but covers those which are most common. Click to expand

Crop production categories

There is no generic classification system for crop production categories, though a number of farm classification schemes have been proposed and used for data surveying or mapping of agricultural areas. These farm classifications have often focused on geographic or economic parameters like local climate zones, presence or absence of irrigation, or degree of farm commercialization (Robinson et al., 2011). 

Our primary interest in categorizing production typologies here, however, is to review the variety of techniques and production philosophies that can be implemented by any crop-producing farmer in any geographical region, which have central influence over environmental impact and productivity. Figure 7 presents an overview of the different production techniques and practices under discussion. Reading the diagram from left to right allows the creation of a pathway that combines several types of crop production philosophies and practices. There are many layered combinations possible among the practices listed, with only certain categories that are incompatible with one another. 

In recent decades, a distinction has been made between conventional agricultural techniques and variously called “sustainable” or “aspirational” agricultural practices. In practice, these are descriptive rather than rigorous terms due to the many possible combinations of techniques they can both encompass. For example, it is quite common to have large-scale organic monocultures, which may or may not implement aspects of conservation agriculture (Goodman, 2000). Likewise, cropping systems can use a combination of Genetically Modified Organisms and typical conservation practices like crop rotation and no-till farming. Many combinations of practices, from what might seem like contradictory philosophies, are possible. 

Perennial vs. Annual Crops

One of the first distinctions between cropping systems is made between perennial (also called permanent) and annual crops. Perennial plants, like fruit trees, berry bushes, and woody vines, live for many years and invest in intensive root and vascular structures before reaching productive maturity. Depending on the type of plant, reaching this stage can take from two years to over a decade. Properly managed perennial cropping systems can enhance soil quality and biodiversity, since these production systems are not annually disturbed and re-planted. Annual crops on the other hand, grow from seed each year, going through a full annual life cycle of flowering, fruiting, and dying. The vast majority of agricultural crops are annual species, requiring a yearly cycle of replanting (cereals, most vegetables, oil crops, etc.).

Perennial crops have been shown to reduce energy use, erosion and nitrogen loss rates to less than 5% compared to annual crops (Gantzer, Anderson, Thompson, & Brown, 1990). However there are currently no domesticated perennial varieties for grains, legumes, or oilseeds, which make up 69% of the current production. The reason why perennial varieties were originally not domesticated by farmers is that wild annual varieties produced higher yields per hectare. Cultivating perennial crop types that are equally productive could theoretically be possible, but would require a long time using artificial selection (Cox, Glover, van Tassel, Cox, & De Haan, 2006). Active research is underway to develop perennial cereal varieties in many parts of the world (Batello et al., 2013). 

There are however some disadvantages to growing perennials when compared with annual crops. Namely, their permanence has a number of consequences, such as a structural water demand that can be difficult to adapt to local rainfall and weather patterns, and the inability to be rotated which increases risk of pest damage.

Growing practices
Soilbased vs. Soil-less systems

Because of their basic biology, perennial crops are generally only grown in soil-based cropping systems, whereas most annuals can also be cultivated using soil-free techniques like hydroponics. There are many varieties of hydroponic systems, which range from deep bed systems where plant roots are directly suspended in water with liquid nutrients, to a variety of systems where species are planted in a soil-less medium (rockwool, coconut fibre, clay pellets, etc.). Aeroponics, another variation, involves applying a fine mist of nutrient solution directly to roots hanging in air without the use of any substrate.   

Annual plants grown using soil-free techniques can present significant advantages over traditional soil-based systems. “Closed” recirculating water supplies are commonly cited to save 60 – 90% of water use and 20 – 30% of fertilizer use over outdoor, soil-based cultivation (Jovicich, Cantliffe, Simonne, & Stoffella, 2007). Combining soil-less plant production with fish cultivation in aquaponics systems (a mixed aquaculture and hydroponics system), provides a source of dissolved nutrients to the plants (from fish waste) and recirculates purified water back to the fish in a symbiotic arrangement. This kind of solution addresses both the problems of nutrient run-off from concentrated fish farming as well as the need for nutrient inputs into plant production systems (F. Blidariu & Grozea, 2011). The yields of soil-less cultivation systems are generally much higher than those of soil-based systems due to more precise levels of control for nutrient delivery, oxygenation, pH, and temperature control (Jensen, 1999).

Despite the demonstrated benefits of soil-free cultivation, it is only applied on a small fraction of global agricultural land (on the order of magnitude of 0.0001%) though in some countries, it holds a significant percentage of farm share for certain types of crop production (authors’ estimate, based on figures presented in Peet & Wells, 2005 and FAO, 2015b). For example, hydroponic techniques are used for the majority of tomato and bell pepper cultivation in the Netherlands (Cantliffe & Vansickle, 2009). 

The primary reason that soil-free systems are not more broadly applied is that they require a great deal of starting capital and a variety of high-tech inputs (precision management tools and software), not only for the systems themselves, but also for the greenhouses in which they are typically located (Peet & Wells, 2005). For this reason, they are commercially economically viable only for a range of high-value vegetable crops and primarily implemented in the developed world. An important note regarding soil-free cultivation systems is that they cannot be certified as organic production systems, based on the standards set forth by international certification bodies, since they do not make use of soil, which is the cornerstone of the organic production philosophy (Goodman, 2000).

Soil-less systems also have a number of disadvantages that are worth mentioning. Due to the materials and format of soil-less systems, they are typically quite energy and fossil fuel dependent, and not easily integrated with the environment.

Protected vs. outdoor cultivation

Soil-free cultivation is generally only ever applied in protected cultivation systems: greenhouses or an emerging class of high-tech indoor farms, which often use fully artificial conditions for plant cultivation, including artificial lighting. Greenhouse cultivation can either be soil-based or soil-less, though the productivity of soil-less greenhouse systems is generally higher (Gołaszewski et al., 2012) Protected cultivation systems, which include both glasshouses and plastic greenhouses, were estimated to occupy 1.6 million hectares in 2005, which would translate to 0.001%, of current global arable land use (authors’ calculations based on Peet & Wells, 2005).

Yields in greenhouse systems are generally far higher than in traditional field production systems, partly owing to the fact that they extend the growing season for crops. This allows them to disproportionately contribute to global production relative to their small footprint. Certain varieties of plants, particularly leafy greens and Asian cabbages, can produce up to 12 harvests per year in a greenhouse system as opposed to one or two annual yields in outdoor fields. A single planting of tomatoes can continue to produce for 11 months out of the year in a greenhouse system, effectively boosting the total productivity of a single area of land (Jensen, 1999). This practice at least partially explains the extreme range found in global tomato yields, which spans from an average of approximately 1 tonne per hectare (in Somalia) to 560 tonnes per hectare (in the Netherlands) (FAO, 2015b).

Greenhouses using highly efficient LED lights
Creative Commons: US Department of Agriculture, 2014

Monoculture vs. polyculture

A further critical distinction in the classification of crop production systems has to do with the number and variety of plants selected for sequential or simultaneous cultivation. There are two broad categories to consider here, though they each have some sub-variations: monoculture and polyculture.


Monoculture, is the practice of growing a single crop on a large tract of land. It is a hallmark of industrial, conventional agriculture, since it is very well suited to supporting mechanisation and presents large economies of scale (Fitzgerald-Moore & Parai, 1996). Continuous cropping, or mono-cropping, refers to growing a single type of plant species year after year on the same soil (C. E. Murphy & Lemerle, 2006). Strictly speaking, continuous cropping is a term only applied to agricultural production systems that do not implement any form of crop rotation (the practice of growing a winter-season crop on fallow land in order to prevent soil erosion and moisture loss, among other potential benefits). True mono-cropping is less common than is typically made out to be the case in discussions of conventional farming. Even in the United States, which is known for its vast tracts of single-crop agriculture, a significant majority of crops (82 – 94%) is grown with some kind of rotation (corn and soybean being a very common example), though cover cropping, an important conservation agriculture technique, remains uncommon (White, 2014). These rotation systems, however, do not qualify as polycultures under stricter definitions of the term, which generally refer to production systems that grow multiple crops simultaneously on the same plot of land.

Large-scale monocultures are widely reported to result in agricultural problems ranging from depletion of soil fertility due to continuous extraction of the same nutrients, to the intensification of pest problems by providing uninterrupted breeding grounds for specialized pests. Because of their design for large-scale productivity, they typically require very high inputs in terms of both chemicals and energy (for operating machinery, for example) (Olesen & Bindi, 2002).


Multiple cropping, also known as poly-cropping or polyculture, involves growing multiple crops on the same plot of land. The intensity and productivity of polyculture systems can range significantly. In general, multiple cropping is associated with stable productivity and on average higher relative yields than are found single crop systems. Certain crop combinations have much higher combined total yields, and can be selected for high productivity (Gliessman, 1985).

Multiple cropping can allow for better pest control through mutualistic interactions, increased microbial activity in the soil, more efficient fertiliser use, better use of time and space with more crops per unit area, pattern disruption for pests, reduction in water evaporation, shared benefits from nitrogen fixation from crops like legumes. There are also potential disadvantages such as difficulty with mechanisation, competition between plants for nutrients, water, and light; difficulty with incorporating fallow periods; and the possibility of allelopathic interactions between plants (Gliessman, 1985). However, agricultural research is typically focused on maximizing single crop yields, instead of thinking of yields on a long-term diversified basis, which has translated into minimal investment in high-yielding polyculture systems  (U.S. Congress Office of Technology Assessment (OTA), 1985). 

Monoculture fields in America
Creative Commons: Daniel Lobo

Water management
Irrigated vs. rainfed agriculture

On average, irrigated agriculture produces more than twice the yields of rainfed agriculture (Steinfeld, H., Gerber, P., Wassenaar, T., Castel, V., Rosales, M., & De Haan, 2006).  However, despite significant bluewater extraction and the doubling of the global irrigated area since the early 1960s (FAO, 2011), rainfed agriculture remains the world’s predominant production system (FAO, 2013) Irrigation is typically associated with large water losses, however new and more efficient irrigation methods such as drip irrigation can reduce water usage considerably; by 15-25% according to one estimate (Mushtaq, Maraseni, & Reardon-Smith, 2013). It is important to emphasize, however, that efficiency does not necessarily equate to sustainability. Improvements in efficiency do not necessarily lead to sustainable use patterns, as total water withdrawal using efficient methods can still result in a net increase in consumption.

Production methods and philosophies
Conventional, non-genetically modified

Despite the possible variability in applying the term “conventional” agriculture, what is typically meant by this phrase is: outdoor crop cultivation in monoculture systems with high levels of mechanisation and artificial inputs, largely implementing the techniques introduced through the Green Revolution (see Chapter 2). Most conventional agricultural systems use high-yielding varieties that have been bred specifically for large-scale monoculture production, and often have features (like shorter stalks) to facilitate mechanical harvesting, boost yields, and prevent crop spoilage. Generally speaking, conventional agricultural techniques also implement ploughing of soils as a technique for homogenizing and breaking up the top layer of soil prior to planting. This combination of techniques results in a high-input, energy-intensive, soil and biodiversity depleting, low labour and high-yield form of agricultural practice (Matson, Parton, Power, & Swift, 1997).

Conventional, genetically modified

A relatively new addition to the repertoire of conventional farming techniques is the genetic modification of cultivated species in order to artificially enhance them with desirable traits. As of 2013, 174 million hectares (12.5% of all arable land) was cultivated with genetically modified (GM) crops (GMO Compass, 2014). GM crops are still primarily limited to a few species such as maize, soy, cotton, and oilseeds, though sugar beet, alfalfa, papaya, and squash are also emerging as more common GM crops. In the U.S., in terms of planted area in 2014, 94% of soybeans, 96% of cotton, and 93% of corn were GM varieties (“USDA Economic Research Service – Adoption of Genetically Engineered Crops in the U.S.,” n.d.).

Genetic modification of crops has most commonly involved the introduction of non-native traits that confer either herbicide-resistance or pest-resistance. Herbicide-resistant crops can be sprayed with herbicides, allowing for the  elimination of weeds without negative effects on the crop itself. Pest-resistance has most commonly been conferred to crops through the expression of foreign inserted genes for the expression of Bt toxins, derived from the bacterium Bacillus thuringiensis. The insecticidal proteins produced by Bt are a class of natural insecticides, which are pest-specific, and also used in powdered or liquid form in organic agriculture (Caldwell, Sideman, Seaman, Shelton, & Smart, 2013).  

GM crops and foods have been the subject of numerous controversies centred around the topics of food safety and potentially unforeseen ecological impacts (Finucane & Holup, 2005). Additionally, concerns have been raised by various civil society groups around the role of GM crops in supporting greater consolidation and corporate control of agricultural supply chains. The controversy continues, with fierce rhetoric and complicated realities clouding both sides of the debate (Gilbert, 2013).

Organic agriculture

Organic agriculture (sometimes referred to as biological or ecological agriculture), is a production philosophy and set of practices that were first defined in the beginning of the 20th century with a focus on healthy soils as the foundation of agricultural productivity (Ma & Joachim, 2006). This production philosophy has been codified in strict guidelines through the definition of organic certifications, which include strong prescriptions against the use of synthetic chemical inputs, genetically modified seeds, antibiotics in the case of livestock rearing, and so forth (Baier, 2005).

The growth in adoption of organic agriculture has been estimated at a compounded annual growth rate of 8.9%, greater than any other form of agricultural practice (Paull, 2011). The total adoption of organic agriculture is now estimated to cover 37.5 million hectares (0.9 of total agricultural land) (Paull, 2011; Ponisio et al., 2014; Willer & Lernoud, 2014). In total, 1.9 million organic producers were reported, with over ¾ of these located in developing countries (Willer & Lernoud, 2014). As a result of the rapid growth in demand for organic food, the production of organic crops has become predominantly a highly intensive monoculture production method (Guzman, 2014).

Organic production has a number of benefits over conventional agriculture. For example, it was found to have a 29% lower energy demand when compared with non-organic systems, averaged across a large subset of products in a UK study commissioned by the FAO (Ziesemer, 2007). Higher use of pesticides and other chemicals in non-organic agriculture leads to the unintentional killing of non-pest insects, which can lead to a decrease in beneficial predatory insect species and a reduction of sources of nutrition for animals higher up the food chain (Kim, 1993). Partly because of these dynamics, organic agriculture has been associated with higher levels of biodiversity. According to a meta-analysis of studies comparing biodiversity with organic and non-organic practices, on-farm biodiversity measures were on average 30% higher with the use of organic practices when compared to non-organic controls (Fuller et al., 2005).

There has been a great deal of debate historically about the sustainability of organic agriculture, particularly from a yield perspective. Organic agriculture has generally been found to result in yields 20% lower on average than in conventional agricultural practices, though with a high variation between crops and farms (De Ponti, Rijk, & Van Ittersum, 2012), leading to concerns around the need for larger amounts of arable land potentially needed to satisfy global food demand under an organic production model (Badgley & Perfecto, 2007; Connor, 2008; De Ponti et al., 2012). A recent meta-study published results indicating that the yield gap between organic agriculture and conventional farming systems is smaller than expected previously (Ponisio et al., 2015). The Rodale Institute, which promotes organic agriculture, released the results of their 30-year trial of side by side controlled plots, maintaining that organic yields matched conventional yields, outperformed conventional in years of drought, built rather than depleting organic matter in soil, and used 45% less energy than conventional systems (The Rodale Institute, 2011). 

Many groups have argued that increased research and development funding targeted specifically at organic practices could lead to an elimination of the organic yield gap (Ponisio et al., 2015). Regardless of this assumption, surveying global data makes it clear that a more dominant cause of low yields is simply less advanced agricultural practice. Organic tomato production in the Netherlands yields 350 tonnes per hectare, while conventional tomato production in otherwise similar conditions ranges from 50 – 120 tonnes per hectare in other parts of Europe (FAO, 2015b; Gołaszewski et al., 2012). This indicates that the yield gap between organic and non-organic forms of production can be much less significant than the yield gap that results simply from lack of knowledge, technique, or sufficient resources. 

There are several other agricultural systems which implement some of the same basic principles as organic production as a basis, for instance biodynamic agriculture and permaculture. Within certain contexts, they can be considered variations on organic production (Nesme, Colomb, Hinsinger, & Watson, 2014).

Integrated farming

Integrated farming attempts to produce food that is better for the environment taking into consideration a large number of factors. The management practice does not ban or require certain practices or inputs, but attempts to optimize practices depending on a number of conditions analysed using a systemic approach. For example, no-till agriculture may reduce energy use under certain conditions, but increase it if additional crop protection measurements are required in exchange. All of the inputs and effects in the entire system as the result of a change in practice are considered (EISA, 2012).

Studies have shown that though energy use and emissions with integrated farming are higher per hectare than in organic production, they are lower per tonne produced than in organic and conventional agriculture (Tuomisto, Hodge, Riordan, & Macdonald, 2012).

Conservation agriculture principles

Conservation agriculture is a term that encompasses three crop management principles: no-till agriculture, crop rotation, and residue retention. It has gained international support in policy circles as a method of improving long-term soil productivity. Currently around 100 million hectares worldwide apply conservation agriculture principles (Sommer et al., 2012). Anecdotal evidence suggests that herbicide-resistant GM crops have facilitated the adoption of conservation tillage practices due to the increased ease of weed control through herbicide application (Fernandez-Cornejo, Hallahan, Nehring, Wechsler, & Grube, 2013). Conservation tillage practices have frequently been associated with an increase in herbicide use, because the lack of ploughing allows weeds to become established. Other methods of weed control, such as mulching and cover cropping, can also be applied with conservation tillage practices, removing dependence on chemical weed control (Bullied, Marginet, & Acker, 2010; Moyer, Roman, Lindwall, & Blackshaw, 1994; Sans, Berner, Armengot, & Mäder, 2011).

No-till farming (also called zero tillage or direct drilling) involves cultivating crops or pasture without using ploughing, thereby maintaining soil ecology, decreasing erosion and compaction, and improving water retention (Holland, 2004). This practice increases soil quality by increasing the amount of infiltrating water and increasing retention of organic matter and nutrient cycling. In addition to reducing soil erosion, it increases new soil formation by promoting the amount and variety of life in the soil, including soil-forming organisms (Martin R. Carter, 1994). In 2014, around 125 million hectares or around 9% of cropland was under no-till cultivation (Pittelkow et al., 2014). In 1999 no-tillage farming was only practiced on around 45 million hectares  worldwide. Its adoption grew at a rate of around 6 million hectares per annum between 1999 and 2009. The practice has been widely adopted in all types of climates and on all types of soils (Derpsch, Friedrich, Kassam, & Hongwen, 2010).

Though no-till agriculture provides many benefits for soil and has been largely adopted for this reason, adopting the practice alone may come with drawbacks. Recent findings indicate that, contrary to common beliefs, no-till agriculture generally has been found to have a negative effect on crop yields, of an average of 5.7% in one meta-study, unless applied with other conservation agriculture principles (crop rotation and residue retention), which then narrow the yield gap (Pittelkow et al., 2014). By contrast, under dry and arid conditions, no-till was found to confer a yield benefit regardless of whether it was applied with other techniques. Additionally, no-till practices may initially require a higher need for fertiliser (Frankinet, Roisin, Baumer, & Ehlers, 1989) and pesticides (Soane et al., 2012) in order to maintain yields. If combined with other practices, such as crop rotation and residue retention, the potentially negative effects of conservation tillage can be avoided and benefits strengthened. Regardless of some drawbacks, conservation tillage has been shown to be one of the only agricultural techniques that reduces the rate of soil erosion to within the background geological rate of soil loss and formation (Montgomery, 2007) (see section 3.1.2). 

A hillisde agricultural village in Uganda
Creative Commons: Rod Waddington, 2015

Crop rotation is the second practice within the conservation agriculture portfolio. Methods such as intercropping or crop rotation can increase resistance against pests and disease, as well as increase soil quality, thus reducing the need for expensive inputs. One estimate states that these methods can reduce U.S. pesticide use by 50% without reducing yields (Pimentel & Lehman, 1993). Additionally, polycultures can improve total yields per area by taking advantage of symbiotic relationships (Naeem et al., 2013).

Residue retention is the third primary component of conservation agriculture. Even when applying no-till agriculture, the removal of crop residues can reduce the fertility of the soil over time. One study showed that organic carbon in the soil was reduced by 75% after 15 years of no-till cropping with residue removal (Chivenge, Murwira, Giller, Mapfumo, & Six, 2007). Several studies have suggested that in addition to improving soil fertility, the combination of residue retention with other conservation agricultural practices leads to an increase in the amount of water available to plants through increased infiltration, reduced runoff and reduced evaporation (Sommer et al., 2012).

Livestock production practices

Livestock production uses almost 80% of global agricultural land, most of which is pastureland. The vast majority of grasslands used for pasture are relatively inexpensive and either low-carbon, arid, cold, steep, or rocky, offering very few options for other food-producing uses (Capper et al, 2013). As illustrated in Figure 8, livestock are produced in either mixed, grassland-based, or industrial (landless) systems. Though pastures have long served as the foundational resource for rearing the world’s domesticated animals,, the livestock sector has gone through a transformation in recent decades fueled by growth in demand for animal products.  Producers have turned from primarily depending on feeding animals using residual materials and pasturing them on low-fertility land, to more intensive production approaches. In Concentrated Animal Feeding Operations (CAFOs), purchased feed crops  and equipment are used to replace land and labour, though one can argue that it simply increases demand for high-quality, crop producing land over larger amounts of low-quality pasture. As an example, the EU imports enough soy to account for the use 18 million hectares of agricultural land outside the EU (Idel, Fehlenberg, & Reichert, 2013), a large part of which is used as a main component in animal feeds.

With livestock now consuming food that would otherwise be suitable for human consumption, meat production for the wealthier part of the population has begun to compete directly with food availability for the global poor. A key factor here is the relatively inefficient conversion rate of cereals into animal protein. UNEP has reported that it takes approximately 3 kg of grain to produce 1 kg of animal protein using cereals as feed  (Nellemann et. al., 2009). Though there is no global shortage of staple crops, competition for cereal crops can drive up prices globally, which reduces the economic availability of food in food insecure regions (UNEP, 2012). 

It is important to note that different animal products have highly varying resource demands in production.  A commonly cited number in discussions of animal production impact is the Feed Conversion Ratio (FCR), which is a measure of an animal’s efficiency in converting feed mass into an increase of a desired output (e.g., milk, meat, eggs). FCR ranges greatly, from an average of 1.6 in fish, 2 in poultry, and 3 in swine, to up to 11 in cattle (Boyd, Tucker, Mcnevin, Bostick, & Clay, 2007).   

An alternative to extensive grazing on monoculture pastures and CAFOs is silvopastoralism, where livestock grazes on mixed vegetation. Less land is required because dry matter production in silvopastoral systems is 27% higher than monoculture pastures. Additionally, silvopastoralism requires fewer and less agricultural inputs, such as fertilisers and pesticides, and less upkeep than monoculture pastures (Broom, Galindo, & Murgueitio, 2013). Additionally, such systems can be more productive than extensive grazing. For example, silvopastoral systems lead to a higher milk production in cows than standard, but highly productive, monoculture pastures (Broom et al., 2013).

A classification of different production methods for livestock

Figure 8: (FAO, 1995)

Fishing practices

The FAO defines the term “fishery” as an activity leading to the harvesting of fish through either wild capture or aquaculture. Fisheries are further defined in terms of at least some of the following:  “people involved, species or type of fish, area of water or seabed, method of fishing, class of boats, and purpose of the activities” (FAO Term Portal, 2015). As such, fishing methods must always be considered in context: any fishing method not appropriately matched to a species, location, or time of year can potentially result in ecological harm. 

Over 90% of fishers involved in global capture fisheries operate in either small-scale or artisanal fisheries. Though many of them are at least partly engaged in fishing for subsistence reasons, they are estimated to produce approximately 50% of fish supply for human consumption (Johnson, 2005).

Small scale  fisheries are not necessarily considered artisanal, and vice versa, though when one term applies to a fishery, the other often does as well. Artisanal fisheries are those that typically utilise relatively low levels of technology and have relatively low levels of capital investment per fisher, often making use of traditional fishing techniques (e.g., hook and line, beach seines, cast and lift nets, fish traps and weirs, manual harvesting). They are typically associated with lower ecological impact, lower running costs and fuel consumption, lower cost of technology, higher versatility, and higher employment opportunities. These relative benefits do not imply that artisanal fisheries do not contribute to overfishing or ecological damage, however, as these fishers can and do overfish available resources and occasionally use ecologically damaging methods such as poison or dynamite (Johnson, 2005).

Large-scale, industrial fisheries employ around 10% of fishers globally and are responsible for an estimated 50% of global fish landings (Johnson, 2005). There are four main large-scale commercial methods of catching fish and seafood; trawls and dredges, line fishing, net fishing, and traps. A description of these methods as well as some others are shown in Table 1, along with a brief overview of their relative impacts.

In general, bottom trawling and net methods have the highest negative impacts, but are the most economical as they catch an enormous volume of fish using relatively little labour and time. A survey about impacts due to different fishing methods showed that experts agreed that bottom trawling produced the largest negative effect on the environment, attributed largely to the direct effect on the seafloor habitat (Chuenpagdee, Morgan, Maxwell, Norse, & Pauly, 2003). Bottom trawling accounts for a large part of the destruction to coral reefs and sponges, around 1 million pounds were destroyed between 1997-1999 in the water around Alaska alone (Lewison, Crowder, Read, & Freeman, 2004). While bottom trawling damages the seafloor habitat by scraping and ploughing the floor up to 30 cm, it also stirs up soil, causing an additional impact through increasing the turbidity of the water (Dayton, Thrush, Agardy, & Hofman, 1995). 

Similar to trawls and dredges, other types of nets have little species selectivity, producing a lot of bycatch, including through lost nets (referred to as ‘ghost fishing’) (Suuronen et al., 2012). In particular, gillnets (or driftnets) are particularly damaging, resulting in the highest bycatch levels of mammals, sea turtles, and seabirds (Chuenpagdee et al., 2003). Such nets were banned in international waters in 1992 by a U.N. resolution, though individual nations can still use driftnets of up to 2.5 km in length in their own waters  (Lewison et al., 2004).

Longline fishing uses relatively little fuel, inexpensive equipment, is relatively species-selective, and generally causes minimal habitat damage (Pham et al., 2014). The most significant downside to longline fishing is that it still results in  capture of non-target species such as marine birds, mammals, and turtles. The method is also labour- and time-intensive, and is dependent on the price of bait (Suuronen et al., 2012). Similarly, trolling, which requires dragging fish lines through the water to attract fish, is more species selective than nets, trawls, or dredges, but also produces a low catch.

Traps and pots catch fish and crustaceans by using barriers that allow fish to enter an area or trap but make it difficult for them to escape. Trap design or bait selection can result in species selectivity. Specialized gear can be very effective at targeting certain species, such as lobsters (FAO, 2001). This method can be a relatively low impact manner of fishing when managed properly, but often old traps are forgotten or discarded, leading to ghost fishing and additional marine debris (Arthur, Sutton-Grier, Murphy, & Bamford, 2014). For example, one program to collect derelict pots and traps around Virginia estimated that 41% of the gear found had been abandoned (Bilkovic, Havens, Stanhope, & Angstadt, 2014).

Aquaculture practices

Aquaculture may offer some benefits over fishing as it does not lead directly to overfishing and can be separate from natural habitats. The problems and solutions associated with aquaculture are generally more similar to those encountered with conventional agriculture. However, the methods that are typically applied for aquaculture can and do have negative effects on the environment. For one, aquaculture isn’t typically separate from marine and freshwater environments. According to the FAO 2012 Fisheries and Aquaculture Yearbook, around 63% of the aquaculture production of fish, crustaceans, and other species, occurred inland, while 37% of the production was marine aquaculture (FAO, 2012). Freshwater aquaculture often comes at the expense of other ecosystems. For example, in Vietnam, 290,000 hectares of wetlands were converted into shrimp aquaculture (McDonough, Gallardo, Berg, Trai, & Yen, 2014). Both freshwater and marine aquaculture produced through methods such as growing fish in pens, can lead to effects such as disease, parasites, and concentrated waste, due to the crowded nature of aquaculture. Additionally, for predator species of fish, fish farming doesn’t entail a detachment from wild ecosystems. Salmon, for example, require a higher volume of wild fish for consumption than they yield in terms of edible meat (Seafood Choices Alliance, 2005). 

An overview of the most common fishing methods, and their impacts to the environment

Table 1: (Monterey Bay Aquarium Seafood Watch, 2015)

1.3 Processing

Food processing can generally be described as the “transformation of agricultural crops, livestock, and seafood into secondary products.” However, the types and intensity of processing vary greatly between products. Processing could refer to, for instance, the simple cleaning and packaging of vegetables, but it also includes the production and packaging of sugar, breakfast cereals, or soft drinks (Monteiro, Levy, Claro, Castro, & Cannon, 2010).

Share of global food products going into processing

The wide variety of options included in the concept of processing results in data inconsistencies, which make it challenging to accurately estimate the total amount of food that is processed globally. However, as mentioned in section 1.1, the FAO’s statistical database does contain information on the processing of primary crops, of which around a fifth are routinely processed into secondary products before consumption. Sugar and oil crops make up the largest share of primary crops going into processing (1,940 million tonnes; 92%), the remainder is split among cereals (89 million tonnes), fruits (55 million tonnes) and roots and tubers (15 million tonnes). The main outputs for human consumption are alcohol, sugars and sweeteners, and vegetable oils, while oil cakes are the main product destined for animal feed (FAO, 2015b).

These figures only tell us something about the share of primary production initially used in processing, but do not say much about the total amount of secondary processing such as breakfast cereals, yoghurts, or soft drinks.  In this regard only broad estimates are available; the United Nations Industrial Organization (UNIDO) estimates that in the percentage of all food going through some form of processing ranges from 30 percent in the Global South, to 98 percent in the Global North (FAO, 2012a).

Food processing industry

Although global data is unavailable, there are indications that the food processing market is experiencing continuous growth. The U.S. food processing industry, for example, has shown market growth of up to 5% annually. Currently, the size of the American food processing market is on the order of $2 trillion, providing jobs for an estimated 1.5 million people (Feldman, 2011; United States Bureau of Labour Statistics, 2011). The growth rate for the European processing industry is more modest (a 3.4 percent growth in turnover between 2011 and 2014); the industry’s growth rates appear to have stabilized in recent years (Food and Drink Europe, 2015).

1/5 of all primary crops are processed before consumption.

The value of processing

Processing can have several objectives, which include complying with food security standards, extending product life, developing special products (e.g. cheese, sausage, vegetable oil) or increasing consumer convenience – with the latter increasingly becoming a driving factor. Legal standards on health and hygiene, technological innovations, as well as consumer demands are additional factors influencing developments the global food processing industry (A. Regattieri, M. Gamberi, 2007; Market Research Reports, 2015). 

Aside from these considerations the main goal of the food processing industry obviously lies in adding value to primary or secondary food products with the purpose of extracting a profit. The value added in food processing and subsequent stages of the production chain such as retail and distribution, is often much larger than that of primary producers such as farmers and fishermen (for a case in point, see Beshah et al., 2015).

Workers in a food processing plant.
Creative Commons: Wikimedia

Nutritional value and bio-availability

Aside from the objectives mentioned above, food processing is associated with a range of negative impacts such as an increase in energy use or, depending on the exact production process, a decrease in the nutritional value of food. The processing of food can have considerable impacts on the nutritional value of food products. Exposure to high levels of heat, light or oxygen can lead to a decisive nutrition loss. High heat levels, for example, destroy certain vitamins and reduce the biological value of proteins (Rong, Hai-Yan, Dongfang, Xingrong, & Aluko, 2013). Oxidation, on the other hand, degenerates lipids and destroys oxygen-sensitive vitamins. Water-soluble vitamins (such as vitamins C and B) are generally more affected by processing than fat-soluble vitamins (vitamins K, A, D, or E) in this respect. High nutritional losses occur, for instance, during the milling and grinding of cereals to remove their fibrous husks; most of the plant’s fibre, B-group vitamins, phytochemicals, and minerals are in these husks. While the freezing of products does not affect the nutritional value of foods, blanching and canning both cause nutritional losses due to high temperature exposures. Nutritional losses also occur through the peeling and trimming of fruits and vegetables to remove their skin, as a major fraction of nutrients tend to lie close to the skin surface (State Government of Victoria, 2014). 

Processing of food products can vary to a great degree. While fresh meat, milk, grains, and vegetables usually undergo few processing steps that include practices like cleaning, removing of inedible fractions, portioning, refrigeration, or bottling to make products more accessible to the consumer, also different highly processed products exist that have been altered a great deal (Monteiro et al., 2010).  Examples of these highly processed foods are breads, biscuits, confectioneries, crisps, cereal products, sugared and soft drinks, and processed meat products.  They often contain additives such as flavours, colours, or other substances that make them more palatable or even habit-forming (Moubarac et al., 2013). Global diets today increasingly consist of highly processed foods. The more processing, usually the lower the nutritional value and the higher the adverse impacts on human health resulting from low nutrient density, insufficient dietary fibre, and a surplus of simple carbohydrates, saturated fats, sodium, and trans fatty acids (Monteiro, Levy, Claro, de Castro, & Cannon, 2011).

The processing of food also can have positive implications and can increase the bioavailability of nutrients from raw food products. Perhaps unexpectedly, frozen vegetables can have a higher nutritional value than ‘fresh’ vegetables (Rickman, Barrett, & Bruhn, 2007). Furthermore, cooking is a traditional form of processing that is essential to ensuring the bioavailability of certain nutrients (see section 1.6 for more on cooking and food preparation). Finally, methods like canning, pasteurization, dehydration and freezing preserve nutrient contents and can make food longer available (Pasha, Saeed, Sultan, Khan, & Rohi, 2014; Weaver et al., 2014).

Resource consumption in food processing

The preparation of processed foods requires resources such as  energy, water, and materials (e.g., for packaging). The high demand for energy in the food processing industry arises mainly from increasing automation and machinery use during this production stage (Canning, Charles, Huang, Polenske, & Waters, 2010). In an advanced food industry like that found in the United States, food processing is responsible for around one third of total energy use in the food system. Up to 1,000 calories of energy are needed per production of 1 calorie of processed food (Verma, 2015). 

From another perspective, looking at the embodied energy in the diet of an EU citizen, it has been estimated that of the total embodied dietary energy, 28% is due to food processing making it the second largest share next to the production stage (33%). All in all, in the year 2013, the European food industry consumed 28.4 MTOE of energy, or 2.6% of the EU-28’s average final energy consumption  (Dallemand et al., 2015). 

As there is a multitude of diverse food products that require different production methods and different numbers of processing steps, energy inputs vary widely per product. Canning of fruits and vegetables (575 kcal/kg) and also freezing of fruits and vegetables (1,815 kcal/kg) have lower energy inputs as opposed to food products that entail more processing steps like the production of breakfast cereals (15,675 kcal/kg) or chocolate (18,591 kcal/kg). With a growth in demand for more convenient or new food products (e.g. pre-cut vegetables, salad mix products) which entails more processing, preparation and packaging, the energy intensity of the sector will also increase (Verma, 2015).

Data for the U.S. food processing industry between 1997 to 2002 confirm this trend showing an annual increase in energy consumption of 8.3% (Verma, 2015). The increase in energy intensity is a long term trend: since the early 20th century yearly increases in energy use between 9.6 and 13 percent have been documented for cereal products, baking products, fresh dairy and snacks, frozen and canned food, spices and condiments (Canning et al., 2010). On the other hand, the EU region has managed to decrease its  food processing sector’s energy consumption over the past years (2005-2013), despite an overall growth in processing (Dallemand et al., 2015).

Aside from inputs in the form of energy, food processing also typically increases the demand for specialized food packaging. The impact of packaging and associated material wastes is discussed in section 3.1.6. 

Global estimates of the share of resources associated with food processing are not available as most studies focus on the production stage of food products where most environmental impact still occurs (Boye & Arcand, 2012).

Sunrise in Peachy Canyon vineyard in California.
Creative Commons: Malcolm Carlaw

The US food processing industry is growing at a rate of 5% annually.

1.4 Trade

Every year, about a billion tonnes of raw and processed food commodities are traded internationally; this amounts accounts for 14% of the world’s food supply. Forty-one percent of this trade happens within regions while 59% takes place between them. Using data from FAO’s statistical database, we analysed trade in food commodities between regions for the year 2011. The high-level results of this analysis are summarized in Figure  9 (FAO, 2015b).

Eggs and algae are the least traded commodities in proportion to their production. On the other hand, for each 100 tonnes of stimulants (like coffee and tea) produced, 109 tonnes are traded. This happens because some countries, mostly European ones, engage in importing and re-exporting these goods. Cereals are the most traded commodities on Earth (accounting for 30% of trade by mass). The U.S. and Canada, Europe, and Oceania are the world’s major exporters; East Asia and the Middle East are its major importers.

Five regions form the core network of international food trade: The U.S. and Canada, Latin America, Europe, East Asia, and South East Asia; 88% of the world’s food trade passes through them. Moreover, there are three main trade patterns in the world: 

  • Intra-European trade movements, which are mostly self-contained.
  • The role of East Asia as the largest food importer region in the world.
  • The role of the U.S. and Canada and Latin America as the largest food exporter regions.

As seen in Figure 9, Europe is the region  most involved in international trade. 38% of global trade by mass involves product movements between its countries or outside of its regional borders, with 30% of the world’s trade taking place entirely within this region. Considering only extra-regional trade, Europe provides 11% of global exports and 18% of global imports, which results in a regional net trade deficit.

Together, the U.S. and Canada and Latin America account for over 60% of the world’s inter-regional trade. Two thirds of all Latin American exports reach Europe and East Asia. The U.S. and Canada region sells 43% of its exports to East Asia alone. This dwarfs the exports destined for Inter-American trade, which account for only 18% of exports. Southeast Asia is the third-largest food exporter, but at a far lower proportion.

East Asia purchases 35% of the world’s traded food products, with which it manages to provide for only 8% of its food supply. Its main trading partners are the regions surrounding the Pacific Ocean (East Asia imports 43% of the US & Canada’s exports, 24% of Latin America’s exports, 39% of Southeast Asia’s exports, 40% of Oceania’s exports, and 44% of its own exports end up as intra-regional flows). 

The Middle East and North Africa region purchases 13% of the world’s extra-regional exports. This amount represents a third of the region’s food supply, making it the most dependent on international trade to meet its food availability needs.

Oceania has the highest participation in international trade relative to its domestic production, but in absolute terms it accounts for less than 3% of trade movements. The Central and South Asia region has little involvement in international food trade. In general, it is a self-sufficient region with a small trade surplus. Finally, Sub-Saharan Africa is the region most disconnected from the world in terms of food trade. Its participation is less than 3% of the world’s extra-regional trade movements, with a small trade deficit. 

Food traded in 2011

Table 2: The total percentage of each food product category that was traded in 2011 (FAO, 2015b)

Global trade flows

Figure 9: An overview of total international trade volumes between regions. This diagram does not include intra-regional trade (Trade flow calculations based on FAO, 2015b for the reference year 2011). Click to expand

East Asia imports the largest volumes of food

East Asia imports only 8% of its food supply, but is the region that imports the largest volumes of food in the world. The region also imports 41% of other products, including seaweed, sugar, and fibres. 

The Middle East and North Africa are the most dependent

The Middle East imports 31% of its food supply, including 55% of its cereals, 91% of its oil and 70% of its sugar. It is the region with highest import dependency for its food supply. In turn, it exports a mere 9% of its food output, mostly crops such as nuts and fruits, but also oils and sugar.  

Oceania has the most favourable trade balance

Oceania imports a mere 7% of its food supply, yet it exports 51% of its production. It has the most favourable food trade balance in the world. Its main exports are cereals, which go to the East Asia and South and Central Asia regions, and oil crops, which are primarily sold to Europe.

Latin America is the largest importer of non-food crops

Latin America imports 15% of its food supply, yet exports 21% of its production. Its main exports are oil crops, which are sold to East Asia, and a range of other products including sugar, fruits and cereals. The region imports mostly non-food crops, such as flowers and live plants, from Europe, but also food products from the U.S. and Canada.

1.5 Retail

The big players in food distribution

Distribution channels have undergone significant changes since the economic reforms of the 1980s and 1990s. Globalisation has created space for large retailers to dominate over much of the developed and developing world (Wrigley & Lowe, 2010). Today,  an estimated 51% of global food sales are purchased through supermarkets and hypermarkets. Food sales through these channels are growing at a rate of 2% annually (Nielsen, 2015). While the world’s largest food retailers were traditionally based in the U.S. and Europe, waves of supermarket development have begun globally out in what has been labeled the “supermarket revolution” (S. Murphy, Burch, & Clapp, 2012).

Supermarkets first spread out in the 1990s to South America, Central Europe, and South Africa. In the early 2000s they only accounted for between 5 to 10% of the food retail market share, however later that decade they grew to 50% of the market. A similar pattern occurred in Central America, South East Asia, and Mexico. The final wave and most recent market expansion has been in China, Vietnam, India, Russia, and Africa. Generally, within nations, the spread of these large retailers has developed out from urbanized cities and middle class regions to rural communities (OECD Competition Committee, 2013).

Managing the big retailers

Globalised food networks, high technological management, diversified product branding, and reduced nutritional content, are all characteristics of the modern food distribution system. Retail giants such as Walmart now use high level ICT systems to improve their logistical management and gain a market edge on their competitors  (OECD Competition Committee, 2013). The ICT boom of the late 1990s enabled the collection of immediate demand-related data which helped retailers to reduce their incumbent investments and improve their supply chain efficiency (Deloitte Touche Tohmatsu Limited, 2014). Because of their scale, scope, and bargaining power, large food retailers have continued to offer generally cheaper priced food commodities than their small-scale competitors (Ruppanner & Mulle, 2010). So-called “supermarket price wars” between large retailers have also led to continuous downward pressure on food prices, which is a burden that food producers (farmers, fishermen) are ultimately forced to bear (Consumers International, 2012).

Processing and health

As discussed in section 1.3, food processing is increasing in both volume and complexity over the last decades. This trend in processing is also connected to the structure of the retail market.  While a majority of supermarket products once consisted of relatively basic raw ingredients and vegetables, large retailers increasingly make their profits from “value added” or processed goods (Deloitte Touche Tohmatsu Limited, 2014). Roughly 80% of supermarkets goods are processed and made by a decreasing number of manufacturing firms due to market consolidation with in this industry. These firms include manufacturers and traders such as a General Mills, Nestlé, Con-Agra, and others (OECD Competition Committee, 2013). While some of these processed foods are relatively benign, “ultra processed foods” that have a high level of additives, fats, salt and sugars and pose significant issues for general health trends in the countries where supermarkets dominate (Bloomberg, 2014). The majority of these manufactured goods are low in price, high in calories, and relatively low in valuable nutritional content (OECD Competition Committee, 2013). This has in part contributed to a global increase in food related illnesses such as heart diseases and diabetes (Bloomberg, 2014).  


Where giant retailers have controlled a large share of the food supply, market power has been increasingly recognized as a potential cause of monopolistic practices (Food & Water Watch, 2013). Recent OECD commission studies have looked at the overall competition within the food retail and manufacturing industry to assess the impacts of consolidation in the market. With fewer food retailers and manufacturers, consumer prices seem to be less closely tied to commodity prices and supplier revenue is decreasing (Giovannucci et al., 2012). This is because large retailers and manufacturers cooperate in buyer groups to buy bulk stock from suppliers and negotiate lower prices for raw food and commodities (Giovannucci et al., 2012). This is a trend which is reducing small farmers’ ability to get paid for the full value of their produce because of a lack of potential buyers and a loss of market power (OECD Competition Committee, 2013). 

In developed countries, the growth of large retailers is decreasing, having gone through its largest expansion in the early 2000s (Ruppanner & Mulle, 2010). While this decrease in growth is complex, it correlates with a growth in traditional food  and local food production and distribution systems in both Europe and the U.S. (United States Department of Agriculture, 2012). However, as urbanisation and wealth in developing countries increases, so does the global market share of the largest food retail firms (Nielsen, 2015).

Food retail channels by region

Figure 10: An overview of the different types of food retail channels in each region of the world.
(Adapted directly from Nielson “The Future of Grocery,” 2015).

Global average daily food consumption (2011)

Figure 11. Daily average global food consumption, divided into major food groups, in both mass and calories.
(FAO 2015b for food consumption volumes; USDA for average caloric data tables). Click expand

Globalisation has created space for large retailers to dominate over much of the developed and developing world. Today 51% of global food sales are purchased through supermarkets and hypermarkets. Food sales through these channels are growing at an annual rate of 2%.


1.6 Consumption

After a long and sometimes extremely complex journey through the global food production line, most food products finally reach their ultimate destination: the proverbial plate. Food consumption patterns largely dictate trends in food production through market response mechanisms, while consumption practices affect environmental and social outcomes.

Consumption patterns

Consumption patterns describe both the types and quantities of food consumed. The evolution of these patterns is constrained by food availability and prices. As countries develop, food expenditures tend to decline as a fraction of total household expenses. For example, in the U.S. and U.K., food budgets constitute an average of 10% of household costs. In many developing countries, food expenses remain a much larger percentage – for example, 70% in Tanzania and 45% in Pakistan (UNEP, 2012). 

In addition to prices determining the amount of food consumed, prices for different types of foods also affect dietary choices. Higher incomes and a lower fraction of income spent on food are associated with a shift towards a more nutritionally diverse diet and replacement of grains with animal products (Regmi, 2001). Within the boundaries of food availability and price, consumption patterns are largely determined by social, personal, cultural preferences and by knowledge.

Between 1950 and 2009, consumption of animal products doubled. If the trend continues, global animal protein consumption will quadruple by 2050.


Consumption quantities

In addition to prices determining the amount of food consumed, prices for different types of foods also affect dietary choices. Higher incomes and a lower fraction of income spent on food are associated with a shift towards a more nutritionally diverse diet and replacement of grains with animal products (Regmi, 2001). Within the boundaries of food availability and price, consumption patterns are largely determined by social, personal, cultural preferences and by knowledge.

Simultaneously, there is an increasing number of overweight individuals, both in developing and developed countries. Nearly 2.5 times as many people are overweight as undernourished, with cases of severe overweight (obesity) rising in parallel. There are a number of factors contributing to rising obesity rates, including food prices. In the United States and many other countries, crops like corn, soy, and wheat are subsidized, while fruit, vegetables and nuts are not (Mortazavi, 2011).  While prices for carbonated sodas (made with corn syrup) fell between 1980 and 2010, prices for fruits and vegetables rose (Powell, Chriqui, Khan, Wada, & Chaloupka, 2013). Processed foods are typically less expensive than fresh foods because they largely consist of cheap (often subsidized) ingredients such as grains, sugar, and oil. These foods also contain more calories when compared to their mass and nutritional value. Figure 11 illustrates how unevenly the mass of food consumed translates into caloric value. Consumption quantities and their surrounding trends are further discussed in sections 2.2 and 3.2.2.

Roughly 1/3 of Indians and 44% of those living in Sub-Saharan Africa suffer from undernourishment.


Global availability of calories per capita (1961) (KCAL)

Global availability of calories per capita (2009) (KCAL)

Figure 12: A comparison of global availability of calories per capita in 1961 and 2009.
(FAO, 2015b). Click expand

Nutritional quality

The composition of diets and quality of nutrition varies globally. Most high-calorie countries have high intakes of nutritionally insubstantial sugar and sweeteners, with North America ranking highest with 15% of calories coming from this category. Moreover, almost all high-calorie regions obtained more than 10% of their calories from meat, whereas low calorie regions obtained less than 5% of their calories from meat (FAO, 2011). Rising incomes and purchasing power results in a nutrition transition, with the largest impact being an increased amount of consumption of animal protein. Between 1950 and 2009, consumption of animal products doubled. If the trend continues, global animal protein consumption will quadruple by 2050, compared to 1950s levels (Nellemann, 2012).

Humans require a large volume of macronutrients such as protein, carbohydrates, and fats for growth, development, energy provision, and many other vital functions. Additionally, there are a large number of micronutrients which are necessary in smaller amounts, including other essential vitamins and minerals. Over- or under-consumption of vitamins and minerals can lead respectively to toxicity or deficiency. While nutritional deficiencies are often coupled with undernourishment, it is possible to consume a sufficient amount of calories and still suffer from a lack of important micro-nutrients. The most prevalent deficiencies of vitamin A, iron, iodine, and zinc in the diet, contribute to an estimated 19% of childhood deaths and 6% of DALYs (disability-adjusted life years) (Black, 2003).

Although there is an indication for increasing conscious and healthy food choices by some consumers, this is a small dynamic in the context of general consumer choices around the world. The concept of healthiness of food varies across cultures, and geographic regions. When considering consumer perceptions of healthiness more broadly, there are a few key points that influence their purchasing behaviour. For one, consumers may not be educated on which types of foods are healthy. Low levels of literacy and general education reduce the ability to understand nutritional labeling and thus the ability to make informed consumption choices (Wagner, 2014).

Food preparation

In addition to the quantity and quality of foods that are consumed, methods of preparing and storing food are important for overall outcomes of the food system.

The process of cooking is important for a number of reasons such as sterilizing harmful bacteria and other microorganisms, removing toxins, and increasing the availability of certain nutrients (Carmody, Weintraub, & Wrangham, 2012; Miglio, Chiavaro, Visconti, Fogliano, & Pellegrini, 2008). At the same time, cooking requires energy, contributing to emissions due to electricity production, and other fuels like wood and gas (Hager & Morawicki, 2013). Around 2.7 billion people rely on burning biomass for cooking globally, leading to further air emissions and health problems for those cooking indoors (IEA, 2014).

The bioavailability of protein is commonly measured by the percentage of nitrogen present that is retained (referred to as the biological value) and varies by source between around 60-70%. Plant-based sources generally have a lower biological value than animal-derived products, requiring a higher volume of protein consumption to ensure an adequate nitrogen balance (Reeds & Garlick, 2003). Proteins are denatured by heat, making them more easily digested by humans when cooked. Proper processing and cooking methods can also decrease anti-nutrients such as phytate, polyphenols, and oxalate, which reduce absorption of nutrients, while also increase bioavailability by freeing nutrients from chemical compounds. Such processing and cooking methods include thermal processing (boiling, steaming), mechanical processing (pounding), soaking, fermentation, and germination (Hotz & Gibson, 2007). 

In some instances however, cooking may reduce the nutritional value as a result of losses and changes in major nutrients, including proteins, carbohydrates, minerals and vitamins (FAO, 1990). In particular, cooking in water or oil which is then drained off and not consumed removes a large portion of nutrients, varying between 35-70% for different nutrients and raw foods (United States Department of Agriculture, 2007).

1.7 Waste

All industrial activities within the current economy, including agriculture, lead to the production of material by-products that do not have an immediate useful function, otherwise known as “wastes.” The food system is no exception to this rule, and is implicated in the generation of many kinds of waste, including, but not limited to: crop residues, agricultural plastics, chemically contaminated waste water, manure, food packaging, and food waste. These topics are individually dealt with in more detail in Chapter 3.

Some of the most salient statistics around waste in the food system to briefly mention in this overview chapter include:

  • An estimated 31% of all food (by mass) is wasted rather than consumed, representing a massive loss in embodied land, water, labour, and energetic resources (FAO, 2015b). Some estimates of food waste go as high as 50% of total production (IMechE, 2013). Figure 13 shows the fraction of food losses and waste taking place at the consumer stage across different geographic regions.
  • Solid waste from food packaging contributes up to half of the volume of municipal waste streams in many countries (Bournay et al., 2006).
  • The food system’s almost 30 billion animals produce over 200 billion tonnes of manure annually, much of which is inappropriately handled and contributes to global nitrogen cycle overloading (FAO, 2006).
  • 80% of all domestic wastewater is untreated, further contributing to imbalances in the global nutrient cycle and leading to “wasted” nutrient streams, which could otherwise be recovered for further use in the food system (UNESCO, 2003).

Per capita food losses and waste by region (kg/year)

Chapter 2

Trends and Behaviour

2.0 Introduction

The enormity of the food system is apparent from the data presented in the previous chapter. Since World War II, the system has tripled its output across many categories of foods to keep pace with population growth and changes in food demand patterns (FAO, 2015b). As we discuss in more detail in Chapter 3, the continued increase in resource throughput accompanying this expansion has placed ever greater stresses on both the biophysical resource base of the food system as well as the people and animals influenced by it. In this chapter we present some of the trends and underlying dynamics of the food system in order to better understand how its current shape and direction have evolved. The resource flows linked to different aspects of the food system have not grown uniformly. With regards to some parameters (land, greenhouse gas emissions) the food system has become much more efficient (though absolute throughput has still increased), whereas with regards to other parameters (pesticides, fertilizers), the food system has become much more resource-intensive over the period examined, with some recent signs of increased efficiency. In addition to looking at the quantifiable outcomes of the food system’s activities (food production, resource consumption), we also examine a few of the driving trends (population, GDP) and emergent behaviours (intensification, consolidation) that have shaped the system and characterise its current functioning. In the discussion section at the end of this chapter, we look at the implications of the food system’s current trajectory for the coming decades using the FAO’s business as usual projections for 2050 as a starting point. Despite its current enormity, the food system is poised for continued expansion due to projected increases in population growth and wealth. This projected increase in demand raises critical questions regarding limits to the system’s expansion under its historic model of development.

Key messages:

  • The Green Revolution played a significant role in establishing intensive agricultural production methods globally and shaping the reigning philosophies in mainstream agricultural practice. Though widely credited with helping avert anticipated large-scale food shortages in the post-WWII era, the intensification practices brought on by the Green Revolution have also been critiqued for driving ecological degradation and entrenching dependency on non-renewable resources like fossil fuels.
  • There is more food produced today per person than ever recorded. Both calories and grams of protein per capita have steadily increased since the 1950s. »Growth in yields has begun to slow in recent decades, with annual yield increases in cereal crops now growing on average at half the rate necessary to reach a (potentially necessary) doubling of food production by 2050. The genetic potential of major crops is being reached and land degradation as well as lack of investment in low-producing regions is leading to overall yield declines.
  • There is enormous global variability in yield, and the global yield gap between the most and least productive farms globally has increased dramatically since the 1950s.
  • The food system’s absolute resource use (water, pesticides, fertilizer, energy) has increased significantly over the period evaluated. However, resource intensity per unit of food output has been improving for certain resources. Emissions intensity measured in tonnes of CO2eq. per tonne of food has decreased.
  • Fertilizer and pesticide intensity have more recently begun to show signs of decline as well. These are indications that the system is becoming more efficient as it expands.
  • Key trends that have been driving the expansion pattern and structure of the food system include increases in global population, wealth, and urbanisation. These increases are associated with changes in consumer dietary preferences, which have led to the increased complexity and resource-intensity of average diets.
  • Policy-supported trends have also led to structural shifts within the food system. Notably, demand for non-food uses of crops, particularly biofuels and biomaterials, is putting significant pressure on the resource base needed to support continued food production.
    The food system exhibits several large scale behavioral trends including intensification, consolidation, specialisation, and regionalisation. As evidenced in steadily increasing yields, intensive practices now define much of the food system. Control of the system has consolidated onto a handful of actors in production, processing, and retail. Intra-regional trade now encompasses the majority of international trade, indicating a slow-down in the effect of globalisation towards a more regional model.
  • Funding for agricultural research and development is not evenly distributed across nations or production methods. This has allowed certain nations and regions to improve, while many low-income nations are excluded. Similarly, funding has been prescriptive in developing specific production methods, allocating little opportunity or funding for alternative practices to take hold.
  • A slowly growing counter-movement to the intensive practices brought on by the Green Revolution has begun to emerge in the form of alternative, lower-impact agricultural systems. However, these practices still make up a small minority of agricultural production worldwide and are generally under-researched. New practices and food processing techniques (advanced greenhouse horticulture, symbiotic agricultural systems like aquaponics, agroecological practices, vertical urban farming, alternative and synthetic protein products), present a small, but promising frontier for food system innovation.

Understanding the history of the food system and the origins of its current development patterns provides vital insights for shaping a more sustainable pathway for its further evolution. In this section we review some of the major trends that have characterized resource throughput in the food system over the last decades, some of the proximate drivers that have shaped these trends, and a few of the key emergent behaviours that have defined larger-scale patterns in the system. An important backdrop for any discussion about trends in the food system is an understanding of the major transformation of agriculture that took place in the 20th century known as the Green Revolution.

The green revolution

The Green Revolution refers to the decades-long technological development and transfer process, which lasted roughly from the 1930s to late 1960s, and centered around the implementation of intensive agricultural production methods that characterise present-day “conventional” agricultural practices (see section 1.2.7). The technologies implemented included  high-yielding crop cultivars, synthetic chemical inputs, mechanisation, modern irrigation, and monocultures (Fitzgerald-Moore & Parai, 1996). Asia was the primary beneficiary of the Green Revolution, where its practices led to unprecedented increases in yields of rice, maize, and wheat (FAO, 2000). 

In the 1980s and early 90s, trade negotiations and agreements such as NAFTA and the Uruguay Round formed new free trade relations, further aiding in the spread of Green Revolution practices (Brainard, 2001). Global markets were flooded with cheap agricultural goods, whose production was enabled by more intensive cultivation techniques. Local producers, who up until then used less-intensive methods, were pressured to adopt intensive agricultural practices in order to remain competitive on the global market. 

Norman Borlaug, the agronomist known as the “Father of the Green Revolution,” received the 1970 Nobel Peace Prize for his work and has been credited with saving over a billion people from starvation through the production increases associated with the new intensive practices (Easterbrook, 1997). Though it may have indeed helped avert global famine as broadly reported (FAO, 2011 ), the Green Revolution also led to many structural changes in the global food system, many of which are now viewed in a less-positive light.

One such example is the resulting increased dependency on fossil fuels and their derivatives, creating a lock-in effect that has been argued to undermine the structural resiliency of the food system (Pfeiffer, 2013). As Green Revolution techniques rely heavily on automation (and its associated fuel use) as well as fossil-fuel derived chemicals (fertilisers, pesticides), the agricultural system is now more tightly bound than ever to the volatility of the fossil-fuel market (see section 3.3). The long-term effects of the Green Revolution have also led to public awareness of environmental degradation issues associated with agriculture, including serious human health effects from pesticide use (Culver, Mauch, & Ritson, 2012). The  negative impacts of the food system, further discussed in Chapter 3, are broad and varied; many of these can, at least in part, be attributed to the intensification of agricultural practice that had its origins in the Green Revolution.

Global agricultural production, land use, and population between 1961 and 2009

Figure 14. Trends in global agricultural and food production, agricultural land use, and global population.
(FAO, 2015b). Click to expand

2.1 Outcome trends

It is likely that without the Green Revolution, the food system would not have been capable of undergoing the expansion that we have witnessed since the end of World War II, which has largely underpinned global capacity for providing an uninterrupted food supply for a growing, wealthier population.  

Agriculture occupies 38% of global land, consumes 69% of global fresh water withdrawals, and uses 30% of the world’s primary energy each year (AQUASTAT, 2014; FAO, 2012a; The World Bank, 2014a). In this section we survey some of the most evident physical trends that have accompanied the expansion of the food system to its current state, both in terms of absolute growth and relative efficiency. First we focus on the “outcome trends;” those that are often seen as performance metrics of the food system, rather than those that have been driving the changes at hand.

Food production

The amount of food produced globally more than tripled from 1961 – 2011, growing at an average rate of 2.30% per year. In 2011, 4.54 billion tonnes of food were produced (FAO, 2015b). In this time period, global meat and crop production more than tripled, growing to 205% and 209% above 1961 levels respectively, while global fisheries output quintupled (416%). Meat, crops, and fisheries production had annual growth rates of 2.26%, 2.28% and 3.34% respectively. Though meat and fisheries production have increased significantly, their collective share of production has remained relatively stable at around 25%. Fisheries, individually, have increased in share of production from 1.8% in 1961 to 3.1% in 2011 (FAO, 2015b). 

Global statistics obscure the localized nature of many of these changes. As described in section 1.2.4,  there are strong regional differences in food production, both in terms of type as well as volume of food produced.  Food production has grown irregularly throughout the world, continuing historical imbalances in food availability. Notably, as discussed in chapter sections 1.6 and 3.2.1, increases in global food production have not led to a commensurate increase in overall global food security, despite the fact that sufficient food is currently produced to provide nutrition for the entire population (FAO, 2015b). This emphasizes the critical importance of economic factors, such as poverty, in the question of food security.

A large part of the variations in food production globally derive from changing patterns in yields, which have also progressed at an uneven pace across regions.

Food production

From 1961 to 2011, global agricultural yield (both food and non-food) increased by 186% at a rate of 2.13% annually. In 1976, the global agricultural system crossed an historic threshold, reaching an average global production level of over one tonne per hectare. By 2011, global average yield had once again almost doubled since this previous milestone, reaching 1.988 tonnes per hectare. Figure 14 illustrates these evolving trends and correlation between population, agricultural output, and land use. The data clearly present a much higher increase in global agricultural production relative to a comparatively low increase in land use, demonstrating significant increases in food output per unit of land area.

Global tomato yield variation in 2011 (tonnes/HA)

Figure 15: An overview of total yield (tonnes per ha) for tomatoes. Only a few countries have been highlighted in this graph. (FAO, 2015b). Click to expand

Yield growth is slowing

The impressive gains in yield largely facilitated by the Green Revolution allowed for food output to exceed population growth for much of the 20th century.  Though population and wealth have continued to rise, recent empirical studies have shown that growth in yield has significantly declined since the early adoption era of intensive practices.

Yield increases for major cereal crops, which are responsible for nearly two-thirds of the calories delivered by agricultural production, are increasing at a much lower rate than they have historically. Ray et al. found that cereal yields are generally growing at an average of half the rate required to reach a doubling of global production by 2050, which is frequently cited as a target figure for avoiding food shortages by 2050 (Ray, Mueller, West, & Foley, 2013). Similar findings are echoed throughout the literature, concluding that the gap between average farm yields and genetic yield potential of major crops is closing and that land degradation is leading to overall yield declines (T. Robinson et al., 2011; Wirsenius, Azar, & Berndes, 2010).

Once again, however, though global statistics provide an important metric, the evolution of yields has varied greatly across regions and is greatly dependent on local context.

Variation in global yields

Variation in yield is enormous across products, geographies, and production systems. As a simple indication of the significant spread in yields, Figures 15 and 16 show the global range in average yields for two products: tomatoes and wheat (FAO, 2015b). The highest average tomato yield (in the Netherlands) is around 500 times greater than the lowest (Somalia). This is a much more extreme range than that which is seen for wheat, where the difference between highest and lowest average yields amounts to around a 10-fold difference. At the same time, maximum average tomato yield per hectare can reach masses of several hundred times than wheat yield (in the order of 500 tonnes per hectare versus 10 tonnes per hectare), showing the significant differences in yields inherent between product types (though it is important to note that nutritional density of these products is also highly variable). In short, certain agricultural products result in inherently greater production yields. These differences in yield result both from the inherent biology of the products and the agricultural practices implemented by farmers.

In addition to showing wheat yields, Figure 16 shows in parallel the total area harvested per country. The clear indication is that many of the countries with the largest areas planted are not the most productive. From this we can conclude that even moderate increases in yield in these low-yielding regions could have dramatically positive impacts on the global food balance.

The yield gap

The un-captured yield potential between what a crop could biologically and technically yield in a given context and what it actually yields is referred to as a “yield gap” (Van Wart, Kersebaum, Peng, Milner, & Cassman, 2013). The global yield gap refers to the total unexploited yield potential across farms globally. This topic is the subject of much study, since capturing this potential could reduce the need for the future expansion of arable land and contribute to improving farmer livelihoods. However, the full scope of the global yield gap is not currently known, because actual yield potentials are highly contextually variable (based on factors like local climate and soil conditions and the potential for irrigation). Efforts are underway to gain more fine-grained insight into the full scope of the global yield gap, through projects such as the Global Yield Gap Atlas (   

In many parts of the world, agricultural intensification has already run its full course exploiting the maximum genetic potential of crops. By contrast, there are many regions in the world where intensification practices were never introduced and yields remain exceedingly low (most notably in Sub-Saharan Africa). Combinations of factors that often go beyond mere technical performance have led to the stagnation of crop yields. These factors include declining research and investment and the increasing opportunity cost of labour (Reardon, Barrett, Berdegué, & Swinnen, 2009). More discussion on  this topic can be found in section 5.2.3.

Resource use

The primary instruments behind the increases in productivity and yield throughout the Green Revolution relied on an intensification of resource inputs such as water, fuel, fertilisers, and pesticides. These increases in inputs, as already discussed in the previous section, allowed for sharp gains in land-use efficiency at the expense of impacts in other parts of the system (see Chapter 3).  In this section we look at some of the trends surrounding the evolution of input use over the last fifty years. 

Global wheat yield variation (tonnes/HA) (in grey) and global land area planted to wheat (in blue)

Figure 16: This graph shows global weigh yield variation (in blue) compared to global land area planted for wheat (in grey).(FAO, 2015b). Click to expand


Land has consistently been a limiting factor to the global agriculture system’s expansion. The moderate growth seen in land use reflects the system’s limitations. From 1961 to 2011, the area of land devoted to food increased by 11%, with an annual growth rate of 0.2%. In 2011, all agricultural land (food and non-food) accounted for 4.54 billion hectares. The total expansion of agricultural land has amounted to roughly 500 million hectares since 1961 (FAO, 2015b).

Although growth in land use has been moderate relative to production trends, the impacts of land use change are often significant. The continued expansion of cropland and pastures is the primary driver of habitat disappearance and fragmentation globally, which in turn is the single largest cause of biodiversity loss (Convention on Biological Diversity, 2015). The conversion of natural ecosystems to agricultural land, resulting in the loss of their carbon sequestration potential, is also one of the more significant sources of global greenhouse gas emissions. Estimates for the contribution of deforestation to global GHG emissions have ranged from 6 – 17% (van der Werf et al., 2009), with more recent research suggesting 10% as the most likely figure (Baccini et al., 2012; Harris et al., 2012). Expansion of arable land is therefore considered highly undesirable, to avoid both biodiversity loss and climate change impacts. While there is some further availability of arable land, analysis shows that land suitable for pasture has been fully exploited worldwide (Robinson et al., 2011).

Despite the fact that expanding agricultural land is not a preferred direction, significant attention has been paid in research to understanding the existing potential for further agricultural land development. This has largely been in response to doubts concerning the feasibility of sufficiently increasing yields on currently developed land resources. The Global Agro-Ecological Zones (GAEZ) study conducted by IIASA and FAO, concluded that a total of 1.4 billion hectares of prime and good agricultural land that could be brought into cultivation if needed (Fischer et al., 2008). Though this assessment did not exclude lands used for pasture, it did exclude land currently under cultivation, forested land, protected land, or land already occupied by non-agricultural uses. In theory, this land could be brought into use for cultivation, though this would often come at the expense of pastures or require considerable investments in infrastructure, soil preparation, or disease eradication. 

Though this may sound like a positive prognosis, a majority of these suitable lands are considered too remote or costly to develop to be worth the investment. Moreover, most of this land is concentrated in just a few countries (60% of it is located in just 13 nations), which is a spatially insufficient distribution of this resource when considering regional demand for food production (Alexandratos & Bruinsma, 2012).

The impacts of and limits to land use change are further discussed in section 3.1.1 in relation to biospheric integrity and in relation to soil management in section 3.1.2.


Though irrigated agriculture covers only one fifth of arable land it contributes nearly 50 percent of crop production, indicating that continued water supply is one of the most critical inputs for increasing yields  (Steinfeld, H., Gerber, P., Wassenaar, T., Castel, V., Rosales, M., & De Haan, 2006). Since the 1960s, the area of irrigated lands has doubled, to around 300 million hectares. Areas limited to rainfed agricultural production face significant disadvantages in terms of yields. Various studies have indicated that global expansion potential for irrigation is limited for reasons including access to sufficient water resources as well as costs of development. The FAO estimates that 180 million hectares remain suitable for expansion, of which they estimate that around 20 million will be developed by 2050 (Alexandratos & Bruinsma, 2012). 

Global freshwater resources are irregularly distributed in both spatial and temporal context. A number of countries worldwide are significantly over-extracting their available water resources. Using more than 20% of renewable water for irrigation is considered entering the threshold of impending water scarcity. 22 developing countries have already passed this threshold, with 13 in the critical, “over 40%” class. On the regional level, North Africa and South Asia already withdraw 52 and 40 percent of their water resources respectively (Alexandratos & Bruinsma, 2012), leaving little room for expansion in these regions where yields are among the lowest globally and undernourishment remains pervasive. 

The impacts associated with the over-consumption of fresh  water are further discussed in section 3.1.3.


Fertilisers are essential for maintaining yield levels as they provide nutrients necessary to support plant growth and maintain soil quality. Over-application of fertilisers is also associated with the disruption of the global nutrient cycle and a plethora of negative impacts, which are further discussed in section 3.1.7. Synthetic fertilisers, derived from fossil fuel sources, were one of the most significant innovations of the Green Revolution.   

From 1961 to 2002, global fertiliser use increased by 353%, with an annual growth rate of 3.75% (FAO, 2015b). In 2002, global fertiliser use was reported at 141 million tonnes. Using fertiliser consumption rates per crop as reported by FAO and 2011 agricultural land use figures, total fertiliser use in 2011 was estimated at 200 million tonnes (FAO, 2007). In the early 1990s global fertiliser use declined significantly. This dip in fertiliser consumption can be attributed to changes in Eastern European growing practices caused by regional restructuring after the dissolution of the USSR.

With respect to yields, fertiliser intensity (tonnes of fertiliser / tonne of crop) increased from 1961 – 2002, but peaked in 1988 at 51 kg / tonne. An increase in fertiliser intensity is expected as yield increases, but a peak in fertiliser intensity suggests that the system is becoming more efficient with regards to fertiliser usage.

Tractors and other farming machinery are among the metrics for evaluating intensification.
Creative Commons: Wikimedia


Pesticide use also increased as the global food system grew. Globally, pesticide use more than doubled from 1990 – 2011, with an annual growth rate of around 2%. According to FAO data, pesticide use peaked in 2007 at 3.68 million tonnes (FAO, 2015b). This report estimates global pesticide use to be 4.4 million tonnes annually, based on per-crop pesticide demands. While reported quantities differ by source, the FAO data provide invaluable historic insight into global trends of pesticide use.

Pesticide intensity follows a similar path to pesticide consumption, peaking in 2007 at 0.42 kg of pesticide / tonne. From 1990 to 2011, pesticide intensity increased by 76%, but had more than doubled as of 2007. While global food production has steadily increased, pesticide and fertiliser use has declined. This shows a slow, but progressive, decoupling between yields and inputs. The impacts associated with the use of pesticides are further discussed in section 3.1.5.

Greenhouse Gas Emissions

Synthetic fertilisers and pesticides increase the food system’s overall energy consumption and emissions because of the high energy use associated with their production. Global emissions from agriculture, defined as IPCC tier 1 emissions, which include embodied emissions of inputs, almost doubled between 1961 and 2011, growing annually at a rate of 1.34% (FAOSTAT, 2015; Intergovernmental Panel on Climate, 2014). However, the largest contributor to agricultural emissions is enteric fermentation – which results in the release of methane gas from the digestive system’s of livestock – at 40%, while synthetic fertilisers account for 13% (Tubiello et al., 2014). When weighed against total agricultural production, the intensity of CO2-equivalent emissions has steadily decreased from 1961 to 2011 to 62% of 1961 intensity, which is a reduction rate of 0.85% annually. These trends show the increasing efficiency of the global food system with regards to greenhouse gas emissions.

Greenhouse gas emissions result in an important feedback loop with the agricultural system. Climate change is expected to have variable effects with regards to agricultural yields in different parts of the world (some positive, some negative), though on balance, it is projected to have negative impacts on yields in some of the most sensitive regions in the world (Alexandratos & Bruinsma, 2012). The impacts associated with GHG emissions and climate change are further discussed in section 3.1.4.


2.2 Driving trends

In this section, we focus on some of the underlying quantifiable trends that have served as drivers for the growth and transformation of the food system. Many such driving trends can be documented. Here, we focus on four which are broadly considered some of the most significant: the global human population, global human wealth (as measured in GDP), changes in consumer diets, and a significant shift towards the production of biofuels and biomaterials.


The vast growth in food and agricultural production can be partially attributed to global population growth. Global population more than doubled between 1961 – 2011, with an annual growth rate of 1.65% (FAOSTAT, 2015). As the food system’s ultimate function is to provide adequate nutrition to the world’s population, major increases in population challenge the food system to produce enough food to adequately meet demand.

While population increases help drive growth in food production, this does not present the complete picture. Food production has outpaced population growth, with food production per person increasing from 1961 – 2011 at an annual rate of 0.64%. In 2011, there were 669 kilograms of food available per person compared to 487 in 1961, an increase of 37% (FAOSTAT, 2015). By this measure, there is more food available per person globally than ever before. Looking more closely at this trend, the availability of energy from food, measured in kcal per capita per day, increased by 31% from 1961 to 2011 at an annual rate of 0.54%. Similarly, available protein, measured in grams of protein per capita per day, also increased by 31% from 1961 to 2011, growing at an annual rate of 0.54% (FAOSTAT, 2015).

Gross domestic product

Production growth has also been influenced by growth in global wealth. From 1961 – 2011, global GDP (constant 2005 USD) increased by 461% while per capita GDP (constant 2005 USD) increased by 148% (The World Bank, 2014b). Increased wealth grants populations access to more food both in quantity and diversity (Gerbens-Leenes, Nonhebel, & Krol, 2010). The significant growth seen in food production relative to GDP and population can be partly explained by the combination of growth in both metrics. The global population is larger and richer than fifty years ago, which has direct implications on food demand patterns and therefore production trends.

Changing consumer diets

The primary shift in consumption patterns since the 1960s has been a large-scale increase in the throughput of food consumption as a result of increases in population. In addition, as discussed in section 1.6, the past decades have witnessed a global shift towards more complex, processed, and resource-intensive diets. The increase in overall food consumption as well as changes in the composition of the global average diet have been driven by at least three underlying global trends: population growth, urbanisation, and increased wealth. 

Global economic trends are driving more people to move to urban areas (Madlener & Sunak, 2011). Urban consumers have access to the global food chain, and thereby a more diverse, nutrient-dense, and resource-intensive diet. Urbanisation is also often followed by increases household income. The process of dietary change has been described to follow two main stages upon the increase in wealth: an “expansion” phase followed by a “substitution phase (Kearney, 2010). The expansion phase is characterised by higher levels of consumption to provide increased caloric input, usually from cheaper, vegetable-based foods. The substitution phase involves a shift from carbohydrate-based staple foods to more desirable and expensive categori