Executive Summary

In complex systems, such as the European food system, solving problems which lead to biophysical and humanitarian impacts requires a new approach. When not considering the problems within the context of the system they exist in, attempts to fix the problems often lead to unintended and unforeseen consequences. 

In this guidebook, we first explain the basics of systems theory. Common patterns we see in system dynamics, which perpetuate problems, are often termed ‘system archetypes’. As a starting point, some of these archetypes are introduced with examples from the food system, to explain why it is difficult to break free of destructive cycles of behaviour. 

Following the introduction to systems theory, we explain our method for finding leverage for change using applied systems thinking. Using the analogy of how a lever works, we introduce the three steps we use to identify the actions which will create the type of outcome we are looking for. 

The first step is identifying the weights or impacts in a system that we want to shift to a more sustainable state. We describe which types of impacts should be prioritized over others, including:

• Long-term, irreversible impacts
• Impacts which undermine the ability of the earth to provide a safe operating space for humans
• Impacts for which the outcomes for people or environment have a high degree of uncertainty

The second step is to sketch a systems map, looking at the processes and system structures which lead to the impacts. The goal here is to identify the fulcrums, or processes, around which the system must shift to reduce impacts. We look specifically for processes which:

• Contribute to a large number of impacts across the system
• Lead to a disproportionately large part of a single impact
• Contribute to reinforcing feedback loops, which will perpetuate the problem unless interrupted

Finally, we must identify the lever itself, or the intervention which will ultimately shift the impact. We explain what interventions are and how they affect the system by returning to examples of archetypes. We prioritize interventions based on their feasibility and effectiveness. For feasibility, we look at economic, social, political, practical, and technical feasibility. When considering potential effectiveness, we consider not only the effect on the single target impact, but also the larger systemic effect, which can be either positive or negative.

While we have worked through this process based on our own experience and expertise, our conclusions are not necessarily the same as the conclusions that other stakeholders would draw. We offer this guidebook as a means to explain the process, which can then be applied by other groups of experts in different fields. For this reason, at the end of the guidebook we also offer our insights into using systems thinking in the process of strategy and roadmap development as an organization.

1 Introduction

1.1 About This Document

In your hands is an introduction to systems thinking and a step-by-step guide to identifying leverage using systems analysis. In this document, we explore the issue of leverage by working through the example of the European food system and discussing how this system can be transitioned to a more sustainable state using systems thinking. The main goal of this report is to show where the functioning of a system can be altered in a structural manner (where there is leverage within the system). Leverage can be defined as an opportunity to change the functioning of the system in order to transition the system as a whole to a more sustainable state. This report will guide you through a process of identifying leverage and thinking systemically about creating effective strategies and roadmaps for change. Based on our own multidisciplinary expertise, we have drawn our own conclusions along each step of the process, though this process can also be followed by experts in specific topic areas. 

The European food system is a perfect case for systems analysis. In the current body of research, it is clear that the food system is the single largest contributor to the depletion of global biodiversity and puts a tremendous pressure on natural and human systems alike. While the food system is already running up against physical boundaries within the global system, it is also poised for a necessary expansion. If we consider a business-as-usual scenario and population growth, food production may need to increase by 60% over 2006 levels by 2050 (Food and Agriculture Organization of the United Nations, 2016). This represents a larger increase from today’s production level than we have achieved through advances of the Green Revolution since the 1960s (Searchinger et al., 2013). At the same time, other demands for biomass (for example for biofuels and biobased materials) compete for land and resources. While it is possible to achieve a food system that works within the planet’s biophysical boundaries, inclusively supports human livelihoods, and ensures food security for a growing population, a fundamental change is needed in the system as a whole in order to achieve these diverse objectives simultaneously.

Chapter index

In this chapter we begin by providing a brief introduction to systems thinking and explain how it can contribute to effective strategy development, while chapters 2-5 serve as a guide to our process of identifying leverage. Chapter 2 provides an introduction to our process, while the following chapters illustrate this process step by step. In this way, we hope to guide the reader to a comprehensive understanding of how to apply a similar process to decision-making for high-impact solutions. At the end of this process, we also present our high-level conclusions on developing effective strategies and roadmaps using systems thinking in Chapter 6. The final chapter provides a reflection on the process and key takeaways for moving forward.

1.2 A brief Introduction to systems thinking

Before we get into the details of systems thinking, it is valuable to discuss what we mean by a system. We define a system as an interconnected set of elements that is coherently interconnected and organized in a way which produces a pattern of behaviours over time (Meadows, 2001). Systems exist at virtually every level of size and abstraction: a single cell and the universe are both systems, while both ecosystems (physical, concrete) and financial systems (immaterial, abstract) fall under the definition of a system. All of these systems are interconnected, and the definition of a discrete system boundary is arbitrary in essence, and defined based on the set of behaviours or outcomes that fits within a certain definition of purpose. The main purpose or function of the food system, for example, is to produce food for consumers, and the boundary is defined by this purpose. 

However, the main function of a system is not the only outcome. While the elements of the food system are organized in a way to provide food, it also leads to a range of additional outcomes, some of which we see as positive (employment and the provision of livelihoods, culturally-significant agricultural landscapes, etc), and some of which are negative (biodiversity loss, greenhouse gas emissions, poor labour conditions, etc). While we would like to address the negative outcomes of a system, due to the interconnected nature of a system, it is difficult to do so without unwittingly contributing to a problem somewhere else within the system. The behaviour or functioning of complex socio-ecological systems, such as the European food system, is difficult to predict, often leading to suboptimal decision-making. 

If we are to have any hope of changing a system for the better, we must first understand what causes the system to function the way it does. Systems theory is a framework which can provide tools for thinking about problems within the context of a system. One key insight within systems theory is that the outcomes or impacts of a system are ultimately caused by its structure. Behavioural patterns are rooted in systemic structures such as biophysical conditions, markets, and political institutions, which are in turn influenced by the mental models and perceptions that guide our decision-making and the establishment of structures.

Building on this notion, Maani and Cavana have introduced what has been called the “four levels of thinking” model, which illustrates these interrelationships. The model shows the hierarchical relationship between four related, but different levels within a system: events (which are impacts or symptoms), patterns (behaviours), systemic structures, and mental models (paradigms) (Maani and Cavana, 2007). 

As shown in Figure 1, events represent only the “tip of the iceberg”. However, because they are most visible and immediate, most policy attention is directed here: at the “end of the pipe,” whether literally in terms of pollution clean-up or figuratively through policies that are aimed only at addressing the symptoms of a problem. Systems theory can provide tools for understanding complex systems and help us move beyond “end of pipe solutions” to address deeper structures and mental models that are at the root of the problem.

1.3 The value of systems thinking as a tool for problem solving

Holism vs reductionism

The complexity of systems often gives rise to suboptimal decision-making when decisions are made by looking at a problem in isolation, instead of within the context of the system in which it occurs. One example of such consequences in the food system is the impact of food aid. According to Mousseau (2005), despite its purpose being the eradication of hunger, “food aid is integrated into policies leading to structural food deficits and increased dependency on food imports.” This kind of dependency, in combination with the limited resources that the poorest countries have to finance their imports, has actually led to an increase in poverty and hunger (Mousseau, 2005).

Archetypes

Some well-known patterns of behaviours which frequently lead to impacts in systems are called archetypes. The dynamics of these problems and how to best address them are well understood, and yet, the same types of mistakes in decision-making continue to occur. Some of these archetypes include the tragedy of the commons and burden shifting (reducing symptoms in the short-term erode the ability to address other goals). Here we provide a few examples of archetypes within the food system and we return to a discussion of archetypes later in the report to evaluate different strategies.

Patterns of behaviours which frequently lead to impacts in systems are called archetypes.

Fixes That Fail archetype

The fixes that fail archetype describes a scenario in which short-term solutions increase the problem in the long term. The cycle is a balancing loop within a stabilized system, which perpetuates behaviour patterns over a long time, often in undesirable ways. Donella Meadows writes that “despite efforts to invent technological or policy ‘fixes,’ the system seems to be intractably stuck, producing the same behaviour every year” (Meadows, 2008). Looking at the food production industry, we can find a classic example of this archetype: the solution to strengthen the short-term resilience of livestock and aquaculture through antibiotics increases their vulnerability in the future. 

The issue

Antibiotics are one of medicine’s most powerful tools against potentially deadly infections. Misuse of antibiotics however, leads to antibiotic-resistant strains of bacteria, posing a health risk for humans. The majority of antibiotic use globally, and more specifically in Europe, is attributed to livestock use (Landers, Cohen, Wittum, & Larson, 2012, Who, 2011). Antibiotics are poorly absorbed in the gut of livestock animals and make their way into the environment after being excreted. They have been found in groundwater, surface waters, and marine sediments. Some of the antibiotics fed to animals end up in our food supply and environment, contributing to the increasing problem of antibiotic resistance in strains of bacteria (Landers et al., 2012). 

How it works

Looking at the causal loop diagram in Figure 2, we can see that the state of the food supply system is determined by both its demand in the market, as well as the use of antibiotics (which ensures this supply of livestock). However, the corrective action of using antibiotics can also lead to unintended consequences; the more the cycle continues, the larger the likelihood for increased antibiotic resistance, which has a detrimental effect on livestock yields in the long-term. 

Moving beyond the archetype

The only way to prevent this consequence is to find an effective point at which to break the cycle. One intervention could be to bypass the cycle entirely by increasing livestock yield to meet the demand through means besides the use of antimicrobials (for example through changes in husbandry and effective biosecurity routines). Another possibility could be to change one element to prevent just the unintended consequence. For example, the cycle could remain unchanged while the element of antibiotic use could be altered, restricting use of newly discovered antibiotic classes to human health care only (Bengtsson and Greko, 2014), thereby also preventing the unintended consequence of antibiotic resistance.

Addiction archetype

The addiction archetype describes a more specific scenario of the fixes that fail archetype, in which a short-term solution is applied again and again, deteriorating its effect at each passing. The more the temporary fix is used, the less effective it becomes, and the worse the long-term problem becomes. In crop production, this is widely seen when farmers utilize fertilizers to increase yields in ways that deteriorate long-term soil quality.

The issue

Many modern agricultural practices include using synthetic fertilizers from industrial production, which means that nutrient cycles are not dependent on the boundaries inherent in natural systems, which often results in nutrient imbalances (Billen, Garnier, & Lassaletta, 2013). In Europe in 2012, around 21,8 million tonnes of nitrogen were deposited on agricultural land, 46% due to fertilizer use. At the same time, only 13,9 million tonnes of nitrogen was utilized by crops, leaving 8 million tons, or 36% of the nitrogen left in the soil (Eurostat, 2012). The nitrification of excessive nitrogen fertilizer (through which bacteria turn ammonia into nitrite), can lead to acidification of soils (Han et al., 2015). This, in turn, leads to decreased soil fertility, lower crop yields, and poor water quality.

How it works

In Figure 3, the main cycle represents the reinforcing feedback loop of using fertilizer to provide the nutrients for crops, while the smaller cycle to the right indicates the consequences of this addiction: the decreasing soil quality. Although adding nitrogen and phosphorous to the soil through fertilizers might create a higher yield on the short-term, it also leads to a nutrient imbalance in the soil. Over time, this reduces the soil quality by creating suboptimal conditions (for example for soil microfauna), which in turn deepens the dependency on additional fertilizers.

Moving beyond the archetype

Although farmers may try to treat soil by adding lime to reduce acidity, or attempt to recover nitrogen and phosphorous from agricultural waste streams in order to create new fertilizer, these practices essentially address the symptoms of the problem and do not change the basic nature of the nutrient dependency cycle. 

At the same time, simply removing the use of synthetic fertilizers is not possible without losing the positive outcomes of the cycle. With drug addictions, treating the addiction requires not only removing the drug, but also providing an alternate pathway forward. Similarly, a more effective long-term solution would replace fertilizer use with retaining high levels of nutrients in soils by cultivating a healthy soil ecosystem of microbes and insects. This solution would involve a number of practices to avoid excessive disruption of the soil environment, which may be either mechanical or chemical. A number of agroecological approaches lend themselves to this approach.

2. Determining leverage in complex systems

2.1 What is leverage?

Leverage points are often described in systems thinking literature as a kind of “silver bullet” solution to solving systemic problems. The idea is that simply by first identifying and then focusing effort on leverage points, your efforts can result in a disproportionately large positive impact with very little effort. However, this description is somewhat misleading. In reality, leverage points, as described by Meadows (1999, 2001), are simply any place in the system where change in one element or process within the system can have a systemic effect. Discovering leverage points is not the solution, as a leverage point alone does not tell you how the system is being changed, or whether the change is in a positive direction. 

Leverage points are a place in the system where change in one element or process can have a systemic effect.

What we are actually looking for is not only effective leverage points (or the place where change needs to occur), but also an understanding of what it is we are trying to change and how we are going to change it. Effective leverage occurs where the mechanism for change is feasible, and, when enacted, will shift the system in a desirable direction (one in which a target outcome is achieved while minimizing other non-target effects).

We use the analogy of a lever as a starting point for explaining leverage (see Figure 4). The idea is that you are trying to shift the outcome of the system away from the state in which the negative impacts are occurring. In the leverage analogy, the impacts or targeted outcomes are similar to the weights you are trying to move. Some examples of weights include soil erosion, CO₂ emissions, or food waste. The leverage points, or fulcrums, can be described as the contributing elements or processes which need to be shifted in the system to create change. This could include soil management practices, vehicle purchasing, or food storage and cooking. Finally, the levers are the mechanisms or strategies for change, which are made up of different interventions that can be implemented, such as education, policy, or subsidies.

2.2 Explanation of methodology

Defining what high leverage is makes it clear that we are not only searching for one thing in a system, but the combination of three things, which together define a high-leverage strategy. We ask three big questions to find leverage and we follow a three-step process, which mirrors the analogy, to identify where to focus. The three big questions and the steps we use to answer them are given in the table below.

The structure of the following three chapters mirrors this three-step process:

The first question in Chapter 3 is:

What needs to be changed the most urgently?

We identify and highlight the impacts we see in the European food system that are the most problematic. There is a short discussion of how you should prioritize impacts, which can be used as a guide for organizations to internally determine priority areas.

Chapter 4 answers the question: 

Where does change need to happen to transition the system to a more sustainable state?

Using systems analysis, the essential processes that must be interrupted are identified. We identify these processes by implementing a systems mapping exercise, connecting outcomes with systemic elements that act as drivers. This exercise is one that organizations can also use to visualize and understand how a system is functioning.

Chapter 5 asks: 

How is it possible to effectively interrupt or alter the processes that need to be changed?

A few essential processes are illustrated using archetypes, in order to provide insight into where interventions are possible. We introduce a framework for evaluating feasibility and effectiveness of interventions and provide a sample evaluation and conclusions. 

3. Which problems need to be addressed most urgently?

To identify which parts of the European food system are most in need of change, we must first answer the question of where the food system leads to the greatest impacts, both directly and indirectly. In this chapter we facilitate the conversation by providing an overview of the most important biophysical and humanitarian impacts of the European food system that we have identified out of the sub-selection of topics we investigated.

Metabolic previously developed a framework for impacts within the food system (Gladek et al., 2016), where we identified 11 main impact categories, which we apply again here:

Europe contributes to these impacts related to the food system directly, through production and consumption of food, and indirectly, mainly through trade, knowledge transfer, and development aid. On the following pages, we briefly illustrate the extent to which the food system contributes to each of these impact types with infographics of direct and indirect impacts. The purpose is not to be exhaustive, but to provide a quick look into the most pressing issues.

3.1 Largest impacts to address in the EU food system

Considering all the different types of impacts associated with the global food system, we carried out research on the main issues at play on the European scale. While our research is far from exhaustive, there are a few criteria which help to identify which impacts to prioritize, including: 

Long-term, irreversible impacts, which should be prioritized above all else 

• Tipping points in the European food system should be identified and carefully avoided. Aside from the extinction of key species and biodiversity loss, other areas where tipping points in the food system are likely to be found include the effects of climate change, disruption to biogeochemical cycles through the input of excess nutrients, or loss of cultural heritage through the extinction of languages, practices, or institutions. 
• Once complex properties of the system such as biodiversity, or culture and heritage are lost they are likely never to emerge in the same form again. 

Impacts which will undermine the ability of the earth to provide a safe operating space for humans 

• The Stockholm Resilience Centre’s Planetary Boundaries define limitations for different biological levels in which humans can safely thrive. Boundaries which have already been crossed or are nearing these thresholds should be prioritized. At the moment, the boundaries for biodiversity loss, the nitrogen cycle, and the phosphorus cycle have been crossed, and these issues should receive the most attention for this reason.

Impacts for which the outcomes for people or environment have a high degree of uncertainty

• The emissions of novel entities and atmospheric aerosols are examples of impacts for which the effects are unknown or uncertain. For these impacts, it is impossible to define safe limits or understand what the long-term effects are of exceeding limits. According to the precautionary principle, these risks are unacceptable outcomes to allow to continue.

On the following pages, we highlight some of our main takeaways in terms of the largest and most significant outcomes within Europe itself and on the global system as a result of the European food system, considering both the criteria just described and the scope and magnitude of certain impacts. This overview depicts some of the impacts, or ‘weights’ in our analogy, which we see as the most pressing issues to consider based on our criteria. However, other organizations with different expertise or focus may prioritize other issues beyond the ones we have highlighted. While the following chapters go into further detail on some of the problems we identified, this same process can be followed irrespective of the target outcomes selected as a starting point.

3.2 Direct outcomes of the European food system

Long-term food production in Europe is under threat from current agricultural practices

• Soil erosion is higher than formation in Europe, while agriculture is responsible for 70% of erosion (Verheijen, Jones, Rickson, & Smith, 2009).
• Desertification is already a problem in Mediterranean countries of Europe (Yassoglou & Kosmas, 2000).
• Without insect pollination, agricultural production in Europe would be reduced by 25-32%, while agriculture itself is one of the main contributors to pollinator decline (Zulian, Maes, & Paracchini, 2013).
• It is estimated that honeybee populations in particular have declined by around 25% in Europe in recent decades (Greenpeace Laboratories Research, 2013).
• Around 12,6% of agricultural land lies within river basins where water extraction is higher than recharge (World Resources Institute, 2015, FAO, 2015).

Europe’s traditional foods and food production systems are disappearing

• The Mediterranean diet has declined by 42,5% in Mediterranean European countries within the past 40 years (Da Silva et. al., 2009).
• From 2007 to 2010 small farmers owning less than 10 ha lost control over 17% of their land, an area bigger than Switzerland. In contrast, the arable land occupied by large farms in the EU slightly increased (+4%) between 2003 and 2010 (European Parliament, 2015).
• Intensive farming in Europe is well on the rise as cropping intensity is projected to increase from 63% in 2009 to 83% by 2050 (FAO, 2011).

Europeans exceed their fair share in terms of food production inputs, contributing to a greater share of the impacts

• Europeans make up only around 11% of the global population, while around a third of global fishing vessels are European (FAO, 2015).
• Europeans consume 45% of global pesticides (De, Bose, Kumar, Mozumdar, 2014).
• 34% of all tractors in the world  (8,5M) are in use in the EU (WorldBank, 2000).
• Out of the global seed market, Europe is responsible for a 20% share of the market (Ragonnaud, 2013).

European consumption habits are unhealthy and unsustainable

• 60% of Europeans are overweight or obese, which may rise to 90% by 2030 (International Association for the Study of Obesity, 2015). The respective global obesity levels are nearly 30% (McKinsey Global Institute, 2014).
• In 2001, 43,5% of total food expenditure in the EU was processed foods (Gracia, 2001), while food processing is highly energy intensive;  28% of embodied energy in European diets is due to food processing (Eurostat, 2015), while 1000 calories of energy is used during food processing for every single calorie in the final food product (Horrigan, Lawrence, and Walker, 2002).
• Europeans waste approximately 280 kg food per capita per year, compared to  the global average of 210 kg (FAO, 2011).

Food production is one of the largest contributors to environmental degradation within Europe

• Aquaculture alone is responsible for 13% of all aquatic invasive species introduced to Europe (Nunes, Katsanevakis, Zenetos, & Cardoso, 2014).
• Food production activities threaten more than 500 endangered and threatened species within Europe (IUCN, 2015).
• In 2012, around 7,8 million tonnes of nitrogen was deposited on agricultural lands that was not utilized by plants and entered the environment (Eurostat, 2012), contributing to eutrophication. 
• Around 62% of the nitrogen entering rivers and 55% entering marine environments originates from agricultures (van Grinsven et al, 2015).
• Around 3,8 kg pesticides were applied per hectare agricultural land in 2010 (FAO, 2015). 

Europe’s food system is highly consolidated, while smaller players find it difficult to keep up

• The top 6% of large farms controlled 67% of agricultural land in Europe in 2010 (Eurostat, 2015).
• In 2010, nearly two-thirds of the cattle in Europe was kept on holdings with more than 100 LSU (standardized livestock units), while around 70% of livestock holdings with pigs had 1000 or more pigs (Eurostat, n.d.).
• The top 10 food retailers hold a nearly 40% market share. (Regiodata, 2014).

3.3 Indirect outcomes of the European food system

Europe is a major exporter of key agricultural inputs, knowledge, and technologies

• The top three agrochemical companies globally are European. Bayer, Syngenta, and BASF alone made up 49% of the global agrochemical industry in 2007 (ETC group, 2008).
• Development aid from Europe has provided many small-scale farmers with seeds, tools, and cash to help them escape poverty (European Union, 2014).
• Five of the top ten seed companies globally are European. Bayer, Syngenta, and Limagrain alone make up 27% of the global vegetable seed market (Ragonnaud, 2013).

Europe controls a large share of the global food chain

• Of the top 100 largest supermarket chains globally, 63 are owned by Europeans (Regmi & Gehlhar, 2005).
• 35 of the top 50 largest food processors globally are European companies, including two of the top three: Nestle and Unilever (Regmi & Gehlhar, 2005).
• The EU market share in global exports is shrinking- from 20,4% in 2000, to 17.8 in 2011- however it remains the world’s largest exporter of food and drink products (FoodDrinkEurope, 2012). Food exports from Europe can push smaller farmers out of business in importing countries, particularly when subsidies lower the prices of European food and distort markets (Willem et al, 2014). 

Europe is a net importer of global embodied impacts of food production

• 42% of Europe’s water footprint is displaced to countries outside of Europe (Food and Agriculture Organization of the United Nations, 2010, Steen-Olsen et al., 2012).
• The EU imports 10% of the world’s embodied deforestation (European Commission, 2013).
• Europe is also a net importer of land footprint (Steen-Olsen, Weinzettel, Cranston, Ercin, & Hertwich, 2012).

Europeans consume more than their fair share of global food production

• Europeans make up only 11% of the global population, but consume 15% of the world’s food (FAO, 2015). 
• Europeans consume around 34% more calories than the global average (FAO, 2015).
• Europe imports a particularly large share of the world’s production of specific commodities. Some examples include coffee and cocoa, of which Europe consumes around 60% (FAO, 2015).
• Europe imports nearly 1/3 of Brazil’s soy production (van Gelder, Kammeraat, Kroes, 2008).

Europe plays a strong role in international affairs that affect global food production

• The European Union is the largest donor of official development assistance (European Union, 2014).
• Some development aid in the past was targeted towards funding commodity associations and supporting commodity sectors, ignoring staple crop production and small-scale farmers (Pelum Association y Practial Action, 2005).
• Structural adjustment programs associated with World Bank or IMF funding often increases dependency of recipient countries on aid and reduces food self-sufficiency by encouraging farmers to switch to export crop production (Heidhues & Obare, 20111, Mousseau & Mittal, 2006, Shah, 2007).

4. Where does change need to happen to transition the system to a more sustainable state?

4.1 Identifying where in the food system to focus

Now that we have identified the main outcomes and problems that we would like to address in the food system, we need to answer the question of where, in the food system, to make change happen. After identifying some of the main impacts within the European food system, the next step is a systems mapping exercise to better understand the interconnections related to the target outcomes. We performed a systems mapping exercise over the course of a week with a small interdisciplinary team at Metabolic, which is described below. This process could easily be replicated with any team of experts by following the same approach.

Figure 5 provides a simplified illustration of the systems mapping exercise. In this process, you begin with the target outcomes broken up into smaller problems, as shown on the right side of the figure. The next step involves identifying all of the primary processes and behaviours (e.g. fertilizer application) which are drivers of these target outcomes. These processes and behaviours are then linked to structures (e.g. subsidy structures, market prices), which are subsequently linked to the core system goals and mental models (such as paradigms), which drive the entire system. This provides the baseline for seeing the system as a whole and beginning to understand the interconnections which result in impacts. 

After the system is mapped out, from the impacts to the system goals and mental models at the core of the system, we quantify the connections as much as possible with available data to provide insight into the scope of different connections. This information enhances the picture, now depicting both the number and the strength of driving elements.

After mapping the system, we used a set of criteria to identify the most essential processes which are crucial in determining the negative outcomes of the system. The criteria we use for selection of these processes from the systems map include: 

1.  Processes which contribute to a large number of different impacts across the system
2.  Processes which lead to a disproportionately large part of a single impact, and
3.  Processes which contribute to reinforcing feedback loops (think back to the archetype of addiction presented in the introduction) and will perpetuate impacts endlessly, unless interrupted

4.2 Most important EU food system processes

Using our criteria, we identified 22 processes and behaviours which we believe contribute most significantly to the negative impacts of the European food system. We estimate that shifting these processes to a more sustainable state, or in some cases interrupting them entirely, would address the largest share of the impacts that arise from the system. The identified processes and behaviours are: 

It is important to note that many of these processes lead to positive as well as negative outcomes, depending on how they are implemented. For this reason, sometimes it is more desirable to transition the process to a more sustainable version (and preserve the positive impacts it leads to where possible), while at other times it may be more desirable to transition away from the process entirely (when it does not also lead to positive outcomes). 

The following chapter goes into details of how to intervene in the system to create change, but first, we explore the processes in more detail to illustrate where the process fits into the larger system and where interventions are possible. While we worked out this picture around each of the 22 processes, we only provide two examples here for the sake of brevity. The two processes selected for illustration on the following pages are synthetic fertilizer application and exports of European food, which are particularly relevant for the current hot topic of reform of the European Common Agricultural Policy (CAP). Additionally, these processes are associated with priority impacts, as discussed here:

Synthetic fertilizer application contributes directly to one of the most severely transgressed planetary boundaries: biogeochemical flows. Additionally, through acidification and eutrophication of soils and water, it contributes to biodiversity loss, the other severely transgressed planetary boundary. In addition to these effects, long-term nutrient excesses decrease soil quality and undermine soil productivity, which is a threat for future food security.

The export of European food to global markets currently leads to indirect effects which threaten the livelihood of farmers, food security, cultural diversity, and biodiversity indirectly. With subsidies in place which depress European food prices, farmers in developing countries find it difficult to compete. When not pushed out of farming entirely, many farmers adopt Western production methods and crop varieties in order to compete, leading to biodiversity loss. Exports of highly processed European food leads to lower quality diets and less traditional diets.

4.3 Process infographics

Synthetic fertilizer consumption

Whereas traditional agricultural systems rely on semi-natural production systems for fertilizer, current production systems have shifted towards synthetic fertilizers which are not naturally limited. Now that it is possible to create synthetic fertilizer with industrial production, the use of fertilizer is entirely disconnected from natural systems (Billen, Garnier, & Lassaletta, 2013).

Synthetic fertilizer production requires fossil fuel use, contributing to resource depletion and emissions. In addition, the excess amounts of nitrogen and phosphorus included in these types of fertilizers cannot be fully absorbed by plants, which creates a situation of an overly nutrient-saturated soil, which can affect crop yields (Spectrum Analytic, 2014). Additionally, nitrification of excess nitrogen can cause soil salinization (Han, Shi, Zeng, Xu, & Wu, 2015). The application of only nitrogen without carbon can lead to a decrease in soil organic matter, as a result of carbon mineralisation from acidifying fertilizers (G. L. Velthof et al., 2014). Soil organic carbon is highly important for agricultural productivity as it increases the fertility of soils, as well as aeration, water infiltration, pH buffering, and provides a food source for soil-forming microorganisms (Jones, Hiederer, Rusco, & Montanarella, 2005). These effects on agricultural soils contribute to increased erosion and degradation and threaten long-term productivity of Europe’s agricultural systems.

Finally, excess nutrients escape to water bodies, which can result in killing off aquatic life and becoming a human health hazard. Higher nutrient loads in the environment and nitrogen deposition can affect the composition of species, leading to the loss of biodiversity. Oligotrophic species that thrive in nutrient-poor soils are replaced by eutrophic species that thrive in high nutrient environments, which has been observed in European forests (Dirnböck et al., 2014). The largest effects of excess nutrients however, are in aquatic environments. Nutrient-rich water entering marine ecosystems via European rivers leads to increased growth of phytoplankton and algae, which can deplete marine oxygen concentrations and promote the risk of harmful algal blooms. These effects lead to loss of benthic fauna and the risk of shellfish poisoning for humans (Romero et al., 2013). 

In addition to all of these direct effects of synthetic fertilizer consumption within Europe, Europe is also a net fertilizer exporter (Romero et al., 2013), contributing to indirect effects globally as well as locally.

Exports of European food

The exports of European food (also in the form of food aid) is one of the main mechanisms by which the European food system affects global food security, labour and livelihoods, and culture and heritage. Ultimately, food exports affects the agriculture of developing countries and results indirectly in effects on biodiversity. This topic is closely related to European agricultural subsidies, which some might say leads to food dumping and gives European producers an unfair advantage over smaller farmers in developing countries without high subsidies. The Overseas Development Institute (ODI) conducted a study of CAP and its potential reforms and concluded that even though subsidies for exports have been removed from the CAP, a number of trade-distorting features remain in the current CAP, such as import tariffs, direct decoupled payments, and export subsidies  (Willem et al., 2014). Though Europe is shifting away from food aid as a type of development aid, it still contributes to floods of cheap European foods on markets of developing countries.

Cheap food in the markets of developing countries reduces the opportunities for domestic farmers to enter the market or gain access to inputs and knowledge necessary for food production, which decreases food security. Farmers struggling to compete with European prices often switch to more industrial farming methods and crop varieties. The result is a loss of crop genetic diversity and farmers using less sustainable agricultural practices, which can lead to biodiversity loss and reduce the productivity of agricultural land over time. While switching to Western farming methods means a loss of traditional practices, the export of cheap processed food can also reduce the consumption of traditional foods, leading to a loss of culture and heritage through dietary changes. 

5. How is it possible to interrupt the processes?

5.1 Formulating interventions

What exactly are interventions?

After we have identified the most important impacts or target outcomes, as well as the processes which need to be addressed to affect them, we need to decide how to make the right kind of changes happen. Interventions are actions which interrupt an undesirable process in order to change it. The goal of an intervention is the prevention of negative outcomes, which would perpetuate over time without systemic change. While leverage points are targeted processes of key influence within a system, interventions are the mechanisms by which the change happens. A leverage point within the European food system, such as the use of synthetic fertilizers, may be influenced by a variety of interventions such as fertilizer taxes, environmental regulations, certification schemes (e.g. SKAL), or consumer awareness campaigns. 

In the previous chapter, we mentioned that interventions can be targeted at two different levels: to prevent a process from occurring or to maintain the process but improve it to avoid the undesirable outcomes. While this is a useful way to frame what it is you are trying to interrupt with an intervention, in reality this is an oversimplification. Before we explore feasibility and effectiveness of different interventions in the following sections, we return to the examples of archetypes, applied to issues around synthetic fertilizer use and exports of European food to illustrate possible interventions. Mapping out archetypes aids in seeing where interventions would actually address the problems and can also provide insight into potential trade-offs or unintended consequences related to these interventions. These two examples show the archetypes, along with the locations of potential interventions.

5.2 Evaluating interventions

Types of interventions

For each of the 22 processes, we produced a sample set of interventions. First, we researched general mechanisms for creating change that are already within the realm of what organizations can work on to create change in the European food system, though this list is by no means exhaustive. The interventions fall into one of the following categories of activities we define.

Evaluating the outcomes of interventions

To evaluate the total expected impact of an intervention, we use the formula:

Impact = Feasibility x Effectiveness

Feasibility and effectiveness are evaluated separately and then brought together for a final score, as described here. Our team worked through the evaluation, though a similar process could be implemented by any group of experts.

Step 1: Evaluating Feasibility

With feasibility, the main question is: How difficult will it be to successfully implement the intervention, given the barriers that exist? The feasibility of the interventions were ranked according to five dimensions of feasibility:

• Social feasibility: The relationship between the effort put into creating change through the intervention and the chance of the change occurring. In this case, the effort would take into account the number of people whose minds would need to be changed, while the chance of occurrence would take into account potential social barriers, such as conflicting paradigms.
• Economic feasibility: The relationship between the amount of money that would need to be spent to create change and the chance of that change occurring. For example, R&D and lobbying has a generally low economic feasibility since they are costly and have no guarantee of success.
• Practical feasibility: The relationship between the practical barriers to making change and the chance of change occurring. Practical barriers would include the number of people that would need to be mobilized, schedules, distance that would need to be travelled, and any physical limitations that exist. 
• Political feasibility: The relationship between political barriers and the chance of change occurring. For example, if an intervention is difficult to promote in the current political system and the chance of the change happening is also very low, then political feasibility would be low.
• Technical feasibility: The relationship between the state of current technology and knowledge and the level that is needed to achieve the desired change. In other words, this dimension would take into account whether or not we have the means to execute the desired plan.

The interventions were ranked within each of the dimensions and then the dimensions themselves were ranked according to their relevance for changing a certain process. The result is a matrix where the most feasible interventions end up in one quadrant (see Figure 10).

Step 2: Evaluating Effectiveness

The main question regarding effectiveness is: If the intervention is fully implemented, what share of the negative impacts of the system can be mitigated? A very feasible intervention, for example, is less of a priority if it has a negligible effect. 

To determine effectiveness of interventions, we consider what the maximum theoretical effect of an intervention would be. For example, the maximum theoretical effect of banning an agricultural practice would be that practice being reduced to near zero, while education of farmers on a new practice would be limited by the adoption rate. From the theoretical maximum effect, we try to determine the share of the main impact that the intervention would mitigate. 

In addition, it is important to consider potential systemic effects: would the intervention only change a single outcome, or have an effect in reducing impacts across the system? While we ranked the interventions in terms of total effect on the main impact area of concern, we also want to take systemic considerations, such as cascading effects and potential trade-offs, into account as well. For this reason, we also rank the interventions on their potential systemic effects. The final result is a matrix similar to the one for feasibility, as shown in Figure 11.

Step 3: Evaluating impact-feasibility and effectiveness

Finally, we can score the interventions in terms of their total expected impact. The scores for both feasibility and effectiveness were normalized to values between 0 and 10 and then multiplied for a final score of 0-100. We calculated the outcomes for feasibility, effectiveness, and total impact, for each of the 22 processes. For illustration, we present here the results for synthetic fertilizer use and exports of European food.

Synthetic fertilizer use:

A. Fund research on symbiotic agricultural processes and technologies that increase soil fertility

B. Lobby for subsidies to implement symbiotic processes and technologies which increase soil fertility without fertilizers

C. Lobby for taxes on synthetic fertilizers 

D. Provide information to farmers on desirable practices amongst farmer networks through social/new media (through innovative projects)

E. Encourage and support European food processors and retailers to implement practical training programs, web resources for farmers that promote the income diversification benefits and cost-savings from practices that increase soil fertility (such as polycultures, etc.).

F. Fund research and pilot projects for closing nutrient cycles in urban regions through urban agriculture.

G. Publish research into techniques and technologies for nutrient recovery/removal from soils and water.

H. Set up a revolving fund for investments in closed loop systems like aquaponics or hydroponics

I. Facilitate partnerships between precision farming companies and farmer’s organizations or large food processors with control over their supply chain, to improve efficiency of synthetic fertilizer application.

Exports of European food to global markets:

A. Lobby for the reduction of subsidies which push the price of food very low

B. Work in a coalition of NGOs to ensure that food aid is not used as a means of dumping

C. Work in a coalition of NGOs to ensure that long-term solutions are selected for addressing hunger over short-term solutions such as food aid

D. Provide food processing companies with information and tools to produce and export foods that fit the traditional diets of the importing regions

E. Participate in international negotiations with the goal of encouraging trade agreements that focus on food self-sufficiency, over trade liberalization. 

5.3 Key focus areas

Based on our assumptions, after looking at a selection of interventions for the 22 high-leverage processes, we identified a few key areas of activity that we estimate would be most impactful for organizations to engage in. This is based on our estimations of feasibility and effectiveness of interventions, as well as activities which have an effect on many different areas. These include the following activities:

CAP and CFP reform:

Changes to incentive structures such as the CAP and the CFP ranked generally low for feasibility, though very high for effectiveness. The low feasibility is partly due to the strong lobby around agriculture and fisheries. Despite this, a handful of potential interventions rose to the top, including increased supports for rural amenities (such as education), shifting the focus of subsidies away from larger farms towards small-scale or traditional farms, and supporting developments in urban agriculture through CAP reform. Reductions in overall food subsidies or selective reductions in subsidies for specific types of foods (animal products, feed crops, etc) did not appear be a viable manner to reduce overall food demand or address global trade distortions, due to potential tradeoffs. Reducing or eliminating fishing subsidies comes with fewer tradeoffs and scores higher for both feasibility and effectiveness than changing agricultural subsidies.

Funding small projects:

Though a large number of potential research and development directions were listed as interventions, they generally scored low for potential impact, mainly because of uncertainty around successful realization of new innovations. On the other hand, the establishment of revolving funds for a number of issues scored rather highly. 

Additionally, the establishment of more sustainable agricultural systems, such as polycultures, silvopastoralism, urban agriculture, or perennial crops, requires larger up-front investments and have a relatively long payback period, which forms a barrier to their establishment. Establishing revolving funds for these types of activities presents an option for bypassing the difficulties and uncertainty involved in trying to change legislation to alter incentive structures, allowing the organization to create change in an ideally cost-neutral way.

Working directly with companies:

Interventions that aim to change consumer demand through awareness campaigns all ranked relatively low for feasibility. However, there are a number of ways to interact with companies directly that lead to more desirable outcomes. Connecting companies and experts and endorsing companies who improve their practices both ranked highly as interventions. Another example would be bringing together experts and companies around nutrient recovery from waste or sustainable aquaculture. Endorsements of companies who go above and beyond by reducing waste and consumption of feed crops, or increasing labour standards or standards of living for animals, can provide an incentive for companies to improve behaviours and simplify more sustainable choices by consumers.

Forming partnerships with NGOs:

Privatization of land, hunger, and the loss of culture and heritage are some examples of problem areas. As we have seen, these problems are also closely related to biodiversity loss. For these types of problems, collaborations with NGOs generally scored high for feasibility and effectiveness. In order to ensure that global hunger, poverty, and losses of traditional cultivation and dietary patterns are addressed sustainably, without further losses to biodiversity, collaboration with other NGOs who specialize in these issues may be the best path.

6. Roadmap development using systems thinking

6.1 Developing a systems based roadmap

With the methods presented in this report, organizations and thought leaders can begin to formulate strategies and roadmaps using a process of applied systems thinking. Where roadmaps outline which specific interventions should be implemented over time to reach a certain end-goal, transition strategies outline the different ways in which an organization can go about realizing these interventions (the how). In this section, we outline a step-by-step approach for developing roadmaps for change, in accordance with the principles of systems thinking. 

The practice of road mapping is widespread in the public and private sector alike. When developing roadmaps for creating change within complex systems, it is highly beneficial to take into account the complexity of navigating many interdependent factors over time. Roadmaps produced by planning systemically have a number of benefits over conventional roadmaps, including the following adapted from Stroh (2015):  

• They incorporate dynamics of reinforcing and balancing feedback, thereby representing how complex socio-ecological systems actually behave and how interventions will affect these dynamics in the short- and long-run. 
• They optimize the relationships in the system instead of the individual parts themselves.
• They include an integrated definition of success, so that a broad range of sustainability dimensions can be addressed rather than one or few isolated themes.
• They take time delays into account when defining sequence and timing.
• They strive for sustainable improvements both in the short-term and in the long-term.

6.2 A step-by-step guide to roadmap development

We suggest a five-step approach to developing systemic roadmaps, which are:

1. Creating a shared vision
2. Defining end-goals and KPIs
3. Defining a package of interventions
4. Planning the order and timing of interventions
5. Continuous re-evaluation

Step 1: Creating a shared vision

The first steps in any attempt to drive change is the creation of a shared understanding of the current state of the system and a shared vision in which the change that an organization wants to see is clearly defined. As in any organization, there are likely to be differences in the way in which people perceive problems and solutions. There may be differences in opinion on which impacts should be prioritized (fertilizer use, land use change, or the impact of extractive fisheries?), or what the best means is for achieving a change (e.g. should there be a focus on production practices or consumer behaviour?), while addressing these differences is key to steering an organization. 

Systems thinking can help overcome potential differences within the organization by:

• Making all those involved aware of the mental paradigms that drive the behaviour of people, departments, and offices within the organization. 
• Taking the discussion beyond actions and events to find common ground and a shared vision for change, even where there are more superficial differences at first sight.
• Letting all stakeholders involved see what their role is within the bigger picture and how they can contribute to realizing the shared vision.

Step 2: Defining specific end-goals and KPIs

Once a vision is clearly outlined, specific goals need to be formulated, ideally accompanied by key-performance indicators that can measure progress towards these goals. To achieve meaningful results, goals should be aimed at creating structural, lasting change. Thus, they should ideally not address symptoms at the events level, but rather structures (physical, or institutional) or mental models which are the root causes behind the impacts and the functioning of the system. 

During the goal-setting process, it is also important to keep in mind that there may be dominant mental models present within the current system, which seem self-evident, but in fact create solutions for the wrong part of the system. One example within the European food system is the continuous hunt for more efficient production methods. One could argue that future trends such as a projected doubling of food demand (United Nations, 2015), make the need for efficient production more pressing than ever before. However, there is no reason to assume that the optimal means satisfying this need is through more intensive production per hectare as was the goal in the Green Revolution.

When measuring efficiency along other measures of success (e.g. ecological footprint, resilience), we can see the European food system has evolved optimally to pursue a goal which does not fit with other goals such as preserving biospheric integrity or maintaining productivity over the long term. The fact that the current system has evolved to maximize short-term yields does not mean that this goal is the most sustainable one, or that this measure of the system’s functioning is the most relevant one for the future. Systems thinking can help in the goal setting process if the following actions are included in this phase:

• Laying bare what goals address the level of events rather than structural change; addictions to short-term fixes with negative consequences in the long run can then be avoided. 
• Determining which sunk costs, or financial, technological, and political lock-ins are present in the current system. When aware of these factors, those setting new goals can consciously consider the desired direction for change. 
• Providing people with the awareness that the system is too complex to predict the exact means by which future goals are best achieved; facilitating performance-based goal-setting which allows for innovation and adaptation.

Step 3: Defining a package of interventions

The next step is to select a package of interventions which can realize these end-goals if successfully implemented. In this report, we have outlined a method for prioritizing impacts and addressing them by changing processes and behaviours through feasible and effective interventions. Obviously both effectiveness and feasibility are key characteristics to take into account when assembling a package of interventions, but the dynamics of complex systems mean that in addition to this, there are other issues which need to be taken into account when looking at the interrelationships between interventions, such as: trade-offs, dependencies, and reinforcing or weakening effects of interventions upon each other. Systems thinking can ensure that the measures selected reinforce each other, facilitate each other’s implementation, and create cascades of changes throughout the system. To achieve this, we recommend the following:

• Select measures with a symbiotic advantage; packages of interventions that reinforce each other’s effect so that the whole package combined has a larger effect than the ‘sum of its parts’.
• Select ‘enabling interventions’ at an early stage, so that the possible range of action in the future grows over time.
• Keep an eye out for pathways forward and the effect that interventions have upon the possibilities to change course and adapt in the future: avoid lock-ins and dead-ends at all costs.

Step 4: Planning the order and timing of interventions

Once the optimal combination of interventions has been selected, one needs to think carefully about the sequence in which they should be implemented, and plot their implementation on a time-scale to see at what point in the future they’ll be realized. Again, it is very important to think about the implications of implementing individual measures for the dynamics of the whole system and to avoid undesirable trade-offs, path-dependencies, and lock-ins. Furthermore, delays within the system need to be taken into account. In essence there are four types of delays (Stroh, 2015):

• The delay between the change in a condition and the awareness that the condition has changed
• The delay between our awareness and our decision to act
• The delay between the decision to act and implementation
• The delay between implementation and corresponding changes in conditions 

Long term planning, and creating the time needed for the effect of interventions to grow is essential.  At the same time, it is crucial to achieve some success quickly to maintain momentum and energy within the organization and in society for the implementation of other interventions. To achieve this, we recommend the following considerations:

• Defining quick wins to create momentum forward, so long as addiction to short terms fixes or fixes aimed at non-structural changes are avoided.
• Defining a number of mid-term goals which are enabled by these first quick-wins and build upon and reinforce these first steps. Likewise, mid-term goals should also pave the way for the interventions which will follow. Where the implementation of mid-term interventions is a prerequisite for more long-term aspirations, these are prioritized above others. Because of a delays within the system, it is very important to take into account the build-up period required for these measures to have an impact.
• Defining one or two more ambitious interventions at the end of the roadmap’s timespan: make sure that interventions proposed on the short and medium term strengthen the feasibility and effectiveness of these measures rather than weakening it or creating lock-ins which hamper their implementation.

Step 5: Continuous re-evaluation

Continuous re-evaluation of the roadmap is essential. It ensures that people can make necessary course corrections in systems that are ultimately too complex and dynamic to fully predict (Stroh, 2015). Continuous monitoring of the effectiveness of interventions that have been implemented, and continuous learning and observation regarding the way in which the system behaves ensures that measures are adapted or dropped when necessary, and that new options for intervening in the system continue to be explored. Re-evaluation of end-goals, or even the vision and values underlying these goals may also be necessary, albeit on a less frequent basis.   

7. Conclusions

7.1 Conclusions: How to further apply systems thinking to strategic development

Often organizations have a wide range of stakeholders, internal interests, large networks, and conflicting priorities, which may at times form a barrier for developing a central vision that everyone within the organization can agree upon and see their role in realizing. The application of systems thinking can serve a role in providing means for bringing together different stakeholders, establishing common ground between them, and enabling them to communicate and think systemically about their role in the realization of larger organizational goals.

While we, as an outside organization, are unable to decide for others which strategic pathways are the best forward, we belive that our methods can be used effectively by individuals and organizations to capitalize on their network’s existing strengths and expertise and to define new, high leverage focus areas. Especially when addressing impacts in complex systems such as the European food system, a systems thinking approach is particularly valuable.

8. References

Anderson, V., & Johnson, L. (1997). Systems Thinking Basic From Concepts to Causal Loops.

BENGTSSON, B., & GREKO, C. (2014). Antibiotic resistance—consequences for animal health, welfare, and food production. Upsala Journal of Medical Sciences, (119), 96–102. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4034566/pdf/UPS-119-96.pdf

Benitez, I. (2013). Spain Leads EU in GM Crops, but No One Knows Where They Are | Inter Press Service. Retrieved September 24, 2015, from http://www.ipsnews.net/2013/03/spain-leads-the-eu-in-gm-crops-but-no-one-knows-where-they-are/

Billen, G., Garnier, J., & Lassaletta, L. (2013). The nitrogen cascade from agricultural soils to the sea: modelling nitrogen transfers at regional watershed and global scales. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 368(1621), 20130123. http://doi.org/10.1098/rstb.2013.0123

Buchholz, T. S., Volk, T. A., & Luzadis, V. A. (2007). A participatory systems approach to modeling social, economic, and ecological components of bioenergy. Energy Policy, 35(12), 6084–6094. http://doi.org/10.1016/j.enpol.2007.08.020

Caritas Europa. (2014). The EU ’ s Role to End Hunger by 2025.

Cogliani, C., Goossens, H., & Greko, C. (2011). Restricting Antimicrobial Use in Food Animals. Microbe, 6(6), 274–279.

Continuous Improvement Associates. (2003). Systems Thinking Archetypes ( Generic Structures ), 1–10.

Dirnböck, T., Grandin, U., Bernhardt-Römermann, M., Beudert, B., Canullo, R., Forsius, M., … Uzieblo, A. K. (2014). Forest floor vegetation response to nitrogen deposition in Europe. Global Change Biology, 20(2), 429–440. http://doi.org/10.1111/gcb.12440

ETC Group. (2008). Who Owns Nature? Corporate Power and the Final Frontier in the Commodification of Life. Who Owns Nature?, etc Group(November), 1–52. http://doi.org/pow

European Association for the Study of Obesity. (2015). Obesity: an underestimated threat.

European Commission. (2010). Final Report – Preparatory Study on Food Waste. Retrieved September 23, 2015, from http://ec.europa.eu/environment/archives/eussd/pdf/bio_foodwaste_report.pdf

European Commission. (2010). Developments in the income situation of the EU agricultural sector, (December 2010), 79.

European Commission. (2013). The impact of EU consumption on deforestation : Comprehensive analysis of the impact of EU consumption on deforestation. http://doi.org/10.2779/822269

European Union. (2014). International Co-Operation and Development. State Government.

European Union. (2013). How many people work in agriculture in the European Union?, (8).

Eurostat. (2015). Eurostat: Your Key to European Statistics. Retrieved September 17, 2015, from http://ec.europa.eu/eurostat/web/energy/data/shares

Eurostat. (2012). Agri-environmental indicator – consumption of pesticides – Statistics Explained. Retrieved September 24, 2015, from http://ec.europa.eu/eurostat/statistics-explained/index.php/Agri-environmental_indicator_-_consumption_of_pesticides

Eurostat. (2012). Gross Nutrient Balance. Retrieved from http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=aei_pr_gnb

Eurostat. (2012). Agri-environmental indicator – nitrate pollution of water – Statistics Explained. Retrieved September 29, 2015, from http://ec.europa.eu/eurostat/statistics-explained/index.php/Agri-environmental_indicator_-_nitrate_pollution_of_water

Eurostat. (2012). Population trends of farmland birds – Statistics Explained. Retrieved October 2, 2015, from http://ec.europa.eu/eurostat/statistics-explained/index.php/Agri-environmental_indicator_-_population_trends_of_farmland_birds

Food and Agriculture Organization of the United Nations. (2010). Agri-environmental indicators: Water withdrawal for agricultural use as a % of total water withdrawal (average from 2001-2010). Retrieved from http://faostat3.fao.org/home/E

Food and Agriculture Organization of the United Nations (FAO). (2015). FAOSTAT. Retrieved August 12, 2015, from http://faostat3.fao.org/home/E

Food and Agriculture Organization of the United Nations (FAO). (2015). FishStat Plus. Retrieved September 7, 2015, from http://www.fao.org/fishery/statistics/software/fishstat/en

Food and Agriculture Organization of the United Nations (FAO). (2015). AQUASTAT – FAO’s Information System on Water and Agriculture. Retrieved September 30, 2015, from http://www.fao.org/nr/water/aquastat/di

Freund, D. (2015). EU Integrity Watch Brussels lobbying in numbers. Brussels. Retrieved from http://www.transparencyinternational.eu/wp-content/uploads/2015/06/23-06-2015-EU-Integrity-Watch-Launch-Press-Pack.pdf

Friends of the Earth Europe. (2012). Publid Money for Public Goods? Common Agricultural Policy 2014-2020, (October 2011).

Gelder, J. W. van, Kammeraat, K., & Kroes, H. (2008). Soy consumption for feed and fuel in the European Union, 22.

Han, J., Shi, J., Zeng, L., Xu, J., & Wu, L. (2015). Effects of nitrogen fertilization on the acidity and salinity of greenhouse soils. Environmental Science and Pollution Research, 22(4), 2976–2986. http://doi.org/10.1007/s11356-014-3542-z

Heidhues, F., & Obare, G. (2011). Lessons from structural adjustment programmes and their effects in Africa. Quarterly Journal of International Agriculture, 50(1), 55–64.

Heinrichs, H., Neuman, K., & Lorenz, U. (2013). Transformative modeling: Identifying drivers, blockages and fundamental leverage points for sustainability transition.

Horrigan, L., Lawrence, R. S., & Walker, P. (2002). How sustainable agriculture can address the environmental and human health harms of industrial agriculture. Environmental Health Perspectives, 110(5), 445–56. Retrieved from http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1240832&tool=pmcentrez&rendertype=abstract

International Union for Conservation of Nature (IUCN). (2015). The IUCN Red List of Threatened Species. Retrieved October 20, 2015, from http://www.iucnredlist.org/

Jeffries, B. (2015). WWF Annual Review 2014. http://doi.org/10.1007/s13398-014-0173-7.2

Jones, R. J. a, Hiederer, R., Rusco, E., & Montanarella, L. (2005). Estimating organic carbon in the soils of Europe for policy support. European Journal of Soil Science, 56(5), 655–671. http://doi.org/10.1111/j.1365-2389.2005.00728.x

Landers, T. F., Cohen, B., Wittum, T. E., & Larson, E. (2012). A review of antibiotic use in food animals: perspective, policy, and potential. Public Health Reports, 127(1).

Lowder, S., & Raney, T. (2005). Food Aid : A Primer. FAO – Food and Agriculture Organization of the United Nations. Retrieved from ftp://ftp.fao.org/docrep/fao/008/ae878e/ae878e00.pdf

Maani, K., & Cavana, R. Y. (2007). Systems thinking, system dynamics: Managing change and complexity. Prentice Hall.

Meadows, D. (1999). Leverage Points: Places to Intervene in a System. The Sustainability Institute, 2–19.

Meadows, D. H., & Wright, D. (2008). Thinking in Systems: A Primer. Chelsea Green Publishing. Retrieved from https://books.google.com/books?hl=en&lr=&id=CpbLAgAAQBAJ&pgis=1

Monat, J. P., & Gannon, T. F. (2015). What is Systems Thinking? A Review of Selected Literature Plus Recommendations. American Journal of Systems Science, 4(1), 11–26. http://doi.org/10.5923/j.ajss.20150401.02

Mousseau, F., & Mittal, A. (2006). Food Sovereignty: Ending World Hunger in Our Time. Humanist, 66(2), 24–26. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&AuthType=ip,cookie,url,uid&db=aph&AN=19862072&lang=es&site=ehost-live

Oxfam. (2015). The Food Index.

Pelum Association y Practial Action. (2005). The crisis in African A more effective role for EC aid ?

Priefer, C., Jörissen, J., & Bräutigam, K. R. (2016). Food waste prevention in Europe – A cause-driven approach to identify the most relevant leverage points for action. Resources, Conservation and Recycling, 109, 155–165. http://doi.org/10.1016/j.resconrec.2016.03.004

Regmi, A., & Gehlhar, M. (2005). New Directions in Global Food Markets.

Romero, E., Garnier, J., Lassaletta, L., Billen, G., Le Gendre, R., Riou, P., & Cugier, P. (2013). Large-scale patterns of river inputs in southwestern Europe: Seasonal and interannual variations and potential eutrophication effects at the coastal zone. Biogeochemistry, 113(1-3), 481–505. http://doi.org/10.1007/s10533-012-9778-0

Searchinger, T., Hanson, C., Ranganathan, J., Lipinski, B., Waite, R., Winterbottom, R., … Heimlich, R. (2013). The great balancing act, (May), 1–16.

Shah, A. (2007). Food Aid. Global Issues. Retrieved from http://www.globalissues.org/article/748/food-aid

Simonette, M. J., Rodrigues, A. M. D. A., Seno, W. P., Plínio Franco Thomaz, Navarro, F. J. K. G., Martinelli, D. P., … Hardman, J. (2008). Efetividade Do Processo De Comunicação Com Base Na Teoria Do Comportamento Informacional: O Caso De Um Organismo Internacional Da Área Da Saúde Pública Sediado No Brasil. Systems Research and Behavioural Science, 8(3), 27–42. http://doi.org/10.1002/sres

Spectrum Analytic. (2014). Agronomic Library: Cation Exchange Capacity (CEC), (3), 1–5.

Steen-Olsen, K., Weinzettel, J., Cranston, G., Ercin, a. E., & Hertwich, E. G. (2012). Carbon, land, and water footprint accounts for the european union: Consumption, production, and displacements through international trade. Environmental Science and Technology, 46(20), 10883–10891. http://doi.org/10.1021/es301949t

Stroh, D. P. (2015). Systems thinking for social change: A practical guide to solving complex problems, avoiding unintended consequences, and achieving lasting results. Chelsea Green Publishing.

Tendall, D. M., Joerin, J., Kopainsky, B., Edwards, P., Shreck, A., Le, Q. B., … Six, J. (2015). Food system resilience: Defining the concept. Global Food Security, 6(October), 17–23. http://doi.org/10.1016/j.gfs.2015.08.001

Van den Belt, M., Forgie, V., Scott, A., Frampton, A., & Obeidat, A. (2013). Solution-oriented Systems Thinking Archetypes; examples from the Manawatu River, New Zealand. 31 St International Conference of the System Dynamics Society, 1–20.

Velthof, G. L., Lesschen, J. P., & Webb, J. (2014). Measuring the impacts of the Nitrates Directive on nitrogen emissions.

Verheijen, F. G. a., Jones, R. J. a., Rickson, R. J., & Smith, C. J. (2009). Tolerable versus actual soil erosion rates in Europe. Earth-Science Reviews, 94(1-4), 23–38. http://doi.org/10.1016/j.earscirev.2009.02.003

Vos, W. & M. H. (1999). Trends in European cultural landscape development: Perspectives for a sustainable future,. Landscape and Urban Planning, 46, 46, 3–14.

Willem, D., Page, S., Cantore, N., King, M., Boysen, O., & Keijzer, N. (2014). the EU’s CAP and development, (79).

World Health Organization (WHO). (2011). Tackling antibiotic resistance from a food safety perspective in Europe. World Health, 1–88.

World Resources Institute. (2015). Aqueduct Global Maps 2.1 Data | World Resources Institute. Retrieved September 4, 2015, from http://www.wri.org/resources/data-sets/aqueduct-global-maps-21-data

World Resources Institute. (2011). World Greenhouse Gas Emissions in 2005 | World Resources Institute. Retrieved September 23, 2015, from http://www.wri.org/publication/world-greenhouse-gas-emissions-2005

WWF. (2007). Performance and Learning Culture, (July).

WWF. (2016). How is WWF run? Retrieved from http://wwf.panda.org/who_we_are/organization/

Colophon

Authors

Erin Kennedy, Eva Gladek, Gerard Roemers

Research Team

Matthew Fraser, Fatema Baheranwala, Oscar Sabag, Evan Wood, Theologos Xenakis