Quantifying the environmental impacts of food waste

(1)

Quantifying the environmental

impacts of food waste

A study to explore the impact of food waste on the carbon, land, and

water footprints of Germany, the Netherlands, and the UK as well as on

the displacements of environmental pressures as a result of the Food

Supply Chain's fragmentation

Master Thesis by Katja Baldus (S2361523) June 2015 (Spring Semester 2015)

University of Groningen

Faculty of Economics and Business

Master Programme International Economics & Business Groningen, the Netherlands

(2)

2

1. I

NTRODUCTION

"The current global population of seven billion is expected to grow to nine billion by 2050. But the number of hungry people need not increase. By reducing food waste, we can save money and resources, minimize environmental impacts and, most importantly, move towards a world where everyone has enough to eat.

Ban Ki-moon, UN Secretary-General1

Feeding an expanding population is a major challenge of today's world according to the Food and Agriculture Organization of the United Nations (FAO). By 2050 the world's population is expected to have increased by about two billion. FAO estimates that in order to nourish these additional two billion people the production of food would need to increase by 60% (FAO, 2014a). Moreover, advisors at the United States (US) State Department emphasize the potential of feeding the world to foster peace and stability (National Geographic Magazine, 2015). While this may be true, our need to eat is a major burden on the planet. Today food production is highly unsustainable as it contributes to diminishing natural resources, vanishing biodiversity, deteriorating ecosystems, and climate change (FAO, 2014a). Therefore, one of the major challenges of the 21st century is to feed an increasing population without harming our planet. This challenge has heightened the need for sustainable food production, and hence a sustainable food supply chain (FSC)

Food wastage occurs at every single stage of the FSC (see figure 1.1): from edible crops not harvested at the farm to consumers buying more than actually needed. It is important to point out that the negative impact of food wastage on sustainability increases as one moves down the FSC since more resources are being used and more value is added to the product.

Figure 1.1: The Food Supply Chain

In this paper, I take the stance that reducing food waste, particularly at the consumption level, can play a major role in making the FSC more sustainable.

(5)

5 Food wasted at the final stage of the FSC embodies the most resources and value lost. Since food at this stage is mainly wasted due to poor planning or personal preferences (Parfitt, Barthel, & Macnaughton, 2010) there are many opportunities for food waste reduction by means of awareness raising measures. In contrast to much of the research on this topic, I argue that reducing food waste can only indirectly contribute to food security by making food production more sustainable. Therefore this paper attempts to discover the potential of reducing food waste in making the FSC more sustainable and ultimately contributing to food security.

With regard to the discipline of sustainable food production and sustainable FSCs it can be stated that the reduction of food loss and food waste2_{(henceforth referred to as food} wastage) is central. It can be claimed that food wastage is a fully unsustainable practice as it negatively impacts the three sustainability dimensions that will be elaborated hereafter: economic, environmental, and social.

Firstly, the negative impacts related to the economic dimension are easily observable, given that direct costs arise as food is wasted. Economic costs of food wastage refer to the economic value of the food that is being wasted. In the year 2007, this amounted to 750 billion US Dollar (USD) (FAO, 2013). Besides the costs arising from the lost economic value of the food wasted, there are also costs related to its disposal. For example, in the US the disposal of 32 million tonnes of food wastage costs local governments about USD 1.5 billion per year (The New York Times, 2015). However, it must be stated that this number merely represents the tip of the iceberg.

Secondly, the critical impacts of food wastage are hidden and are related to both the social and the environmental dimension of sustainability. These impacts are often not directly observable as they do not necessarily occur where the product is ultimately consumed. This makes it easy to ignore their existence although, the consequences will be borne by society at large and particularly by future generations. The social impacts of food waste relate to the moral and ethical aspects of wasting food (Papargyropoulou,

(6)

6 Lozano, & Steinberger, 2014) with food security3_{at the centre of attention.} Papargyropoulou, Lozano, & Steinberger (2014) name the reduction of food wastage as a "first step towards achieving food security". In this context, it seems like a morbid joke, or even immoral (Parfitt, Barthel, & Macnaughton, 2010), that one third of food produced destined for human consumption is being lost or wasted while in 2014 805 million people worldwide were still suffering from chronic undernourishment. These figures show that there is a wide gap between developing and developed countries. And lastly, the environmental impacts of food wastage are primarily related to the waste of resources. Gustavsson, Cederberg, Sonesson, van Otterdijk, and Meybeck (2011) describe food wastage as representing "a waste of resources used in production such as land, water, energy and inputs". In this regard it can be stated that owing to unsustainable agriculture, a quarter of the world's land is degraded, nearly a third of the worldwide fish stocks overfished, and biodiversity lost, while no more than four of 30,000 plants dominate our daily diet (FAO, 2014a). In addition, agriculture is accountable for 70% of total water use, from which it follows that it can also be held responsible for progressing water scarcity and deteriorating water quality (FAO, 2014b). Consequently, agriculture heavily impacts on natural resources, ecosystems and biodiversity, and where food wastage is concerned it does so in vain.

With regard to the academic debate on food waste and more precisely the three sustainability dimensions outlined afore, it can be stated that literature has grown in recent years. To mention just a few, several studies have documented the scale of food wastage at a global level. One of the first studies to quantify global food loss and waste is the research of Gustavsson et al. (2011), who estimate that about one third of the global food production, i.e. roughly 1.3 billion tonnes per year, is lost or wasted on the way to final consumption. In an attempt to take a holistic look at the impact of food wastage, the FAO (FAO, 2014b) was the first to include the environmental and social costs of food wastage in its research. Results show that, using a full-cost accounting framework, the costs of the externalities associated with the environmental and social impacts caused by food wastage, can be measured and subsequently valued in monetary terms.

(7)

7 Moreover, it must be pointed out that by looking exclusively at the economic costs of food wastage , one underestimates the problem. For example, to the USD 1 trillion of economic costs per year, one has to add about USD 700 billion of environmental costs and USD 900 billion of social costs. A study looking at the effects of reducing food wastage on food security is the research by Munesue, Masui, and Fushima (2015). They examine the effects of food wastage reduction in developed countries, throughout the entire FSC, and its impact on food security and global resources. According to this research, a 50% reduction of food wastage in developed countries can cause the number of undernourished people in developing countries to decrease by 63.3 million.

When looking at the national level, it can be observed that there is a growing number of studies commissioned by national governments or supranational institutions such as the European Union. These studies predominantly examine the economic costs of food wastage. In a preparatory study of the European Commission (Monier, et al., 2010) with the aim of exploring the scale of European food wastage throughout the FSC (excluding agriculture), annual European (EU27) food wastage is estimated to amount to 89 megatonnes (Mt), i.e. 179 kg per capita4_{. European households can be accounted for} lion's share of the total food wastage disposing 42% or 38 Mt of the total. This is followed by the manufacturing sector, with 35 Mt of food loss per year. The study also provides an outlook for the level of European food wastage, which is expected to increase by 37 Mt to 126 Mt by 2020 (Monier, et al., 2010). The most comprehensive research on the national level can be found in the United Kingdom (UK). Based on the results of a detailed waste analysis across the UK, in addition to the outcomes of food and drink diaries from 950 households and a synthesis of local waste data, British food waste is determined at 7 Mt. This number is equivalent to around 17 million tonnes of generated CO2 (tCO2e) (Quested, Ingle, & Parry, 2013).

From the detailed overview of academic research outlined afore, it becomes apparent that the indirect effects of food wastage have been neglected.

Indirect effects of food wastage arise as a result of globalisation and the concomitant fragmentation of the FSC. Both developments have important implications on food wastage. Since products are no longer "made in Germany" or "made in the UK", but

(8)

8 "made in the world", the effects on reducing food wastage go beyond direct effects on resources. This can be illustrated by taking a look at a cup of Starbucks coffee: your regular coffee, consumed in a Starbucks in the Netherlands, may contain coffee from Colombia, Peru, or Indonesia; sugar from Brazil, and the paper for the cup is sourced from Sweden (International Networks Achive, 2015). If instead of drinking this cup of coffee you decide to discard it, because it got cold the list of wasted resources goes beyond those named above. For the Swedish paper cup paper might be imported from Poland, which in turn imported cellulose from Ukraine, which in turn imported wood from Russia, which in turn imported fertiliser for the trees from the United States. This example illustrates how present research on food wastage is likely to underestimate the potential of reducing food wastage in making the FSC more sustainable.

Taking the aforementioned complexity into consideration, and filling the gap in academic research this research seeks to firstly quantify the impact of food waste on carbon, land, and water footprints at the consumption stage in Germany, the Netherlands and the UK. Secondly, to explore the displacement of environmental pressures across countries, and possible changes caused by reduced food waste at the consumption stage.

With regard to the scope of this research, it must be stated that due to the complexity of quantifying the ethical aspects of food wastage, the focus will limited to the environmental impacts. Moreover, this research will utilize the carbon, land, and water footprints to visualise the impacts of food wastage on the environment as footprints make comparison possible.

The research method employed in this paper is an Input-Output (IO) Analysis as it provides a means to analyse the indirect effects of reduced food waste on the footprints. In addition, it is particularly useful in studying the displacement of environmental pressures. This is of interest as it is expected to shed light on how food waste in one country generates different environmental pressures across the globe.

(9)

9 changes in the displacement of environmental pressures as a result of reducing food waste. Due to data constraints research will be limited to the analysis of food waste occurring in households in the years 2007 and 2010. Data is drawn from three main sources: Data for Germany is taken from a study commissioned by the Federal Ministry of Food and Agriculture of Germany in 2012 (Kranert, et al., 2012).

Data for the UK, it taken from a study from the non-profit organization WRAP on estimates of food and drink waste in British households (Quested, Ingle, & Parry, 2013). Finally the data for the Netherlands is based on a study executed by CREM5_, commissioned by the Dutch Ministry of Infrastructure and the Environment, and under the lead of Agentschap NL6_{(van Westerhoven & Steenhuisen, 2010).}

This paper is organised as follows: Chapter two begins by laying out the theoretical dimensions of the research. This will be followed by a description of the data used and a specification of the final sample in chapter three. The fourth chapter deals with the methodology used. In chapter five empirical results are presented and lastly discussion of the results and their implications will be provided in chapter six.

(10)

10

2. T

HEORY

In this chapter I will lay out the theoretical dimensions of this research. I will start by defining food loss and food waste, followed by a more detailed description of food wastage and what it entails. I will finish this chapter with a description of the footprint family.

2.1 D

EFINING FOOD LOSS AND FOOD WASTE

While a variety of definitions of food wastage have been suggested (FAO, 1981; Stuart, 2009; Smil, 2004; Parfitt, Barthel, & Macnaughton, 2010) this paper will employ the definition first proposed by Gustavsson et al. (2011). The definition, makes a distinction between food loss and food waste. The two concepts are defined as follows: "the masses

of food lost or wasted in the part of food chains leading to edible products going to human consumption." This definition was chosen for this research as it allows for a holistic

approach to food wastage, taking into account food lost or wasted that is subsequently allotted to non-food uses such a animal feed or bioenergy. Moreover, this definition helps distinguish between “planned” non-food uses and “unplanned” non-food uses, which result from the need of recycling food losses. It is necessary to clarify what is meant by food loss and food waste. Firstly, food loss refers to the part of food wastage occurring at the first stages of the FSC, i.e. at the agriculture stage (including postharvest), the food processing and manufacturing stage (Gustavsson et al. 2011). Food waste on the other hand refers to food wastage occurring at the final stages of the FSC, i.e. at the retail and consumption stage ( The Government Office for Science, 2011; Parfitt, Barthel, & Macnaughton, 2010).

Figure 2.1: Composition of Food Waste

For the purpose of this research, food waste is further classified according to its prevention potential (see figure 2.1), in compliance with the classification set out in the WRAP report (Quested, Ingle, & Parry, 2013). In this paper, the term avoidable food waste will be used to refer to all "food and drink thrown away because it is no longer

(11)

11 wanted or has been allowed to go past its best" (Quested, Ingle, & Parry, 2013). Possibly avoidable food waste can be defined as "food and drink that some people eat and others do not" (Quested, Ingle, & Parry, 2013), additionally it encompasses food items such as bread crust, apple skin, or potato peel. Analogous to avoidable food waste, possibly avoidable food waste is or has been edible at some point before being thrown away. Finally the term unavoidable food waste refers to "waste arising from food and drink preparation that is not, and has not been, edible under normal circumstances" (Quested, Ingle, & Parry, 2013) i.e. e.g. egg shells, meat bones, residues from brewing tea or coffee.

2.2 T

YPES OF FOOD WASTAGE

Food is lost or wasted all along the FSC (see figure 2.2): from edible crops not harvested at the farm to consumers buying more than actually needed The problem is present in both developed and developing countries, though the causes differ greatly (see appendix 2). In developing countries food losses prevail. Food is lost as a result of inadequate or poorly developed infrastructure, processing facilities, and harvesting techniques; as well as difficulties to adhere to food safety standards. In developed countries food is mostly wasted at the final stages of the FSC. Here, the main drivers of food waste are consumer preferences on the one hand and lack of coordination between the individual chain members, resulting in production exceeding demand on the other hand.

Figure 2.2: Activities throughout the FSC giving rise to food wastage, adapted from (Parfitt, Barthel, & Macnaughton,

2010) Agriculture

•Edible crops left in field, loss in food quality due to suboptimal timing of harvest •Crop damaged during harvesting

•Out-grades at farm to comply to cosmetic standards

•Loss owing to spoiling/bruising, pests, disease, spillage, contamination, natural drying out of food

Processing & manufacturing

•Primary processing (cleaning, drying, sieving,...) and secondary processing (mixing, cooking, cutting, ...): •Process losses and contaminations causing loss of quality

•Product discarded/out-grades •Inappropriate packaging

Retail

•Damage during transport

•Losses caused by lack of cooling/cold storage •Inappropriate packaging

•Poor stock management

Consumption

•Leftovers

•Poor storage/stock management in homes •Poor food preparation techniques

(12)

12 At the individual stages of the FSC different activities give rise to food being lost or wasted (figure 2.2). At the agricultural level crops are not harvested due to suboptimal timing of the harvest or they are eaten by animals. Food is also lost when it is damaged during harvesting or subsequently out-graded in order to upgrade the quality of the harvest and to comply to "cosmetic standards". At the post-harvest level food is lost owing to damages and spoilage occurring during transportation. Moreover, food is lost owing to diseases or pests infecting the food during storing. At the processing and manufacturing level food is lost as a result of primary activities, such as drying, milling and grinding; or secondary activities, such as cooking, cutting, and mixing. In both phases food is lost owing to process inefficiencies or food contamination. Furthermore, at this level food is out-graded for cosmetic or quality reasons. At the retail level food waste can take place during transport, owing to inadequate storage or cooling facilities or inappropriate packaging, or at the retail shops due to poor stock management. At the consumption level, food waste above all occurs as a result of poor storage and stock management, implying that more is bought than is ultimately consumed.

2.3 T

HE FOOTPRINT FAMILY

Even though the footprint concept has been widely used in research, (Galli, et al., 2012) can be said to use the term footprint family first. The authors define the footprint family as "a set of indicators – characterized by a consumption – based perspective – able to track

human pressure on the surrounding environment". They claim that only the conjoint use

and interpretation of environmental indicators allows a comprehensive analysis of the human pressures on the environment. Moreover, they define human pressure on the environment as the "appropriation of biological natural resources and CO2 uptake,

emission of GHGs7_{, and consumption and pollution of global freshwater resources". Three} footprints allow to monitor these pressures by overseeing three key ecosystem areas which are the biosphere, atmosphere, and hydrosphere (see appendix 3 for an overview on the footprint family).

The first footprint is the Ecological Footprint, which measures the appropriation of productive land necessary to enable the production of goods in order to enable a certain

(13)

13 lifestyle. The Ecological Footprint is expressed in global hectares (gha), which refer to hectares of average productivity. The second footprint, the Carbon Footprint, measures the GHG emissions embodied in the consumption of a nation's economic actors, which comprise of households, governments, organizations, companies, and the industry sector. The Carbon Footprint is expressed in mass units, meaning in kg or tonnes of CO2. It can also be expressed in terms of CO2-equivalent (CO2-e) if more GHGs are considered. In order to make them comparable, the mass of GHGs other than CO2 is multiplied by their global warming potential (GWP) to derive at their CO2-equivalent mass. The Carbon Footprint also considers trade. Consequently a nation's carbon footprint is equal to the aggregate of all its emissions which are related to consumption, including imports but deducting exports. Finally, the Water Footprint measures the direct and indirect use of freshwater by both producers and consumers. It encompasses the use of Blue, Green, and Grey Water. Blue Water refers to surface and ground water, Green Water to rainwater consumed by plants. Finally Grey Water refers to "the volume of freshwater required to assimilate the load of pollutants based on existing ambient water quality standards" (Galli, et al., 2012).

(14)

14

3. D

ATA

This chapter is concerned with the data used for this research. The level of detail described here focuses to the extent that is necessary for the understanding of the research (please refer to appendix 4 for more details). A full technical description of the data collection process, i.e. the collection and analysis of waste, of the individual countries is beyond the scope of this paper.

3.1 C

ASE STUDIES

As mentioned in the introduction the research data in this thesis is drawn from three main sources, which will be discussed in more detail in the following paragraphs.

3.1.1DEFINITION OF FOOD WASTE

Data for Germany is taken from a study commissioned by the Federal Ministry of Food and Agriculture of Germany in 2012 (Kranert, et al., Ermittlung der weggeworfenen Lebensmittelmengen und Vorschläge zur Verminderung der Wegwerfrate bei Lebensmitteln in Deutschland, 2012). The data for the Netherlands is based on a study executed by CREM8_{, commissioned by the Dutch Ministry of Infrastructure and the} Environment, and under the lead of Agentschap NL9_{(van Westerhoven & Steenhuisen,} 2010). Finally, the study consulted for the UK is WRAP's10_{2012 update on its 2007} estimates of food and drink waste in British households (Quested, Ingle, & Parry, 2013). The UK is the country with by large the most extensive data available on food waste, which is a major contribution of WRAP.

The food waste data presented in both the Dutch and the British report is based on a detailed measurement of the weight and composition of food waste found in the waste streams collected by or on behalf of the local authorities. This data is subsequently complemented with an approximation of food and drinks disposed of via waste streams other than those collected by the local authorities. For the Netherlands the approximation is generated based on the results of a residents survey, for the UK it is based on food and drink diaries collected from 950 households. As no such waste analysis exist for Germany, food waste is estimated based on national, European and international statistics, literature and studies on food wastage, and data collected by

8_{CREM is the agency for sustainable development in the Netherlands} 9_{Agentschap NL is an agency affiliated to the Ministry of Economic Affairs}

(15)

15 industrial unions. The quantity of German food waste is subsequently projected on the basis of the researched data background.

The British study looks at avoidable, possibly avoidable and unavoidable food waste, which is grouped according to 15 food groups and 150 food type subgroups. Classification of food and drinks is according to their state when they leave the home. If for example eggs are used to prepare a cake, they will be categorised as belonging to the food group bakery. The German study differentiates between nine food groups and considers avoidable, possibly avoidable, and unavoidable food waste. In contrast to the British and the German study, the Dutch study only differentiates between avoidable and unavoidable food and drink waste, with a further division into 22 subcategories categories (see appendix 5).

With respect to the causes of food waste, five principal reasons emerge from the three studies:

 Not used in time: food gone bad (rotten, gone mouldy) or passed its date label  Cooked, prepared or served too much: leftovers from food cooked or

prepared at home

 Personal preference: allergies or other health related reasons, or not feeling like eating this food

 Accidents: e.g. burnt or spoiled food

 All other reasons: cupboard clear out, dregs 3.1.2THE AMOUNT OF FOOD WASTED

(16)

16 with 110 kg per year in 2012, and even 136 kg in 2007 is considerably larger than the Dutch and German per capita food waste with 73kg and 86 kg respectively. The pie charts in appendix 6 show that in all three case countries the larger part of food waste is avoidable.

Weight generated (tonnes) Monetary value (mln $) Per capita

Country Year Total Avoidable Unavoidable Total Avoidable Weight _(kg) Value _($) Germany 2010 7,024,638 3,966,181 3,058,457 44,264 24,500 86 $ 541 The Netherlands 2010 1,214,253 727,256 486,997 5,514 3,271 73 $ 332 The UK 2007 8,313,000 6,802,000 1,511,000 40,825 33,441 136 $ 666 The UK 2012 7,030,000 5,431,000 1,599,000 39,929 30,732 110 $ 627 Average 5,895,473 4,231,609 1,663,864 32,633 22,986 101 $ 541 Table 3.1: Weight and monetary value of food waste per country, split by avoidability

With respect to the composition of food waste, the pie charts in appendix 711_{show that} the food groups generating the bulk of the food waste in all three case countries are the fresh produces such as fruit, and vegetables. They contribute on average 41% of the countries' total food waste. From the bar charts in appendix 8 we can see that more than half of the vegetable and fruit waste could be avoided. The unavoidable part mainly consists of fruit and vegetable skins or stalks. Another item that is frequently disposed of is bread, a waste that could be avoided in the majority of cases.

Two popular reason for disposing of food are first products having gone past their expiry date/not being used in time, and second waste arising due to products being cooked, prepared or served too much (Quested, Ingle, & Parry, 2013; van Westerhoven & Steenhuisen, 2010). Besides not being used in time, freshness appears to be a major argument for disposing of bread, vegetables and fruits. It must be stated though that the most influential reasons for disposal vary according to the individual food groups. 'Not used in time' is a critical reason for products with short shelf-lives such a fresh produces, bakery, dairy and eggs. Products with longer shelf-lives, such as meals and

(17)

17 drinks, were above all disposed of as a result of having prepared, cooked or served too much.

3.1.3LIMITATIONS

I would like to conclude this subchapter by naming a number of important limitations with respect to the data used. Table 3.1 clearly shows that the weight and above all the monetary value of food waste in Germany, the Netherlands, and the UK largely differs. This can be owing to a variety of reasons. Evidently, differences in the quantity of food wasted can be simply due to food waste being more relevant problem in some countries than in others. However differences can be also due to the data collection processes employed to derive at the data. The table in appendix 4 illustrates that not only the methodology differed among the countries, but also their definition of food waste, i.e. the disposal routes, and the product groups. Regarding the monetary value of food waste, differences can be either due to the composition of food waste or likewise due to the differing data collection processes. If above all relatively expensive food products or drinks are disposed of, this will be reflected in a higher monetary value of a country’s food waste, which will ultimately be used in this research to derive the environmental impacts of food waste. Especially when comparing the German with the British data it can be seen that despite having wasted a lower quantity of food, the monetary value of total German waste is higher than that of British waste. Considering that the composition of food waste in these two countries is largely similar this puzzle can almost certainly be explained with different data collection techniques employed by the countries. As long as no standardised definition of food waste, and no standardised data collection processes exist, data taken from these studies must be interpreted with caution. Also it must be highlighted that the studies rely on rather small sample sizes . Therefore robust estimates might not be ensured. Finally, by not considering food disposed off away from home, for example at the workplace or food eaten "on the go", an important waste stream is excluded in the data collection process.

3.2 WIOD

(18)

(19)

19

4. M

ETHODS

In this chapter I will outline the methodology used to achieve this research’s aim of quantifying the impact of food waste on Carbon, Land, and Water footprint of Germany, the Netherlands, and the UK. First I will give a brief introduction into IO Analysis (based on textbook of Miller & Blair, 2009) followed by a description of the model construction process.

4.1 I

NPUT

-

OUTPUT

A

NALYSIS

For this research Input-Output (IO) Analysis was used to explore the effect of food waste on resources worldwide. IO Analysis based methods provide a means to go beyond the direct effects of food waste by also looking at the indirect effects which result from the globalisation of the food supply chain.

For illustration purposes, the methodological explanation in this section will be limited to a description of a single-country IO system. The application on a multi-country system does not differ.

Figure 4.1: The Input-Output table (in monetary terms)

(20)

20 Hotels and Restaurants sector with intermediate inputs worth USD 3,000, this value would appear in the intersecting cell of the row belonging to the agricultural sector and the column belonging to the Hotels and Restaurants sector. This implies that there is a money flow from sector to sector . The -by- matrix is the final demands matrix, with -categories of final demand including private consumption, private investments, government consumption and investments, changes in stocks, and exports. Element gives us the goods deliveries (in monetary terms) from sector to the final demand category . If the agricultural sector directly delivers, i.e. without intermediation, potatoes worth e.g. USD 500 to the private consumer this amount would appear in the intersection of the row belonging to the agricultural sector and the column belonging to private consumption. Again there is a payment from the private consumer to the agricultural sector, i.e. a money flow from final demand to sector . The matrix is a -by- matrix which provides the primary inputs, with -types of primary inputs, including wages and salaries, employers’ contribution, capital depreciation, indirect taxes minus subsidies, operating surplus (other income), and imports. The element shows the money value of primary input deliveries of type to sector . If the Hotels and Restaurants Sector pays USD 5,000 for wages and salaries to its employees this value can be found again in the intersection cell of the row belonging the primary input wages

and salaries and the column belonging to the Hotels and Restaurants sector. The

column vector gives us the gross outputs of sector , that is the monetary value of sector 's total production, including intermediate deliveries as well as deliveries destined for the final demand. The row vector (the transposed column vector ) gives the total payments of sector for all its production inputs, including intermediate deliveries as well as primary inputs such as wages and salaries. As a result from double entry bookkeeping providing the basis for the IO tables, each sector's gross output equals the sum of all its inputs or total payments for its inputs. Therefore the each sector's row sum is equivalent to its column sum. Q4 is the so-called "fourth quadrant" which includes different taxes as well as imports by households destined for the private consumption. This matrix is typically assumed, without a loss of generality, to be empty. With these matrixes it is now possible to calculate the total outputs in all industries necessary to satisfy a certain final demand vector.

(21)

21 to final demand for sector products, given by , where is the sum of all final demand categories. This can be depicted in a simple equation that shows the distribution of sector 's products to the other sectors , and to final demand:

(1)

Thus, total output of sector is equal to the sum of intermediate deliveries to sector , and to final demand. Such an equation can be constructed for each of the -sectors' output sales: (2) Let (3) Then the output sales of each of the -sectors can compactly be written in matrix notation12_as

(4)

is a column vector of 1's, also called a summation vector13_{. By post-multiplying the} matrix with a summation vector we receive a column vector with the row sums of the . Looking at a single column of only, one can see the sources of sector 's intermediate inputs as well as their magnitudes (in USD). It is assumed that the interindustry flows displayed in (for a given time period) depend entirely on sector 's total output ( in that same period). The idea behind this assumption is that for instance the more tomato ketchup is produced in the food sector in a given year, the more tomatoes will be needed from the agricultural sector. This relationship is reflected in the input coefficients which give the ratio of sector 's input to sector 's output:

12_{Lower-case bold letters refer to vectors and upper-case bold letters refer to matrices}

(22)

22

(5) The input coefficient , e.g. can be interpreted the following way: per USD 1 worth of output sector three receives inputs worth USD 0.03 of sector one. The above described relationship is a fixed one, which implies that in IO analysis economies of scale are ignored.

It follows that equation 2-4 can be rewritten, with each being replaced by :

(6) Equation 6 highlights the relationship, or rather dependence, of interindustry flows on each sector's total outputs. Based on this relationship one can find the answer to the question of the extra outputs necessary to be produced in each sector to satisfy a certain final demand vector.

To answer this questions we have already given the forecasts for the final demands, thus the values of are known. Furthermore we know the input coefficients, . What is to be found are the outputs necessary to supply the forecasted final demands. Based on these information we can rewrite equation 6 as:

(7) With another small modification, grouping together all , all ,... all we arrive at the following equations:

(23)

23

(9)

where the denotes a diagonal matrix with the vector's elements aligned on the main diagonal. Thus, taking into account the definition of an inverse 14_{, is equal to}

(10) Post-multiplying by gives us a new matrix in which each element of column of has been multiplied by or, in other words, been divided by . From equation 8 it follows that equation 6 can be written, in matrix form, as:

(11)

From the definition of the identity matrix and of the input coefficients it follows that:

(12)

Now we can write the equation system in equation 7 in the more compact matrix form:

(13)

Since and are known numbers it is now, with another small modification, possible to find an answer to the initial question of necessary output from each of the sectors to supply a forecasted set of final demands:

₍₁₄₎

where

represents the total requirements matrix, better known as the Leontief inverse. The Leontief inverse makes the dependence of outputs on the final demands more explicit. The typical element of the Leontief inverse shows the necessary extra output of sector if final demands for product increase by USD 1. The effect of food waste on the Carbon, Land, and Water Footprint is calculated using multipliers. The concept of multipliers can be used for a variety of variables such as labour inputs, imports and output. For this research's purpose though only the (blue) water consumption, (GHG) emissions and land-use multipliers are of importance. I am

14_{The inverse can be defined as} _{, where is the identity matrix which has ones of the}

(24)

24 going to explain how the multipliers are calculated on the basis of the emissions multiplier. The first step is to calculate the emission coefficients. Emissions per sector are contained in the emission matrix , which displays emissions of sector emitted for the production of intermediate inputs for sector . Total emission per sector are divided by the sector's output, that is or in matrix notation:

(15)

The typical element of the emission coefficients vector shows emissions emitted by industry per USD 1 of output in sector . From equation 14 and 15 it follows that:

(16)

i.e. total emissions are the product of the emissions coefficients and output per sector. If we define now

(17)

then

(18)

Where is the vector with the emissions multiplier. Equation 18 clearly shows that also total emissions emitted per sector are driven by final demands. The typical element of the emission coefficients vector displays total emissions in all sectors together that are emitted to satisfy one extra US dollar of final demands for product . The exact same procedure is applied to land-use and water-use in order to derive the land-use and the water consumption multipliers.

4.2 M

ODEL CONSTRUCTION

(25)

25 data into a WIOD-compatible format, i.e. a format that can be used to execute the IO Analysis.

The model used in this research was constructed in a process consisting of five steps, which will be described in detail in the following paragraphs.

4.2.1DATA COLLECTION

The first step was to gather the data from the previously discussed reports on food waste and complement it where deemed necessary. In the German report (Kranert, et al., 2012) for example the quantity of food waste and prices per kg (per food group) are given in ranges. For that reason the average was taken for further calculations. Food waste data is only available on a per capita basis. It was projected onto the whole German population to estimate the weight and the monetary value of total food waste. The Dutch report (van Westerhoven & Steenhuisen, 2010) provides data on the kg of food wasted per capita per year, divided into an avoidable and an unavoidable part. Just as with the German data, food waste data was projected onto the total Dutch population to estimate total food waste in tonnes, and the monetary value of the food wasted. Due to the unavailability of information on the monetary value of Dutch food waste, prices were taken from Statistics Netherlands (Centraal Bureau voor de Statistiek, 2015) to derive the total monetary value of the food wasted.

4.2.2SYNCHRONISATION

The collected data was synchronised with the final goal in mind to bring it to the same, WIOD-compatible, format. First, the differing food groups of the individual reports were aggregated (see appendix 5) to be covered by ten food groups, which are the following:

1. Vegetables & fruits 2. Drinks

3. Meat & fish 4. Dairy & eggs 5. Bakery 6. Staple foods 7. Meals

(26)

26 This classification is largely based on the British classification. For the German and British data it was moreover necessary to aggregate avoidable and possibly-avoidable food waste. The rationale behind this decision is that possibly avoidable food waste often is closely related to personal preferences. Therefore it can be assumed that this part of the food waste can also be avoided. An example of this first format draft is enclosed in appendix 9. As previously mentioned the WIOD solely provides environmental accounts up to the year 2009. Consequently German, Dutch and British data, which are available for the years 2010 and 2012 respectively, had to be transformed into 2009 data. British data shows that food waste decreased from 2007 to 2012 due to awareness-raising measures. This decrease was taken into account when deriving the 2009 data by calculating the decrease per year which subsequently was added back to the 2012 data to arrive at the food waste of 2009. For Germany and the Netherlands no such comparative data is available. The studies used in this report are in fact the first high level researches on this topic in these countries. Therefore food waste was assumed to neither have increased nor decreased. By reason of the data in the WIOD being given in current prices, the 2010 and 2012 data were corrected for inflation. Where available the individual food group's Consumer Price Index (CPI) was used15 (European Commission, 2015), otherwise the general CPI for food was used. Finally all monetary values, for 2007 as well as 2009, were converted into current USD using official exchange rates (Eurostat, 2015)

4.2.3TRANSFORMATION

In the third step the synchronised data was transformed to be WIOD-compatible. Three basic assumptions were made beforehand. First of all it is assumed that if food waste is reduced this implies a direct reduction of final consumption expenditure by households. There are two likely scenarios that give support to this assumption. In the first scenario food produces that would otherwise be waste will not be bought in the first place. The other scenario is that the products that otherwise would be wasted are consumed. However then they would replace other food products that, as a consequence, will not be bought. Either way there is a reduction of final consumption expenditure taking place. Therefore food waste, in the context of this research, occurs at the final demand level,

(27)

27 more specifically at Final consumption expenditure by households, and it is directly deducted from the respective sector's final demand deliveries. Secondly, only avoidable food waste, which includes possibly avoidable food waste, is considered in the calculations. As the name already reveals, unavoidable food waste is very unlikely to decrease. Therefore including unavoidable food waste in the calculations would produce biased results. Thirdly, as previously explained the tables included in the WIOD contain data on 35 sectors. Of those sectors only two can be considered to be related to food waste, namely the Agriculture, Hunting, Forestry and Fishing sector as well as the Food, Beverages and Tobacco sector. The above described food groups were regrouped as follows:

1. Agriculture, Hunting, Forestry and Fishing: Vegetables & fruits; Meat & fish; Dairy & eggs

2. Food, Beverages and Tobacco: Drinks; Bakery; Staple foods; Meals; Condiments, sauces, oil & fat; Confectionary & snacks; Other

For the model it is further assumed that the share of domestically-produced and imported food that is wasted is according to the domestic production and import shares in the WIOT. Let us take Germany as an illustration. If 0.5% of total final consumption by households of products from the Agriculture, Hunting, Forestry and Fishing sector is imported from Austria, the 0.5% of total agricultural food waste will be deducted from Austrian deliveries to German final consumption by households of products from the Agriculture, Hunting, Forestry and Fishing sector. Thus food waste per sector is divided over the countries according to their share in total Final consumption expenditure by

households. This model comes close to the reality as it distributes food waste over all

countries and hence imports are taken into account. Furthermore this model allows us to look at the countries' environmental pressure trade balances and changes in those due to reduced food waste.

(28)

28 Final consumption expenditure

a) b)

Germany '09 153,926 128,906

Netherlands '09 31,806 28,686

UK '09 95,774 66,022

UK '07 121,200 87,759

a) before deducting food waste

b) after deducting food waste

Table 4.1: Final consumption expenditure by households of agricultural and food products, before and after

deducting food waste

A final note on the UK, which is the only country for which there is data for two periods available. It can be observed that final consumption expenditure by households of agri-food products decreased from 2007 to 2009: from 121,200 to 95,774 , which is a decrease of 21%. According to the WRAP report food waste could be decreased by 21% from 2007 to 2009 due to the awareness raising measures implemented throughout the UK. If we assume, as discussed before, that an elimination of food waste implies a reduction of final consumption expenditure by households of agri-food products, these 21% might be the 21% reduction reflected in the WIOD data.

4.2.4INPUT-OUTPUT ANALYSIS

The fourth step was to perform the IO-Analysis. By deducting avoidable food waste from the Final consumption expenditure by households new final demand vectors for the individual case countries were created. Thus the task was now to calculate the total outputs in all industries necessary to satisfy a certain final demand vector, and the concomitant environmental pressures.

(29)

29 The WIOD provides data on two sources of emissions: emission by the industry and emissions by households. The emissions by industry change due to reduced food waste, whereas the emissions by households remain the same. Furthermore the WIOD considers eight different air pollutants which are: carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), nitrogen oxides (NOX), sulphur oxides (SOX), carbon monoxide (CO), non-methane volatile organic compounds (NMVOC), and ammonia (NH3). However only the first three air pollutants, CO2, CH4, and N2O, are considered as greenhouse gases which are embodied in the carbon footprint. CH4 and N2O must be “converted” into C02 equivalents in order to make them comparable, which is done by means of their 1-year global warming potential (GWP)16_{(see table 4.2). After adding the three greenhouse} gases to generate a single emissions vector, two emissions matrices could be created by means of the emissions multiplier: one using the original final demand vector and one using the new final demand vector, excluding food waste. A country's carbon footprint is equal to its domestic emissions, which are equal to the sum of total emissions produced in domestic production processes and emissions by households, minus emissions exports, plus emissions imports. The carbon footprint per capita can then be calculated by dividing total emissions by the country's population. A similar procedure was subsequently applied to derive the land use footprint and then the water use footprint. The WIOD provides data on the use of blue, green as well as grey water. In the context of this research only the use of blue water is considered, for the reasons mentioned in chapter 1. Therefore the impact of eliminated food waste on water use footprint was calculated, based on the use of blue water only. In contrast to the CO2 emissions and water use, land use is only given for the industry. A country's land use footprint therefore equals its total land used in domestic production processes, minus land use exports, plus land use imports. Thus after creating two land use matrices by means of the land use multiplier the country's (per capita) land use footprint can be determined.

GHG Chemical Name 100-year GWP source Carbon dioxide CO2 1 by definition

Methane CH4 34 Mhyre, 2013, p.714

Nitrous oxide N20 298 Mhyre, 2013, p.714

Table 4.2: Global warming potentials of Carbon dioxide, Methane, and Nitrous oxide, source (Myhre, et al., 2013)

(30)

30 4.2.5EVALUATION

(31)

31

5. E

MPIRICAL RESULTS

In this section I will evaluate the results that emerged from the IO-analysis. First I will analyse the impact of food waste on the carbon, land, and water footprints by assuming that the avoidable part of food waste could be prevented in the three case countries altogether. Thereafter follows an analysis of the potential changes of displaced environmental pressures, resulting from the elimination of avoidable food waste at the consumption stage. In the third and final part of this chapter, I will compare the results of this study to the results of previous studies in order to analyse whether the impacts of food waste were indeed underestimated. The focus will lie on the 2009 results for Germany, the Netherlands and the UK. The results of 2007, which are only available for the UK, will be consulted for comparison purposes.

5.1 I

MPACT OF FOOD WASTE ON FOOTPRINTS

In the following paragraphs I will analyse the impact of reducing avoidable food waste on their carbon, land, and water footprints in Germany, the Netherlands, and the UK. 5.1.1CARBON FOOTPRINT

In 2009 worldwide carbon emissions amounted to 42,275 megatonnes of CO2 -equivalent (mtCO2e), compared to 42,641 mtCO2e in 2007 (i.e. a percentage decrease of 0.86). 15% of the world wide carbon emissions (6,345 mtCO2e) can be accounted for by the EU27 countries, though they constitute no more than 7% of the world population. Interestingly, Germany, and the UK together were responsible for one third of European emissions (see figure 5.1). The worldwide average per capita carbon footprint for the year 2009 was 12.30 tonnes of CO2 equivalent (tCO2e) per capita. The EU27 average lied above the worldwide average with 12.92 tCO2e per capita and so did the per capita carbon footprints of Germany and the Netherlands (see appendix 11).

(32)

32 The impact of reduced food waste was found to be smallest with the carbon footprint. This is due to the nature of the FSC which is highly land- and also water-intensive. From the results (see appendix 11) we can see that reduced food waste had the largest impact on the British per capita carbon footprint, both in 2007 and in 2009. In 2007 the British per capita carbon footprint could be reduced by 0.57 tCO2e , i.e. a percentage decrease of 3.29%. In 2009 per capita carbon footprints decreased by 0.52 tCO2e, which is a reduction of 3.71%. The German per capita carbon footprint reduction was the second largest: the per capita carbon footprint decreased by 0.3 tCO2e (2.04%). The Dutch carbon footprint decreased by 0.31 tCO2e (1.73%). Despite their reductions, the per capita carbon footprints of all case countries still lied above the worldwide as well as the EU27 average. Total worldwide carbon emissions and total European carbon emissions (see table 5.1) reduced by about 62 mtCO2e. That is a 0.15% reduction of total worldwide emissions and a reduction of 0.98% of total European emissions.

2009 Carbon Footprint Land Footprint Water Footprint

in MtCO2e in Mha in km3 a) b) a) b) a) b) World 42,275 42,213 7,058 7,031 1,987 1,983 EU27 6,345 6,282 612 585 204 200 Germany 1,233 1,208 101 90 30 28 The Netherlands 293 288 30 27 7 6 The UK 871 839 76 62 17 16

a) before deducting food waste

b) after deducting food waste

2007 Carbon Footprint Land Footprint Water Footprint

in MtCO2e in Mha in km3 a) b) a) b) a) b) World 42,641 42,607 7,363 7,348 1,940 1,938 The UK 1,049 1,015 93 78 22 21

a) before deducting food waste

b) after deducting food waste

(33)

33 5.1.2LAND FOOTPRINT

Worldwide land use decreased by 4% from 7,363 mega hectare (Mha) in 2007 to 7,058 Mha in 2009. In 2009 8.7% of worldwide land use can be ascribed to EU countries, of which about one third can be attributed to Germany and the UK. The worldwide average per capita land footprint was 1.76 ha/capita in 2009. The European average was with 1.38 ha/capita lower than the worldwide average. Moreover, it must be stated that the Netherlands were the sole European country with a per capita land footprint above the worldwide average (1.81 ha/capita).

It is apparent from table 5.1 and appendix 11 that reduced food waste had the strongest impact on land use. The case countries' land footprints could be reduced by, on average, as much as 12.81% in 2009. Once more the largest impact could be observed with the reduction of British footprints. In 2009 the per capita land footprint could be reduced by nearly a fifth, i.e. 0.23 ha/capita or 18.64%, compared to its pre-food waste elimination levels (appendix 11). In 2007 the land footprint declined by nearly 16%, from 1.53 ha/capita to 1.28 ha/capita. The reduction of German and Dutch land footprints is much the same, with the German land footprint reducing by 0.12 ha/capita or 10.13%, and the Dutch land footprint reducing by 0.18 ha/capita (9.65%). The stronger impact of reduced food waste on land use is also reflected in worldwide land use, and in European land use, which reduced by 27 Mha, that is a percentage reduction of 0.38% of worldwide land use and a percentage reduction of as much as 4.44% of European land use.

5.1.3WATER FOOTPRINT

(34)

34 Compared to the effect on carbon footprints, which was rather small, and on land footprints, which can be considered too large, the effect on water footprints are said to be medium. On average the per capita footprints of Germany, the Netherlands and the UK decreased by 6.72% as a result of reducing avoidable food waste. Once more the footprint decrease as a result of food waste reduction is found to be most pronounced with the British footprint. As a result of reducing food waste, the British water footprint was reduced by 139 m3_{/capita (i.e. a percentage decrease of 8%), and by nearly 10% in} 2009, i.e. by 27 m3_{/capita. Exceptionally, the reduction of the Dutch water footprint is} larger than the reduction of the German footprint, which was reduced by 5.66% and 4.79% respectively. The effect on worldwide blue water use and on carbon emissions are similarly small: worldwide water use was reduced by 0.17%. The effect on total European water use, with a reduction of 1.7%, can considered to be medium.

5.2 D

ISPLACEMENT OF ENVIRONMENTAL PRESSURES

(35)

35 commodities to be produced domestically, it can be expected that there is a slight tendency in the UK for the reduction of domestic environmental pressure to predominate. For the Netherlands it can be expected that imports of environmental pressures will be reduced significantly. However, owing to the agri-food based export structure of the Netherlands, one might also expect reduced food waste in Germany, and the UK to be impacting on Dutch exports of environmental pressures. The second question I aim to answer is which countries were most affected by reducing avoidable food waste in the case countries. It is expected that this will be most likely countries with large agri-food exports.

The rest of the world (ROW) will be excluded from this analysis, as it does not represent a country on its own, but rather an aggregate of the remaining 155 countries in the world that are not separately listed in the WIOD. Therefore analysing the impact of reduced food waste on environmental pressure displaced to the ROW can be misleading, as supposedly it will be rather large. Furthermore in this part of the chapter I will consider total changes i.e. differences, rather than percentage differences. The example of Luxemburg's export of land use provided subsequently, will illustrate the reasoning behind my choice. According to the percentage decrease, following the reduction of food waste in Germany, the Netherlands, and the UK in 2009, Luxemburg's land use exports experienced the second largest decrease. The reduction amounted to 7.29%. At first sight this seems rather large, however when looking at the total decrease, it can be observed that the reduction only amounts to 0.012 Mha, which puts into perspective the large percentage decrease.

(36)

36 5.2.1DISPLACEMENT OF CARBON EMISSIONS

From the figure in appendix 14, it becomes apparent that while both domestic emissions and imported/displaced emissions decreased considerably, following the reduction of avoidable food waste, the impact on reduced displaced emissions clearly dominates. In general, the composition of reduced carbon footprints is largely import-based, with imports of emissions making up a minimum of 60% of the total reduction. As predicted, there is a particularly strong decrease of displaced emissions in the Netherlands. While domestic emissions decreased by 0.7 mtCO2e, the decrease of displaced emissions was about six times larger.

Emissions exports in MtCO2e a) b) 1) Brazil 248 246 2) The Netherlands 123 120 3) France 125 123

a) before deducting food waste

b) after deducting food waste

Table 5.2: Top-3 countries with the largest reductions in emissions exports.

The top three of emission exports reduction is headed by Brazil, in terms of country with the largest reduction in emissions exports (see table 5.2). Following the reduction of avoidable food waste in Germany, the Netherlands, and the UK in 2009, emission exports decreased by nearly 3 mtCO2e. As expected, the reduction of avoidable food waste in the case countries also had a considerable impact on the Dutch exports of emissions, which was similar in size to the impact on Brazilian emission exports. Another country strongly affected by reduced food waste in the case countries was France, whose emission exports decreased by 2 mtCO2e.

5.2.2DISPLACEMENT OF LAND USE

(37)

37

Land use exports

in Mha a) b) 1) Brazil 140 138 2) Australia 133 132 3) Spain 13 12

b) after deducting food waste Table 5.3: Top-3 countries with the largest reductions in land use exports.

Again the top three of export reductions is headed by Brazil (see table 5.3), whose land use exports decreased by about 2 Mha as a result of reduced avoidable food waste in the case countries. Besides Brazil, the reductions of Australian land use exports, with about 1 Mha, and of Spanish land use exports, with about 0.7 Mha are also quite remarkable. 5.2.3DISPLACEMENT OF WATER USE

The displacement of environmental pressures becomes evident when looking at the composition of the water use footprint reduction of the three case countries. Appendix 14 is quite revealing in this sense. It shows that domestic water use only reduced marginally, and the larger part (i.e. more than 80% for all three case countries) of the footprint reduction is caused by reduced water use imports. Once again, the Dutch trend towards importing both products from the agricultural, and the food and beverages sector is strongly reflected in the composition of the Dutch water-use footprint: the decrease of water use imports was about 40 times larger than that of domestic water use.

(38)

38

Water use exports

in km3 a) b) 1) Spain 6.55 6.25 2) India 24.55 24.37 3) USA 26.70 26.56

b) after deducting food waste

Table 5.4: Top-3 countries with the largest reductions in water use exports.

5.3 C

OMPARING ENVIRONMENTAL IMPACTS

In the introduction of this research paper, I claimed that traditional methods used to calculate the environmental impacts of food waste produce downward biased estimates. In this part of the chapter, I will compare the results from this research's IO Analysis to available estimates of the environmental impacts of food waste in the case countries. The only case country that has such estimates available is the UK. In the supplementary report to the study by Quested, Ingle, and Parry (2013), estimates of the GHG emissions and land use requirements associated with the avoidable part of British food waste are provided. Based on the assumption that if avoidable food waste were to be eliminated, the resources used for its production would be available for other uses, or not be consumed, the authors estimate that 17 mln tCO2e17 and 19,000 km2 can be associated with avoidable food waste generated by British households in 2009.

The IO Analysis employed in this research paper generated the following results: comparing the British national carbon footprint before and after reducing avoidable food waste, a difference of 32 mtCO2e, i.e. 32 mln tCO2e, becomes apparent. This difference embodies the GHG emissions that can be associated with British avoidable food waste. Applying the same procedure to the British land footprint, a difference of 14 Mha emerges, i.e. 140,000 km2_{. Incontestably, IO Analysis produced significantly larger} results with regards to the environmental impacts of food waste than those found in the British food waste report.

(39)

39

6. D

ISCUSSION

&

CONCLUSION

This chapter will start with a discussion of the results that emerged from the IO Analysis. Subsequently, I will put the results into perspective by looking at their implications. This will be followed by a part in which I look at existing measures that tackle the issue of food waste. Moreover, attention will also be paid to the limitations of this research and suggestions for future research will be provided. Lastly, a a brief conclusion will be provided.

6.1 D

ISCUSSION OF THE EMPIRICAL RESULTS

The results of the IO Analysis are quite distressing but, at the same time they highlight the potential of reducing food waste in making the FSC more sustainable and in lowering worldwide environmental pressures.

(40)

40 emitted in vain in Brazil (United States Environmental Protection Agency, 2014). The size of Brazilian land that has unnecessarily been stressed for food exports to German, the Netherlands, or British final demand, where it was wasted, is about equal to Slovenia's or El Salvador's total land area which are 20,140 km2_and _{20,720 km}2 respectively (The World Bank, 2015). With 300 billion litres of water saved in Spain, owing to reduced food waste in the case countries, about 6 billion people could have had enough water to fulfil their basic needs of eating and personal hygiene (Institute Water for Africa, 2015). These quantifications make the technical results more tangible and highlight the strong need to take measures against food waste as well as to highlight the potential of reducing food waste in making the FSC more sustainable, and above all lowering worldwide environmental pressures.

6.2 I

MPLICATIONS

Quantifying the environmental impacts of food waste