Abstract
Throughout the developed world, food is treated as a disposable commodity. Between a third and half of all food produced for human consumption globally is estimated to be wasted. However, attempts to quantify the actual magnitude of food wasted globally are constrained by limited data, particularly from developing countries. This article attempts to quantify total food waste generation (including both pre-consumer food losses, as well as post-consumer food waste) in South Africa. The estimates are based on available food supply data for South Africa and on estimates of average food waste generation at each step of the food supply chain for sub-Saharan Africa. The preliminary estimate of the magnitude of food waste generation in South Africa is in the order of 9.04 million tonnes per annum. On a per capita basis, overall food waste in South Africa in 2007 is estimated at 177 kg/capita/annum and consumption waste at 7 kg/capita/annum. However, these preliminary figures should be used with caution and are subject to verification through ongoing research.
Introduction
Attempts to quantify food waste are motivated by the need to assess the scale of waste in relation to global malnutrition (Parfitt et al., 2010) and to improve the climate for recovering and composting source-separated organic waste (Massachusetts Department of Environmental Protection, 2002). Food waste is composed of raw or cooked food materials and includes food waste before, during and after meal preparation in the household, as well as food losses in the process of food manufacturing, distribution, retail and food services activities (European Commission, 2010). Food waste is a potentially important factor in efforts to combat hunger and improve food security in poor countries (Gustavsson et al., 2011). At the same time, organic waste (including garden and food waste) is a problematic waste stream in landfills, where it contributes to greenhouse gas emissions (Hartman and Ahring, 2006).
The aim of this article is to quantify the magnitude of food waste in South Africa, in order to inform future action to reduce and better manage the food waste problem. The goal of the article is to provide a preliminary assessment of food waste generation at various stages in the food supply chain in South Africa, as part of ongoing research.
Problems associated with food waste
Food waste gives rise to three inter-related problems. The first relates to food security. According to the 2010 State of Food Insecurity in the World report (FAO, 2010), a total of 925 million people globally—almost one in seven people—are estimated to be undernourished. Developing countries account for 98% of the world’s undernourished people (FAO, 2010). A 2008 study revealed that 70% of poor urban households in South Africa live in conditions of ‘significant’ or ‘severe’ food insecurity (Frayne et al., 2009). To meet the challenge of feeding growing populations and addressing food insecurity, massive reductions in the amount of food wasted after production are needed (Lundqvist et al., 2008).
The second problem relates to greenhouse gas emissions along the food supply chain. On average, greenhouse gas emissions throughout the food supply chain range between 2.8 and 4.14 tonnes of carbon dioxide equivalent (tCO2e) per tonne of food, broken down as follows (Bakas, 2010):
manufacturing/processing (including agriculture): between 1.95 and 2.29 tCO2 e per tonne of food;
distribution and retail: between 0.1 and 0.8 tCO2 e per tonne of food;
consumption: between 0.3 and 0.6 tCO2 e per tonne of food;
end-of-life (landfill): 0.45 tCO2e per tonne of food.
Potential inefficiencies in the food value chain therefore have the potential to contribute up to 4.14 tCO2e per tonne of food wasted, contributing a significant component of a country’s emission footprint. While the disposal of organic waste (including food waste) is estimated to contribute 4.3% to South Africa’s greenhouse gas emissions, agriculture contributes 9.3% to the country’s greenhouse gas emissions (DEA, 2009, 2010).
Finally, the third problem relates to waste disposal. Landfilling is generally considered the most practical and cheapest waste management method in South Africa. However, the scarcity of available land in close proximity to areas of waste generation, as well as emissions of landfill gas (with high concentrations of methane), have made landfilling a less attractive management option (Hartmann and Ahring, 2006). For example, the disposal of organic waste (including food waste) is estimated to contribute 4.3% to South Africa’s greenhouse gas emissions. Introducing alternative end-of-life treatment technologies for food waste could, therefore, reduce the greenhouse gas emissions from food waste and should be considered.
Indeed, disposal of organic waste (including food waste) to landfill is outlawed in many countries (DEA, 2010), including Germany, Sweden, Canada; the phasing out of these practices is now also a priority in South Africa (DEA, 2011). Municipalities have to take responsibility for diverting organic waste to composting or biogas digesters (DEA, 2011). Waste from food processing could also be a source of valuable, functional compounds, such as antioxidants (Schieber et al., 2001). Cheap, renewable and abundant sources of antioxidant compounds are in high demand (Moure et al., 2001). Continued research in this field is therefore essential, but requires fairly accurate data on waste quantities and composition at the different stages in the food supply chain. Furthermore, accurate estimates of food waste would assist municipalities to meet their obligations in terms of national priorities.
Food losses in the food supply chain
Food losses occur at all stages in the food supply chain, from initial agricultural production to final household consumption (Gustavsson et al., 2011), including during food storage, transportation, food processing, at retailers, and in the kitchens of restaurants, hotels and households (Lundqvist et al., 2008). A review of food waste in the global food supply chain (Parfitt et al., 2010) found that ‘as much as half of all food grown is lost or wasted before and after it reaches the consumer’ (Lundqvist et al., 2008: 4). Other studies suggest that roughly a third of food produced for human consumption is lost or wasted globally (Gustavsson et al., 2011). This means that resources used in the production of this food are also wasted, while the greenhouse gas emissions caused by the production of wasted food are generated in vain (Gustavsson et al., 2011).
There is a diversity of causes of food waste throughout the food supply chain. Post-harvest losses are influenced by available technologies and the extent to which markets for agricultural produce have developed (Parfitt et al., 2010). Causes of household food waste vary considerably from one area to another, as a result of cultural practices, climate, diet and socio-economic factors (e.g. household size, household income and frequency of eating out) (European Commission, 2010). Parfitt et al. (2010) identified three inter-related drivers influencing food waste:
urbanisation and the contraction of agricultural markets—extended food supply chains are required to supply food to an increasing urban population. These require improvements in road, transportation and marketing infrastructure, while keeping food affordable for the poor;
dietary transition—growth in household incomes, especially in developing countries, results in changes in dietary patterns from consumption of starchy food staples to more diversified diets. Dietary transition leads to increased consumption of vulnerable, short shelf-life items such as fruit and vegetables, resulting in greater food waste (Lundqvist et al., 2008);
increased globalisation of trade—agricultural exports may present a threat to the development of domestic markets. Global food trade increases transport distances in the food supply chain, consequently increasing food losses.
Total per capita food losses throughout the food supply chain in Europe and North America amount to 280–300 kg/annum compared with 120–170 kg/annum in sub-Saharan Africa and South/Southeast Asia (Gustavsson et al., 2011). The proportion of food wasted specifically by consumers on a per capita basis is much higher in developed countries (95–115 kg/annum in Europe and North America) than in developing countries (6–11 kg/annum in sub-Saharan Africa and South/Southeast Asia) (Gustavsson et al., 2011). The causes of food waste in low-income countries are mainly connected to financial, managerial and technical limitations in harvesting techniques, storage and cooling facilities in difficult climatic conditions, infrastructure, and packaging and marketing systems. By contrast, the causes of food waste in medium/high income countries mainly relate to consumer behaviour and to a lack of coordination between different actors in the supply chain (Gustavsson et al. 2011).
A study on food wasted in the UK (WRAP, 2008) showed that consumers throw away about a third of the food that they buy. Parfitt et al. (2010) quote international studies conducted in various countries, indicating that food waste at the household level varies between 8% (the Netherlands) and 25% (USA) of food purchased. Much of this food ‘waste’ (e.g. 61% in the UK) is still suitable for human consumption (WRAP, 2008) and can therefore be considered avoidable. The most common reason provided by consumers for food being wasted is that it is left unused (61% of the avoidable waste) or that too much has been cooked or prepared (WRAP, 2008). Production of food that is not consumed while fit for human consumption results in wasted resources at the production stage while it further contributes to food insecurity. Furthermore, food waste disposed of on landfills contributes significantly to the environmental impacts of waste.
Food waste in South Africa
No primary data collection has been undertaken during this research to date. Therefore, the results are based on available data and assumptions as reported in literature and described herein. The results should be used with caution, as they reflect a preliminary assessment of food waste generation in South Africa. However, research is ongoing to provide more accurate results in the future. An extensive search for available data on food waste generation in South Africa was conducted. Landfill site records, waste information databases, as well as municipal integrated waste management plans were scrutinised for data on food waste. Data extracted from the South African waste information system (SAWIS) was disappointing, with only four sites reporting food waste data between 2007 and 2010. Landfill data obtained from metropolitan municipalities revealed that, in general, municipalities record pre-consumer food waste, requiring safe disposal certificates, although this data is not necessarily reported. This lack of reported data on food waste could be explained by the fact that waste separation at source only became a legal requirement in South Africa in 2009 (Republic of South Africa, 2008), while reporting to the SAWIS is still voluntary and not yet required at this level of detail. Reporting to the SAWIS is expected to become mandatory in 2012 (O, Baloyi, personal communication).
Of the 112 municipal waste management plans reviewed, only two reported food waste figures. This is not surprising when considering that very few waste characterisation studies have been undertaken in South Africa. Sibernagl (2011) reports that two studies were undertaken in Johannesburg and another two in the Western Cape, with some work in Rustenburg and Bloemfontein. A report on another such study conducted in Limpopo was published in 2011 (Ogola et al., 2011).
The waste characterisation study by Jarrod Ball and Associates (2001) conducted in Johannesburg found that putrescible waste (most of which is assumed to be food waste) varies depending on household income level and site. From the available studies it seems that waste from low income households in urban areas contains a higher proportion of food waste by weight (12–26.2%) compared with high income households (7–7.6%). This can be explained by the fact that food is a basic need—poor households therefore spend a higher proportion of their budgets on food compared with high income households (Martins, 2007). In addition, poor households can be expected to throw away less of other waste types owing to their restricted budget for buying non-food items. However, the Limpopo study (covering a more rural area) reports higher food waste proportions in high income households compared with low income households (Ogola et al., 2011). The seemingly contradictory results from the Limpopo study highlight the urgent need for more representative waste characterisation studies covering both rural and urban areas in South Africa.
The available waste-related data was therefore not very useful in estimating food waste at a national level, as they are not statistically representative. It was only possible to estimate food waste generation figures for South Africa based on data on food production (FAOSTAT 2010a, 2010b) and on estimates of proportions of food waste at various stages of the food supply chain (Gustavsson et al., 2011), as explained in the following section.
Research method
Our estimates of food waste in South Africa are based on some of the results of studies carried out by the Swedish Institute for Food and Biotechnology (SIK) as reported by the Food and Agriculture Organisation (FAO) of the United Nations (UN) (Gustavsson et al., 2011) for sub-Saharan African countries. These studies highlighted food losses (referring only to edible parts of the commodities) occurring along the food supply chain, and made assessments and assumptions of the magnitude of these losses, focusing on quantitative weight losses. Estimates of food waste at each stage of the food supply chain in sub-Saharan Africa (as a percentage by weight of the food entering each step), as estimated for various commodity groups by Gustavsson et al. (2011), are presented in Table 1.
Estimated/assumed waste percentage for each commodity group in each step of the food supply chain for sub-Saharan Africa. (Reprinted from Food and Agriculture Organization of the United Nations (2011), Gustavsson J, Cederberg C, Sonesson U, van Otterdijk R and Meybeck A, Global food losses and food waste: Extent, Causes and Prevention with permission from the FAO).
Based on the estimates in Table 1, although using a different methodology from Gustavsson et al. (2011), it was possible to calculate the food losses at each stage of the food supply chain in South Africa based on production figures for each corresponding commodity group. The production of food commodities in each country, including in South Africa, is recorded and reported annually by the FAO of the UN. South African food production figures are summarised in Table 2. The production volumes for all commodities except fish and seafood were collected from the FAO Statistical Yearbook 2010 (FAOSTAT, 2010b). Data for fish and seafood were collected from the FAO Food Balance Sheets (FAOSTAT, 2010a). The same commodity groupings as used by Gustavsson et al. (2011) were used. Food production figures per commodity group in South Africa are presented in Table 2.
Food production per commodity group in South Africa (source: FAOSTAT, 2010a, 2010b).
Food waste at each step of the food supply chain for each commodity group in South Africa was calculated using the percentages as outlined in Table 1. As production figures for the different commodities vary year-on-year, it was decided to use the average production figures over a three-year period (2007–2009) as indicated in the last column of Table 2 to estimate annual food waste generation. As Gustavsson et al.’s (2011) food loss percentages refer to percentages (by weight) of food entering each stage, it was assumed that the amount of food entering each stage equals the amount of food entering the previous stage, less food losses at that stage. The food loss calculated at each stage of the food supply chain was therefore subtracted from the production figure for that stage before the calculation for the next stage was done. For example, agricultural production, minus losses during agricultural production, equals the input figure for the calculation of waste generated during the post-harvest handling and storage stage of the food supply chain, and so on. Note that Gustavsson et al. (2011) used conversion factors to convert primary production (reported in the food balance sheets) to the edible part of commodities. It could, therefore, be expected that the calculations in this article (in which conversion factors were not used) will be higher compared with the results reported in Gustavsson et al. (2011).
The international food waste estimates presented by Gustavsson et al. (2011) were based on 2007 data. Thus, to present food waste estimates for South Africa in per capita terms (useful for comparison with other countries), it was necessary to use food production and population data from the same year, namely 2007. Population statistics for 2007 were obtained from Statistics South Africa (StatsSA, 2010).
Results and discussion
The results of the analysis are presented in Table 3. Using the method described in the previous section, the preliminary average estimate for food waste in South Africa is in the order of 9.04 million tonnes per annum. This amounts to 31.4% of average annual agricultural production (28.79 million tonnes per annum) (Table 3).
Calculated average per annum food waste generation figures for South Africa (2007–2009).
The percentage contribution to total food waste at each step in the food supply chain is illustrated in Figure 1. The split in waste generation between agricultural production, post-harvest handling and storage, and processing and packaging is fairly equal. The majority (8.67 million tonnes per annum or 95.9%) of the food waste generated annually is generated during the pre- consumer stages. Pre-consumer food waste amounts to 30.1% of average annual agricultural production. Only 4.1% (0.37 million tonnes) of the total food waste is generated at the consumption (post-consumer) stage—amounting to 1.3% of agricultural production.
Percentage contribution of each step in the food supply chain to total average food waste in South Africa.
The contribution of each commodity group to total food waste is illustrated in Figure 2. It is evident that fruit and vegetables, combined with roots and tubers, contribute 57% of the overall food waste stream, while fish, seafood and meat contribute only 6%. Focused efforts to reduce wastage of fruit and vegetables could, therefore, have a significant impact on overall food waste generation rates in South Africa. However, efforts to reduce meat waste are also important in order to reduce the environmental impacts of food waste [the production of meat protein has a much higher environmental load than processed protein food based on soybeans (Reijnders and Soret, 2003)].
Percentage contribution of each commodity group to total average food waste in South Africa.
In order to compare the results of this study with findings in the literature (Gustavsson et al., 2011), the 2007 per capita food waste generation for South Africa was calculated following the same approach as described above, but using the 2007 food production dataset rather than the three-year average as used above. The calculated food waste generation for 2007 is provided in Table 4.
Calculated food waste in South Africa in 2007.
In 2007, 8.52 million tonnes of food waste was generated by a population of 48.26 million (StatsSA, 2010). Per capita food waste generation in South Africa in 2007 is therefore calculated as 177 kg/capita. This figure seems reasonably accurate when compared with the per capita estimated food loss of 170 kg/annum for sub-Saharan Africa estimated by Gustavsson et al. (2011). Post-consumer food waste (353,000 tonnes per annum) accounts for 7 kg/per capita/annum compared with 6 kg/per capita/annum in sub-Saharan Africa estimated by Gustavsson et al. (2011). These differences are, in actual fact, small when considering that the calculations by Gustavsson et al. (2011) are based on only the edible portion of commodities. In addition, a higher per capita figure for South Africa compared with sub-Saharan Africa is expected, as South Africa has the highest gross domestic product (GDP) of all sub-Saharan African countries (World Bank, 2011) and there are indications of a strong correlation between a country’s GDP and waste generation (United Nations Environment Programme, 2011). The above estimates are likely to understate total food waste generated in South Africa. FAOSTAT data for 2007 (FAOSTAT, 2007) suggest that South Africa is a net importer of food, while the above calculations were based on local production figures only. Food loss in the food supply chain of imported products should therefore also be included in the calculations in order to provide a more realistic picture. In addition, South Africa is characterised by two parallel economies: the so-called ‘first’ or ‘formal’ economy, which is similar in nature to a developed country economy, and the ‘second’ or ‘marginalised’ economy (Vermeulen and Bienabe, 2007). In parallel to well-developed formal retail chain groups in the formal economy, a very large and growing informal market, especially for fresh fruit and vegetables, characterises the ‘second’ economy. The calculations above are based only on agricultural production in the formal economy. The informal market could, potentially, generate significant amounts of pre-consumer food waste due to inadequate transport, cooling and storage facilities, which is not accounted for here.
Conclusions
Food waste has a triple negative impact. Firstly, disposal of food waste at landfill impacts on the environment through atmospheric (particularly greenhouse gas) emissions, leachate, odours and pests. Secondly, atmospheric emissions caused by food production, processing and distribution of food to consumers are wasteful if the food is not used. Thirdly, wasted food exacerbates the problem of food insecurity.
In light of the threats associated with global change and considering the concern over food security in large parts of the developing world (Gustavsson et al., 2011), it is important to understand the magnitude of food waste globally. This paper sheds some light on the situation in South Africa, and forms part of ongoing research.
The calculated food waste estimate of approximately 9.04 million tonnes per annum in South Africa is conservative and likely to be an underestimate given the assumptions made. Therefore, it is recommended that the highly aggregated results presented here be verified through primary data collection throughout the food supply chain. A more accurate assessment of food waste could facilitate better waste management practices towards reducing greenhouse gas emissions and improving food security in South Africa.
Footnotes
Acknowledgements
The authors would like to thank the officials of various metropolitan municipalities for sharing landfill data. The efforts of Ms Manja Schubert in scrutinising municipal integrated waste management plans and of Ms Linda Godfrey in extracting data from the SAWIS and providing valuable comments on earlier versions of this paper are also acknowledged. The constructive comments of two anonymous reviewers are also acknowledged.
Funding
This research was funded by the CSIR through the Parliamentary Grant budget.