Assessing the vulnerability of Stellenbosch's food system and possibilities for a local food economy

Abstract

Food insecurity is high in Stellenbosch, despite a strong agricultural context. While the causes are complex, it is clear that the situation will be worsened by the effects of climate change and the end of cheap oil on the unstable global food system. This research compiled existing statistical information to present an overview of the current status of food production, distribution and consumption in Stellenbosch to determine key vulnerabilities and opportunities to strengthen resilience. The region produces predominantly wine grapes and fruit for export, while relying on imported produce for consumption. Key vulnerabilities of the food system were identified as: focus on wine production for export, dominance of large-scale commercial agriculture, proliferation of conventional farming methods, long value chains, dominance of national retailers, food insecurity and poor nutrition. Suggestions are made for increasing the localisation of the food system to address inequities and increase resilience to the polycrisis.

Keywords:

• food security
• local food economies
• sustainable development

1. Introduction

Global food riots in 2008 brought to international attention the increasing price of basic food and the alarming global food crisis that is directly affecting the lives of millions of people and impeding human development (FAO, 2008:2). While market forces have largely been blamed for recent events (FAO, 2008:9), much deeper, more complex concerns lie beneath the surface of food insecurity because of the way the modern food system functions. A major concern is that the commoditisation of food has resulted in the social and environmental costs of food production, distribution and consumption being externalised and forced onto other actors, timescales and localities (Pretty, 2002:52).

At the same time, multiple alarms are being raised, signalling that our planet is a system in crisis. This state of global polycrisis ‘consists of a multiple set of nested crises that tend to reinforce one another’ (Swilling, 2009) and that are not reducible to singular cause and effect relationships. These nested crises can be neatly summarised through several key international reports (see Table 1).

Table 1: The global polycrisis and related key documents

Issue	Reference	Message
Climate change	IPCC (2007)	Climate change is now unavoidable and mitigation is essential.
Inequality	UNDP (1998)	Policy focus must shift from poverty to inequality, consumption must be de-linked from development.
Urbanisation	UN DESA (2008)	By the end of 2008, over half the world's population lived in cities (with massive urbanisation predicted in the developing world).
Slums	UN-HABITAT (2003)	One-third of the world's population lives in slums, ‘locus of poverty’ is shifting to cities.
Peak oil	ASPO (2008), IEA (2008)	Oil production will peak soon (ASPO), the era of cheap oil is over (IEA).
Ecosystem degradation	MA (2005)	Human activity has damaged ecosystems to the extent that their ability to support human life is threatened.
Agriculture	IAASTD (2008)	Agriculture needs to find ways to improve livelihoods of small-scale farmers and increase productivity without damaging the environment.

Building local food economies has been proposed as a way to address some of the negative effects of the modern food system and promote more sustainable, equitable, resilient and empowered local communities (Feenstra, 1997; Shuman, 1997; Norberg-Hodge et al., 2002; Winter, 2003; McKibben, 2007; Hopkins, 2008). Localisation involves prioritising the flow of resources (financial, human, social, environmental and others) within a network of community-based enterprises that produce and distribute food at the local scale for local consumption (Schulschenk, 2010). A ‘local food economy’ thus consists of local food initiatives at the community level within the context of the global modern food system.

This research aimed to assess the extent to which local food economies can promote sustainability through a case study of the Stellenbosch Municipal Area (hereafter referred to simply as ‘Stellenbosch’). Covering an area of 831 square kilometres and with a current estimated population of 208 950 (SM, 2009:03), Stellenbosch is a system small enough to be studied critically yet large enough to reflect the complexities of human settlements. It thus presented an interesting and noteworthy case study for the research.

The area has a comparative advantage in the production of wine, largely exported, yet high levels of food insecurity, poverty and inequality persist here. This leads one to question whether conventional theories of trade are working for the poorest in the Stellenbosch community. This paper is a thought experiment. It aims to highlight the existing issues, risks and inequities in the Stellenbosch food system and then to explain the threats posed to Stellenbosch's food system by the polycrisis. This required an investigation into the current status of the food system. Having assessed this status, our next step was to think about steps that could be taken to increase the resilience of Stellenbosch's food system and the food security of its most vulnerable residents, using the local food economies literature as a reference.

2. Overview of Stellenbosch's current challenges

This section gives an overview of the region and highlights the challenges it faces in dealing with problems in the areas of agriculture, food security and inequality. Stellenbosch's population is a diverse mix of ethnic groups: 55% coloured, 26% African, 18.5% white and 0.5% Asian (Zietsman, 2007) and its settlement types vary from dense urban to remote rural. Agriculture accounts for over 80% of land use and the predominant farming activity is wine production (StatsSA, 2006). However, the main revenue streams are derived from other sectors, including finance, business services and manufacturing. The largest employer is manufacturing, followed by wholesale and retail, and community and personal services. Manufacturing is closely linked to the agricultural sector and includes agro-processing activities, so while agricultural production is not a large employer it indirectly supports a number of other sectors, up and down the value chain. The proportion of workers in the agricultural sector has declined dramatically since 2001, from 24% to 6.9% of total employment (SM, 2009:20); this may be due to shifts to other sectors but no further research could be found to explain this decrease.

Stellenbosch currently has some of the greatest discrepancies between poverty and wealth (SM, 2009:20) in a country which has a Gini coefficient of 0.72 (StatsSA, 2008:03), which is considered to place South Africa in the top 10 most unequal countries globally (UNDP, 2008). Major challenges recognised by the ward councillors for Stellenbosch are housing, unemployment, crime, poverty alleviation, urban growth, conservation and cleansing (SM, 2009:26–32).

No exact figures are available for the level of food insecurity in Stellenbosch or South Africa, although a range of estimates can be found. Altman et al. say that national estimates range from 12.2% to 52% of households regularly experiencing hunger (2009). Statistics South Africa says that 14.5% of Western Cape residents have inadequate or severely inadequate access to food (StatsSA, 2010),1 while the Urban Food Security Baseline Survey conducted in late 2008 found that food insecurity among the urban poor in the cities of Johannesburg, Cape Town and Msunduzi can be as high as 70% (Frayne et al., 2009). Gericke and Labadarios (2007) argue that the 2005 National Food Consumption Survey's Fortification Baseline showed that the proportion of households experiencing hunger was as high as 51%, with those at risk of hunger at 28%, translating into only one in five households being food secure. The Department of Human Nutrition at the University of Stellenbosch has correlated population density with a synthesis of several indicators representing a composite index of human needs that can imply food insecurity; part of Stellenbosch was identified as having high food insecurity (personal communication, A Van Niekerk, Director, Centre for Geographical Analysis, 30 October 2009). The authors' personal experience of living and working in Stellenbosch confirms high levels of hunger among school-going children in the Lynedoch area.

It is estimated that poor people in South Africa may spend 60 to 80% of their incomes on staple food and constantly increasing food prices are forcing these families into conditions of poorer nutrition (Naylor, 2008). It is also important to bear in mind that households have a number of objectives and may choose to forgo food in order to preserve other assets or choices (Ericksen, 2008:236). According to Ericksen, people's food choices are not always rational: ‘prices are influential, as are income levels, cultural traditions or preferences, social values, education and health status’ (2008:238). These issues point to problems of poverty and deep inequality that need to be addressed through targeted growth that values social upliftment and integration above purely economic considerations.

These social problems will be compounded by the resource-based problems that Stellenbosch is already facing, which mirror those being faced on an international scale, including the problem of how to overcome the current polycrisis on the path towards sustainable development. Water shortages are commonplace during summer and expected to worsen as population growth and climate change place more demand on dwindling water supplies. This will have a direct impact on the local economy as well as the quality of life for residents. The national energy provider, Eskom, has placed limits on growth and expansion through electricity restrictions (SM, 2008:37). The landfill is full and sewerage treatment plants are at capacity, with overflow commonplace in the winter months. Stellenbosch lies within the Cape Floral Kingdom, an internationally recognised biodiversity hotspot, and less than 10% of the critically endangered Renosterveld vegetation type endemic to the region remains today.

Stellenbosch faces threats to the basic ecosystems that provide people with a clean and healthy environment that sustains life, protects them from dangers such as flooding, and provides food or energy. A degraded environment severely limits the capacity for development and the extent to which poverty and inequality can be successfully overcome. In the context of climate change, oil production peaking and population growth, Stellenbosch cannot afford to continue down an energy-dependent and waste-intensive path of development. But its problems also present an opportunity for building an alternative path to development, founded on environmental sustainability and social equality.

3. The global polycrisis and local food economies

This section highlights the implications of the global polycrisis for the food system and looks at the benefits of increased localisation. Each of the crises shown in Table 1 implies restrictions to our choice of future development paths. Living conditions for the poor must be improved in ways that do not cause further damage to ecosystems but are environmentally and socially sustainable.

The global food system has also contributed to the ecosystem aspect of the polycrisis: 70% of all fresh water and 25% of the earth's land surface are used for food cultivation (MA, 2005). It also has significant climate change consequences: about 70% of anthropogenic nitrous oxide emissions are attributable to agriculture, mostly from land conversion and nitrogen fertiliser use (MA, 2005). The food system is both causing and being increasingly constrained by the polycrisis.

Growth in agricultural productivity is flattening (UNEP, 2009:16), signalling a system in crisis as demand continues to grow in the context of growing constraints on food production. The Food and Agriculture Organisation reported in 2010 that 925 million people were suffering from hunger, ‘even after the recent food and financial crises have largely passed’, and said that this ‘indicates a deeper structural problem that gravely threatens the ability to achieve internationally agreed goals on hunger reduction’ (FAO, 2010:4). The problems will only be further compounded by climate change (which both affects and is affected by the modern food system), and the end of cheap oil on which we depend.

History has shown that even when the world food system has produced enough food to feed everyone on earth, high levels of malnourishment have persisted, indicating that conventional food systems fail the poorest and most vulnerable in society. Via Campesina claims that we have lost our connection to the basic food systems that sustain life and thereby lost control over determining our own food, livelihood and environmental security (2008).

Localisation of food systems is presented as a counter movement to the modern food system. While it cannot be conclusively shown that localisation reduces greenhouse gas emissions, it can be shown that it promotes greater sustainability by building community networks and embeddedness that strengthen community resilience and accountability and that it allows for the social and environmental costs of the food system to be considered and included in the prices of food (Feenstra, 1997:28). In other words, going local is not an argument about reducing greenhouse gas emissions, but rather about building the resilience of the local food system to the current and future impacts of the polycrisis (especially rising oil prices), while increasing the equity of the food system.

Equity is increased through the multiplier effect of local spending. A study by Sonntag found that ‘locally directed spending by consumers more than doubles the number of dollars circulating among businesses in the community’ (2008:v). A further study by Ward and Lewis (2002:20) found that for every £10 paid to a local producer through an organic box scheme,2 the money circulated within the local economy to make a total contribution of £25 to the local economy (compared to £14 to the local economy if the £10 was spent at a local supermarket). In this way, money spent on locally produced food generates nearly twice as much income for the local economy (NEF, 2001) and promotes stronger social cohesion for the entire local community (Taylor et al., 2005).

Though it must be realised that localisation does not automatically mean increased sustainability or equity, and must of course be inclusive and critically reflective, some researchers claim that a local food economy based on principles of agro-ecological production, environmentally sensitive and ethical distribution and nutritional consumption has tremendous potential to promote sustainability for a local community (Hinrichs, 2003:34; Peters et al., 2008:02). Marsden & Smith argue that these ‘innovative regional and local forms of development need to be seen in the context of two major countervailing forces, within which local “value-capture” has to fit: globalisation and agrarian (agro-industrial) modernisation’ (2005:441). As Sonntag recognises, ‘what we are witnessing in the emergence of the local food economy is changing the idea of what makes for healthy economies – from growth based on commoditising resources to community stewardship of resource flows’ (2008:v). Developments like this also mean that the benefits of the food system can be more fairly spread among all the members of the community – from farm workers to small storekeepers and consumers (Norberg-Hodge et al., 2002:31).

Local food economies have their limits. Many local areas cannot produce all food groups, the local food producers may be unable to compete with cheap imported produce, and certain geographic and political contexts may be unsuitable for realising a healthy economy of this kind. But this option should be pursued in order to address the issues of democracy, stability and environmental security that remain critical to human development in the broader context and to promote stronger food systems. The crises listed in Table 1 are expected to alter method of food production and distribution drastically in the near future. Local food economies are presented as an alternative food system that builds social, economic and environmental capital at a community level while increasing community resilience to future shocks.

4. Research methods used to investigate the current Stellenbosch food system

An analysis of the current food system was conducted to provide a baseline for making recommendations on localising the food system. Enquiries at several Stellenbosch University departments and other research institutions revealed that no study had previously been undertaken to assess Stellenbosch's entire food system. Empirical data were sourced to build an understanding of the current status of local food production, distribution and consumption.

Production information was based largely on the Commercial Agricultural Census of 2002 (StatsSA, 2006), although certain figures were updated where possible. For example, an emerging farmers' survey conducted in 2006 had not been included in the Commercial Census (personal communication, J Murdoch, Western Cape Department of Agriculture, 27 May 2009), so farmers in Stellenbosch were contacted and interviewed and asked if they knew of other small farmers.

Distribution information was sought to provide a clear overview of the flow of food in Stellenbosch (i.e. where local produce goes and where locally consumed food comes from). Most Stellenbosch fresh produce was found to be for export oriented markets (Louw, 2009:40; SAWIS, 2009:26). Data on local distribution were extremely difficult to source, mostly because tracking systems were not in place (especially for vegetables) and because there were restrictions on data release for information that was tracked by the Red Meat Association, local abattoirs and the National Fresh Produce Markets (personal communication, E Becker, Operations Manager, Cape Town Fresh Produce Market, Epping, 31 August 2009; and W Malan, Information Manager, Statistics, Western Cape Department of Agriculture, 10 September 2009). Interviews were therefore conducted with selected local farmers, regional distributors and local retailers in order to discover the complexities and possible inefficiencies of the distribution system. It proved to be very difficult to determine the distribution without conducting extensive primary research.

To determine consumption patterns, interviews were conducted at the Department of Human Nutrition at Stellenbosch University and extensive research undertaken. These revealed that no food consumption survey had been undertaken specifically in Stellenbosch and no survey had been done for the Western Cape since 1973 (personal communication, N Steyn, Chief Research Specialist: Knowledge Systems (Nutrition), Human Sciences Research Council, 24 August 2009). Consumption patterns for the population of Stellenbosch were inferred from the National Food Consumption Report (Nel & Steyn, 2002), which is a collective report on several food consumption surveys completed between 1983 and 2000. This report was recommended as the best baseline for inferring consumption for the Stellenbosch region based on available research (personal communication, N Steyn, 24 August 2009). The report describes the average typical food consumption for the entire nation by age group and is not disaggregated by income or ethnicity.

The food consumption by food product and group was weighted by age group based on the findings of the Community Survey 2007 (StatsSA, 2007), which was the most recent population survey for Stellenbosch (Zietsman, 2009). The research can therefore be seen as an estimation of what Stellenbosch might consume based on national trends using the only data available at the time of the research. The population demographics of Stellenbosch and the Western Cape are very different from the rest of the country, so this population's typical diet is likely to differ from that of the rest of the country because of different cultural preferences and consumption patterns. Further empirical data were sourced to help understand the implications of realising nutritionally optimal consumption (i.e. what residents should be consuming for a healthy diet) so that recommendations could be made for localising the food system.

A comparison of ideal versus actual research methodologies for investigating production, distribution and consumption are shown in Table 2.

Table 2: Research ideal and actual methodology, and outcomes and opportunities

Empirical data collection	Ideal researchmethodology	Actual researchmethodology	Researchoutcomes	Researchopportunities
Production	GIS map with ERF specific values for both productive capacity and actual output.	Data used from the Commercial agriculture census data (2002), Emerging farmer census data (2006/2009), findings of Cape Winelands Area Based Plan on productive capacity (2006) and the Specialist Soil and Land Study for the Cape Winelands (2007).	Estimated production for Stellenbosch per product in tonnes. Potential productive capacity for the Stellenbosch region by current land use and broad productive potential.	Detailed surveys and mapping of both actual and potential productivity.
Distribution	Detailed surveys of where farmers sell their food to.	No data available from farmers on where they sell their food to. General statistics on exports for certain commodities (Louw, 2009; SAWIS, 2009).	Broad overview of distribution patterns for parts of the distribution chains through inferences and estimations.	Detailed surveys of where farmers sell their food to.
	Accurate systems from local retailers to show where they source their food.	Estimates from retailers on percentage of food locally sourced.		Systems for retailers to show where they source their food.
	Detailed surveys of where people purchase their food.	National estimates of typical purchasing habits (BFAP, 2009).		Detailed surveys of where people purchase their food.
Consumption	Representative profile of typical consumption patterns across various demographic groups in Stellenbosch. Locally suitable nutritionally optimal diet across age groups.	National food consumption report (2002) compared with multiple Food Pyramid studies (2009).	Estimated consumption patterns for local residents. Nutritionally optimal diet inferred from multiple studies.	Detailed surveys on typical consumption patterns for multiple demographic profiles of Stellenbosch. Locally suitable nutritionally optimal diet across age groups.

An overview of production describes current production by both land use and volume. Because of the difficulties the authors encountered in accessing distribution data, the distribution component of the Stellenbosch food system was described through a series of examples to highlight current practice and complexities. Consumption patterns were inferred for the Stellenbosch population. A summary of all the findings is presented in Table 3.

Table 3: Summary of research findings

	Current production					Current consumption			Nutritionally optimal consumption
Food group	Commercial (kg/year)	Emerging (kg/year)	Total(kg/year)	Total (tonnes/year)	Area (ha)	Total(kg/year)	Total (tonnes/year)	Required (ha)	Total(kg/year)	Total (tonnes/year)	Required (ha)
Cereals	0	0	0	0	0	53 411 519	53 412	5 341	44 218 159	44 218	4 422
Sugar	0	0	0	0	0	7 334 221	7 334	n/a	0	0	0
Stimulants	43 000	0	43 000	43	361	216 441	216	n/a	0	0	0
Milk	6 486 298	0	6 486 298	6 486	504	7 550 860	7 551	8	25 917 692	25 918	2 013
Meat	2 158 164	0	2 158 164	2 158	577	6 751 598	6 752	10	2 434 411	2 434	4
Vegetables	5 211 000	1 051 625	6 262 625	6 263	424	6 264 002	6 264	313	32 901 107	32 901	1 645
Vegetable oils	0	0	0	0	0	695 780	696	696	3 836 074	3 836	3 836
Roots	0	2 750	2 750	3	1	3 739 778	3 740	n/a	0	0	0
Fruit	19 233 000	6 000	19 239 000	19 239	2 143	5 663 066	5 663	126	20 709 158	20 709	460
Wine grapes	86 284 000	0	86 284 000	86 284	8 924	0	0	0	0	0	0
Eggs	2 418 424	0	2 418 424	2 418	0	1 089 214	1 089	0	2 498 400	2 498	0
Pulses	0	0	0	0	0	2 391 953	2 392	199	3 836 074	3 836	320
Nuts	25 000	0	25 000	25	2	134 157	134	27	3 653 840	3 654	731
Fish	55 000	0	55 000	55	0	917 552	918	n/a	3 981 990	3 982	n/a
Human milk and BMS	0	0	0	0	0	163 641	164	n/a	n/a	n/a	n/a
Soups	0	0	0	0	0	301 805	302	n/a	n/a	n/a	n/a
Infant foods	0	0	0	0	0	18 977	19	n/a	n/a	n/a	n/a
Condiments	0	0	0	0	0	41 798	42	n/a	n/a	n/a	n/a
Animal oils and fats	0	0	0	0	0	75 977	76	n/a	n/a	n/a	n/a
Spices	0	0	0	0	0	116	0	n/a	n/a	n/a	n/a
Supplements	0	0	0	0	0	7 489	7	n/a	n/a	n/a	n/a
Alcohol	0	0	0	0	0	3 738 518	3 739	n/a	n/a	n/a	n/a
Total	35 629 886	1 060 375	36 690 261	122 974	12 935	100 508 462	100 508	6 720	143 986 907	143 987	13 430
Notes: BMS = breast milk substitute; n/a = not available.

4.1 Production

Wine and stone fruit farming are the largest agricultural activities (both by land use and rand value) with vegetable and essential oil production on a smaller scale. Deciduous fruits (including wine grapes) contribute 87.5% to gross farm income, vegetables 9.9% and other horticultural products 2.6% (StatsSA, 2006). Wine grapes use the largest percentage of land (71.5%), followed by peaches (9.6%) (StatsSA, 2006). Both the wine grape and fruit markets are export oriented (Louw, 2009:40). Commercial vegetable production (5211 tonnes per annum) is predominantly cabbages, tomatoes, onions and green beans (StatsSA, 2006). Emerging farmers produce these, as well as strawberries, butternut, carrots, spinach, celery, green peppers and other crops to a total of 1051 tonnes per annum. Smaller quantities of tea (43 tonnes) and nuts (25 tonnes) are also produced. The vegetable market is almost entirely nationally oriented (Louw, 2009:50). The region is not considered suitable for sugar cane production and cereals (with the exception of some winter grains) and root crops are difficult to grow in the local climate. Formal livestock farming is relatively small, producing only 2158 tonnes per annum (over half being chickens) and using only 577 ha of land (most for cattle) (personal communication, L Hoffman, Head of Animal Science Department, Stellenbosch University, 25 August 2009; and M Prinsloo, Statistician, South African Poultry Association, 12 May 2009).

4.2 Distribution

Tracking the flows of food in and out of Stellenbosch proved difficult, as much of the information is either not recorded or not openly available. The findings below were selected in order to give an overview of the complexity of distribution rather than a detailed account of flows.

Most locally grown produce is for export markets (Louw, 2009:40). The export value of trade and investment in the Cape Winelands District Municipality has increased by over 400% over the past decade (Louw, 2009:8). Export figures specific to Stellenbosch were not available, but national exports of wine and fruit were 53.9% and 74% of total production (Louw, 2009:50). These export estimates would also suggest that 74 220 tonnes or 60.35% of Stellenbosch's produce is being exported out of the country (StatsSA, 2006). Because of the diversified nature of the system and lack of access to sufficient data, it was not possible to estimate what remains in Stellenbosch or what is distributed across South Africa.

Vegetables, meat and dairy produced locally were found to be distributed within the Western Cape or nationally. Since information was not publicly available for the meat and dairy sector, it was not possible to track the flow of livestock products. For vegetables, only 4% of South Africa's total vegetable production is exported and the majority of the vegetables produced in the Western Cape are consumed locally or nationally (Louw, 2009:50). A large percentage of vegetables is distributed through the national fresh produce markets (estimated at 45 to 50% of total fresh produce sales) and the Cape Town Fresh Produce Market (CTFPM) is the nearest national market for Stellenbosch, where most of its fresh vegetables are both sent to and purchased from (personal communication, E Becker, 31 August 2009). It is estimated that the CTFPM trades approximately 150 million tonnes of vegetables per annum (Louw, 2009:50). Vegetables not sold to this market are sold directly to retailers or informal traders, but there are no records available for these sales.

Interviews conducted with emerging farmers found that they sell their produce through a variety of markets, including directly to local vendors, pack sheds, retailers or central distribution markets such as CTFPM. While it is often difficult for the farmers to transport their produce through to Cape Town and they do not always receive a good price, they are at least guaranteed to sell it at the market.

In South Africa, four of the largest retailers (Pick n Pay, Woolworths, Shoprite Checkers and Spar) account for over 60% of national sales (Van Rooyen, 2009). Almost all major retailers source their produce from central packing warehouses that source either directly from the farmers or from local packsheds. Spar and Pick n Pay Family Stores allow store managers to source their produce directly from local farmers and most stores stock a combination of produce directly sourced from local farmers and produce sourced from the central distribution warehouse.

Local vendors and other informal trading markets account for most of the remaining sales. Informal vendors sell loose fresh produce and small amounts of basic food produce which they source mainly from central distribution markets such as CTFPM or Kraaifontein, where prices are lowest (personal communication, N Lebo, Spaza shop owner, Stellenbosch, 27 August 2009). A very small percentage of food is grown by the community itself for home consumption but this still plays a critical role in supporting food security at household level for families engaged in these food gardens (personal communication, E Mbambalale, food garden participant, Stellenbosch, 2009).

It is estimated that most of the food consumed in Stellenbosch originates from other areas. Cereals and sugars are entirely sourced from other regions. Vegetables, fruits and livestock products may originate either from Stellenbosch or from outside, but as Stellenbosch consumes more than it is currently producing, it can be inferred that most of its food is brought into the region.

Stellenbosch-grown produce is often transported out of the region to central distribution points before being sent back to the large retail outlets or to informal traders, the aim being to achieve the economies of scale and diversity required to meet consumer demands for a wide variety of affordable food year-round (personal communication, W Coetzee, Manager, Fruit n Veg City, Stellenbosch, 2 September 2009). The findings suggest the distribution system is determined by price and efficiency for the large retailers as an overriding decider in all situations.

4.3 Consumption

Table 4 shows the main food groups consumed by weight for the Stellenbosch community. These findings are limited in that they are not based on primary research, but derived from a compilation of national studies that reflect national average consumption patterns. Given that the demographic profile for Stellenbosch differs greatly from the rest of the country, these findings may not reflect actual patterns. Furthermore, the study does not allow for variations in income group or cultural preferences.

Table 4: Current consumption composition by weight

Food category	Percentage of total diet by weight
Cereals	53
Milk	8
Meat	7
Sugar	7
Vegetables	6
Fruits	6
Roots	4
Alcohol	4

Van Niekerk (personal communication, 30 October 2009) described high levels of food insecurity in certain Stellenbosch districts and this motivated our investigation into the food requirements for a nutritionally optimal diet for the Stellenbosch population, given the critical importance of food security in realising sustainable development (Khoza et al., 2009). The estimates in our study used the same food groups that are used to estimate current total production and consumption, and only food groups that contribute to nutritional security and can be satisfied through primary production were considered (i.e. omitting alcohol, sugars, supplements and condiments).

The main findings from comparing current consumption with nutritionally optimal consumption (see Figure 1) suggest that there would need to be large shifts in both the quantity and the types of food consumed. The research suggests that the Stellenbosch population needs to consume smaller quantities of cereals and meat and considerably larger quantities of vegetables, milk, fruit, pulses, fish, nuts, vegetable oils and eggs. In other words, people need to change the composition of their diet, reducing the proportion that is made up of meat and cereals and increasing the proportion that is made up of vegetables, fruit, etc. Overall, more food needs to be consumed, i.e. people are on average consuming too little. This has important implications for food security strategies, which should be linked to increasing the local production of key nutritional crops such as vegetables.

Figure 1: Current vs nutritionally optimal annual consumption (kg) by food group for Stellenbosch

4.4 Current issues and future vulnerability of the food system

Table 5 highlights existing problems with Stellenbosch's food system and shows its vulnerability to the polycrisis.

Table 5: Weaknesses and vulnerabilities of Stellenbosch's current food system

Weakness	Description
Production
Focus on export markets	Export markets will become increasingly less attractive and competitive as oil prices rise. While many point out the foreign exchange and job creation benefits of export markets, questions should be asked about how fairly the benefits are shared and about the security and quality of jobs for lower income members of the community.
Large-scale commercial agriculture dominates	Land holdings (and equity) in Stellenbosch have not changed much since 1994. The poorer members of the community do not have access to land from which to earn a living, and high mechanisation levels on such farms means fewer jobs are available in the agricultural sector.
Proliferation of conventional farming methods	These methods make heavy use of fertilisers and pesticides derived from oil, which will make this method of farming increasingly less competitive as oil prices rise. They have also been shown to have negative environmental impacts (IAASTD, 2008).
Distribution
Long value chains	Food travels long distances to and from Stellenbosch, which will no longer be feasible under peak oil conditions.
Dominance of national retailers	These supermarkets export profits out of Stellenbosch, reducing any multiplier effects of food spending. They also tend to exclude small-scale farmers who cannot meet their requirements.
Consumption
High levels of food insecurity and poor food choices	Many Stellenbosch residents are under-nourished or malnourished because of poverty and poor food choices. This hampers their development and that of the broader community.

5. Conclusions and recommendations for resilience strategies

The key focus areas for reforms of Stellenbosch's food system are reducing food insecurity and building resilience to future shocks. We propose that increasing localisation can achieve both of these aims. This would include not only increased local production of food (dairy, poultry, livestock, vegetables and fruit) using sustainable methods, but also localised food processing, distribution and retail.

Despite the problems in the local production system and tough competition from both local and international producers, there are opportunities for promoting a stronger local food system by increasing local food production and connecting consumers more directly to producers. The Stellenbosch Municipality owns large portions of land in Stellenbosch, held as municipal commonage, that could be made available to emerging farmers for agricultural production on long-term leases. There is significant potential for increasing local food production through such a land reform programme but it would require significant institutional support to overcome the multiple barriers currently being experienced by emerging farmers.

An important issue raised by comparing current production with consumption (both current and nutritionally optimal) is the appropriateness of growing wine grapes for export in the context of the high levels of food insecurity in the region and the dependence on imported produce (from neighbouring regions) and rising fuel costs. Furthermore, prices received by wine producers have declined or remained constant since 2003 because of increasing international competition which, when coupled with increasing costs of production and packaging, is reducing farm viability (Louw, 2009). These trends threaten both employment and the wider communities that depend on the wine industry for livelihoods. The impact of climate change on producing quality wines is a further concern (Louw, 2009).

This is not an argument for replacing all commercial wine estates with emerging food farmers but rather a conclusion that further investigation is warranted into appropriate land use for the region. Non-viable wine estates should be reconsidered for other agricultural purposes. This does not necessarily mean vegetable production, given the soils, climate and availability of water (vegetable production requires up to five times more water than wine grape production in Stellenbosch). The impact on secondary industries (including the tourism sector supported by wine production) should also be considered. Much of the land in Stellenbosch currently under vineyard is suitable for deciduous fruit, olive, nut or even livestock production that could contribute to local food security where the local demand will always remain high. It is rather put forward that food production by emerging farmers should be strongly considered as part of a contribution to local food and livelihood security. The suitability of local soils and climate in conjunction with the needs of the local community for food and livelihood security should be considered when assessing best land use in Stellenbosch.

Alternative local food economy distribution networks such as community supported agriculture initiatives,3 local markets and consumer cooperatives should be developed. These shorten value chains, thus giving farmers a greater share of the retail price. Another suggestion is a local distribution hub for locally produced foods to directly connect farmers and consumers and maximise the benefits of central coordination. By connecting local consumers more directly to the food economies that support them and the individuals behind that food economy, closer bonds can be forged within a community around local food.

The research revealed that areas within Stellenbosch are experiencing food insecurity and that a shift towards a nutritionally optimal diet would mean reduced consumption of cereals and meats, and increased consumption of vegetables, milk and fruit (Schulschenk, 2010). While Stellenbosch residents are not expected to make a complete shift (given the barriers, such as cultural preferences, inaccessibility and unaffordability), there is potential for targeted programmes to increase the volumes of nutritious fresh produce being consumed. Food education should be a priority for the community, especially about affordable nutritious home cooking.

Urban agriculture could contribute to a large increase in the consumption of fresh produce at affordable prices for local residents. An extensive knowledge base, supported by major learning institutions, is already established in Stellenbosch. One aspect that would need to be considered is finding suitable water sources for urban agriculture (e.g. harvested rainwater or grey water) to ensure that scarce potable water is not used. The local agricultural system already supports a strong secondary economy of agricultural businesses and there is potential for strengthening the local food economy by building on the region's resources and reputation.

Food security programmes, such as urban agriculture and feeding schemes, play a critical role in promoting local food security across communities with comparatively small inputs and should be supported further. Experiences with local food initiatives show they can be drivers of greater awareness and change around social and environmental issues. Community connections and building of social capital play a critical role in overcoming the barriers of economies of scale and global competition associated with the modern food system.

In the context of peaking oil production, climate change and population growth, Stellenbosch will have to prioritise local food production to ensure resilience against future shocks. The transition from fossil fuel dependency to post-oil communities will carry significant social and environmental costs if preparations are not made now for promoting resilient local communities. Such preparations will require small but strategic investments in the short term to avoid devastating human (as well as financial) cost in the long term. In the light of both the impacts of the modern food system and the benefits of local food economies, Stellenbosch could become more sustainable, equitable and resilient through building a stronger local food economy.

To aid this development, further research is needed on food consumption trends in the Stellenbosch region, and in particular extensive primary research is needed to establish the food distribution patterns.

Notes

¹When asked for detail at the level of Stellenbosch, StatsSA said they interviewed only 39 Stellenbosch households for this study and could not give statistically significant data below the provincial level (personal communication, I Schmidt, Executive Manager: Social Analysis, Statistics South Africa, Pretoria, 27 January 2011).

²This is a scheme in which consumers receive a box of organic produce weekly, either from one farm or many, either pre-paid or paid on delivery, and either delivered direct to their home or collected from a pick-up point.

³CSA initiatives aim to ensure a better and fairer price for the farmer, and fresh, affordably-priced organic produce for the consumer. Members pay a fee upfront or buy a share in a harvest at the beginning of the season, which then provides the farmers with the necessary input capital to grow their food. In return, members receive a weekly delivery of produce.