Links between the Local Trade in Natural Products, Livelihoods and Poverty Alleviation in a Semi-arid Region of South Africa

World Development Vol. 36, No. 3, pp. 505–526, 2008 Ó 2007 Elsevier Ltd. All rights reserved 0305-750X/$ - see front matter www.elsevier.com/locate/worlddev

doi:10.1016/j.worlddev.2007.03.003

Links between the Local Trade in Natural Products, Livelihoods and Poverty Alleviation in a Semi-arid Region of South Africa SHEONA SHACKLETON Rhodes University, Grahamstown, South Africa BRUCE CAMPBELL Charles Darwin University, Darwin, Australia Centre for International Forestry Research, Bogor, Indonesia HEILA LOTZ-SISITKA and CHARLIE SHACKLETON Rhodes University, Grahamstown, South Africa

*

Summary. — Can the local commercialization of natural products contribute to reduced poverty and vulnerability? Commentary on this issue is mixed, with some observers being quite optimistic, while others hold a counterview. This paper explores the poverty alleviation potential of four products traded in Bushbuckridge, South Africa—traditional brooms, reed mats, woodcraft, and ‘‘marula’’ beer. While key in enhancing the livelihood security of the poorest households, these products were unlikely to provide a route out of poverty for most, although there were exceptions. Incomes often surpassed local wage rates, and some producers obtained returns equivalent to the minimum wage. Non-financial benefits such as the opportunity to work from home were highly rated, and the trade was found to represent a range of livelihood strategies both within and across products. Ó 2007 Elsevier Ltd. All rights reserved. Key words — sub-Saharan Africa, South Africa, natural products, local trade, livelihoods, poverty, commercialization, micro-enterprises

1. INTRODUCTION Poverty alleviation has been placed high on the international development agenda following the adoption of the United Nations Millennium Development Goals (MDGs) in 2000. The first and foremost of these goals is to halve the proportion of people living on less than US$1 per day, and those suffering from hunger, by 2015 (UN Millennium Declaration, 2000). The obligation to address the MDGs, together with the high levels of spatial concurrence between regions rich in biodiversity and the majority of the world’s rural poor (WRI, 2000), has compelled scholars and practitioners operating at the environment–development interface to seek 505

* We extend our sincere appreciation to all producers and traders for sharing their work and insights with us. Sibongile Ndlovu and Sam Moropane provided valuable assistance with fieldwork. For help with statistical analysis, we thank Ramadhani Achdiawan (CIFOR) and Sarah Radloff (Rhodes University). Financial support of this study was provided by BP South Africa, National Research Foundation (NRF), the South African–Netherlands Partnership in Development Alternatives (SANPAD), the Forest Research Programme (FRP) of the UK Department for International Development (DFID), and the Centre for International Forestry Research (CIFOR). All views expressed are those of the authors. We are grateful to Freerk Wiersum, Stephen Turner, and four anonymous reviewers for comments on earlier drafts of this paper. Final revision accepted: March 21, 2007.

506

WORLD DEVELOPMENT

solutions to poverty that include natural resource-based activities (Adams et al., 2004; Bird & Dickson, 2005; FAO, 2003; Oksanen & Mersmann, 2003; Roe, 2004; World Bank, 2002; WRI, 2005). The commercialization of natural products is seen as an important potential vehicle for achieving this (Scherr, White, & Kaimowitz, 2004). Priority has thus been placed on understanding more fully the contribution biodiversity and the natural product trade can make to achieving the goals of poverty alleviation, with examination of the links between the use and trade in natural products, livelihood security, and environmental sustainability (another of the MDGs) forming a focal area for research attention (Roe, 2004; WRI, 2005). Numerous studies worldwide have left little doubt that biodiversity and natural products contribute to the well-being and, sometimes, the very survival of millions of poor rural households (Arnold, 2002a; Belcher, RuizPe´rez, & Achdiawan, 2005; Fisher, 2004; Godoy et al., 2000; Narendran et al., 2001; Scherr et al., 2004). These products may be harvested for subsistence purposes and/or as commodities that can be offered for sale in the market in raw or processed form. The use and sale of products may take place on a regular basis, seasonally as a gap filler, or only in times of emergency providing an important fallback option or safety net (McSweeny, 2004; Takasaki, Barham, & Coomes, 2004). Natural product markets have been shown to be significant in assisting rural households to realize some, if not all, of their cash requirements (Arnold & Townson, 1998; Marshall & Newton, 2003; Marshall, Schreckenberg, & Newton, 2006; Narendran et al., 2001; Ndoye, Ruiz-Pe´rez, & Eyebe, 1997), and are particularly crucial for the most marginalized and vulnerable segments of society (Beck & Nesmith, 2001; Cavendish, 2000; Reddy & Chakravarty, 1999). The processing and sale of natural products certainly offers a low barrier to entry and widely available option to generate cash income that is progressively being taken up by rural dwellers in many countries, including South Africa, as a means to cope with economic hardship. Indeed, in some situations, such as in the dry woodlands of southern Africa, the trade in natural products may be one of the few accessible local income generating options available to the rural poor, and women in particular (Campbell et al., 2002; Shackleton & Shackleton, 2004). While this ‘‘picture’’ of the significance of the natural product trade for livelihoods exists, the

situation in reality is complex and variable, and limited empirical data from across a range of regions, vegetation types, and socio-economic contexts are available to assess the ability of these products to create lasting opportunities for local livelihood enhancement (FAO, 2003; Sunderlin et al., 2005). Indeed, the refocusing of the development agenda on poverty has led to recent reassessment of the role that biodiversity plays in livelihoods and poverty alleviation. A profusion of new commentary has emerged. This poses many fresh questions, and, to some extent, tempers previous optimism regarding the ability of this sector to make a difference by providing a more subtle and complex picture of livelihood–biodiversity linkages (e.g., Arnold, 2002a; Belcher, 2005; Koziell, 2001; Lawrence, 2003; Ros-Tonen & Wiersum, 2005; Scherr et al., 2004; Wunder, 2001). Central to these new enquiries is a more perceptive and nuanced appreciation of (a) what is meant by poverty alleviation in relation to natural resources, distinguishing between poverty prevention or mitigation and poverty elimination; (b) the links between natural resource dependence and the potential of the natural product trade to provide pathways out of poverty; (c) the safety net role of the natural product trade and when and how this might translate into a ‘‘poverty trap;’’ and (d) the extent to which opportunities associated with natural product production and sale can be made more propoor and thus contribute to the efforts to combat poverty and vulnerability (FAO, 2003; Koziell, 2001). A key area of debate is whether the trade in natural products can assist in improving livelihoods and income, or alternatively, whether it offers limited options serving only as a last resort, possibly contributing to persistent poverty (Belcher, 2005; Ros-Tonen & Wiersum, 2005; Wunder, 2001). In this paper we explore these issues by considering the contribution of the widespread, but often neglected and relatively invisible, local trade in natural products for household livelihood security in a semi-arid savanna region of South Africa. Local and regional markets generally account for the bulk of natural products sales and often include the greatest numbers of participants, yet are frequently poorly acknowledged and under-researched (Arnold, 2002b; Taylor, 1999). Most literature on natural product commercialization focuses on international value chains for charismatic products with high levels of external intervention (Taylor, 1999). In addition, few case comparisons to date (e.g.,

LINKS BETWEEN THE LOCAL TRADE IN NATURAL PRODUCTS

Belcher et al., 2005; Marshall et al., 2006) have sought to systematically investigate the livelihood and poverty outcomes from the commercialization of a range of products produced and marketed within the same geographic setting. Previous comparative work has revealed that socio-economic context, in particular the nature of property rights, degree of underdevelopment, accessibility of markets, and the opportunity costs of labor and land, is often critical in influencing the benefits participants obtain from the natural product trade as well as its potential to contribute to poverty alleviation (Belcher, 2005; Belcher et al., 2005). In this study, by focussing on a single region, it was possible to delve more deeply into other factors, including participant’s own motivations, which may play a part in determining the differential role and outcomes of the trade both across and within specific product types. Furthermore, being situated in a semi-arid region, this study also provides case study evidence from a contextual setting that is notably different to the humid tropics upon which most analysis to date is based (Sunderlin et al., 2005; Wunder, 2001). 2. STUDY AREA AND CONTEXT The area selected for study was the Bushbuckridge local municipality, located in the northeast of South Africa in Limpopo Province (30° 30 0 —31° 35 0 E; 24° 30 0 —25° 0 0 S). It comprises an area of 2,417 km2 of which 85% is communal rangeland. The vegetation is typical semi-arid savanna, with rainfall varying from 1,000 mm in the west to 500 mm in the east. Inter- and intra-seasonal drought is frequent, occurring every 3–4 years, limiting dry land arable production. Like all former homeland 1 areas, infrastructure and services are poor, and the area is densely populated with between 150–300 people per km2 and a total population of approximately 550,000 people dispersed across 65 settlements. Formal employment opportunities are limited and much of the population depends on state welfare grants, in particular old age pensions, and migrant remittances for cash income. The majority of households live below the poverty line, with some 40% reporting no cash income whatsoever (Stats SA, 2004). This socio-economic context is consistent with that for much of rural, communal South Africa, and is further exacerbated by growing unemployment, high rates of

507

job loss, high costs of living, a lack of arable land, and increasing vulnerability and poverty arising from the HIV/AIDS epidemic (Aliber, 2003; Bryceson, 2002; Meth & Dias, 2004; Nattrass, 2004). As a result, people are increasingly turning to forms of self-employment to make ends meet, with the natural product trade providing one such opportunity (Kepe, 2002; Marcus, 2000). Data from numerous studies across the savanna areas of the country indicate that some 3–14% of rural households trade in at least one natural product on a regular or ad hoc basis (Shackleton, 2005). Presently about 2% of all households in Bushbuckridge (out of 112,000 in total) trade in at least one of the products chosen for this study (Shackleton, 2005; Stats SA, 2004). 3. APPROACH AND KEY QUESTIONS Detailed case studies of four locally traded natural products, namely woodcraft (furniture and carvings), reed mats, traditional brooms (‘‘indoor’’ grass and ‘‘outdoor’’ twig brooms), and a traditional beer made from the fruits of the marula tree, Sclerocarya birrea subsp. caffra were undertaken (Table 1). For each case study, data were collected on the individual and household profiles of producers and traders, 2 the history and motives for involvement in the trade, the annual returns derived from the trade relative to other sources of cash income, and any non-financial livelihood benefits obtained (Shackleton, 2005). Both quantitative (e.g., repeated semi-structured interviews) and qualitative (e.g., life histories, participatory exercises, ranking) methods were employed. Revenues from product sales were determined in slightly different ways for each product (Shackleton, 2005), but were calculated based on either the amounts of raw material harvested or the quantity of product produced and sold within an appropriate and easily recalled time period, the duration of the selling season, local price and all direct costs incurred during harvesting, processing, and marketing. For seasonal products all fieldwork was conducted during the production season to minimize recall problems. Total household cash income, a notoriously difficult variable to quantify due to its private nature and the need to rely on interviewees’ recall and honesty, was determined through careful questioning around the range of income sources households drew on. Respondents were asked whether their

508

WORLD DEVELOPMENT

Table 1. Key characteristics of the trade in the four case study products (the sample sizes for each case study are shown in parentheses) Case study/product

Primary species used

Key features

Reference

Reed mats (n = 114)

Cyperus sexangularis Schoenoplectus corymbosus

Seasonal harvesting but finished mats stored for sale year round Intermediate value addition Low barriers to entry Schoenoplectus is harvested from communal wetlands and Cyperus from irrigation canals on nearby private farm land—few access problems Resource supply secure and sustainable Sold in local markets—mainly from home Simple market chain—same people harvest, weave and sell to end consumers

Shackleton (2005)

Traditional grass and twig brooms (n = 70—producers, 40—traders)

Athrixia phylicoides (twigs) Festuca costata (grass)

Seasonal—available for 7–8 months Low value addition Intermediate to low entry barriers Harvested mainly from distant higher altitude wooded grasslands on state and private land—access insecure Resource supply secure and sustainable Two step market chain—producers harvest and process and sell to intermediate traders who then sell to end consumers from informal market stalls along with other products (these traders trade within Bushbuckridge)

Shackleton (2005)

Sclerocarya birrea subsp. caffra

Highly seasonal—2–3 months in mid-summer Very low entry barriers Intermediate value addition Resource abundant and locally available in communal and farm lands Sold in local markets in the towns in Bushbuckridge Simple market chain—same people harvest, process and sell to end consumers

Shackleton (2004, 2005)

Pterocarpus angolensis Dalbergia melanoxylon Spirostachys africana Berchemia discolor

High value addition Full-time and permanent activity Intermediate barriers to entry Raw material use unsustainable with scarcity being a major constraint Harvested mainly from communal lands—but the majority are protected species so access is controlled by laws Carvings sold in regional tourist markets through retailers and informal vendors Furniture sold directly to end consumers

Shackleton and Steenkamp (2004) and Shackleton (2005)

Marula beer (n = 51)

Woodcraft— hardwood-based crafts and furniture (n = 40 and 16)

households received pensions, child grants, or other state grants (all fixed amounts per month), and whether they had formally- or self-employed household members. The occu-

pations of these members were then noted, as well as whether they contributed all or some of their earnings to the household. The latter provided a means to crosscheck and probe the

LINKS BETWEEN THE LOCAL TRADE IN NATURAL PRODUCTS

actual figures offered, and to deduce contributions when respondents were unwilling to share these. For this synthesis, comparative and exploratory analyses were carried out using the case study data as well as data from a random household survey of 142 households conducted in the study area in 2001/2002 (Shackleton & Shackleton, 2005). Particular attention was paid to capturing and highlighting the inherent variation within case studies as well as making comparisons across them. The following questions, relating to some of the fundamental issues identified within the biodiversity-poverty discourse, were used to guide the analysis and, in turn, provide the structure for this paper: (a) What is the trend in the natural product trade, in particular its importance for rural people as a safety net under increasing hardship? (b) Who trades in natural products and why? (c) In what way does the trade contribute to household cash income and other livelihood assets? (d) How does the trade fit into the broader household livelihood portfolio? (e) What conclusions can be drawn regarding the scope of the natural product trade to contribute to poverty alleviation? 4. RESULTS AND DISCUSSION (a) What trends are evident in the natural product trade? The numbers of participants in the commercial trade in three of the case study products have expanded in recent years (Table 2). Only woodcraft showed little growth, primarily due to shortages of raw material, although 57% of producers mentioned that there were also new entrants to this trade. For the broom and mat trades, about one third of participants joined the trade during 1998–2003 (the year the data

509

were collected), and up to half during 1993– 2003. The marula beer trade, in particular, grew from nothing in 1998 to over 250 sellers in 2002 (Shackleton, 2004). These findings corresponded with participants’ own perceptions of changes in the trade, with the growth leading to a situation of oversupply for some products (Shackleton, 2005). Producers and traders ascribed the increase in numbers of participants to a lack of employment opportunities, widespread retrenchment, HIV/AIDS, and rising costs of living. This was supported by national level statistics, which demonstrated that, at the time of the study, unemployment rates in South Africa were rising, as was the level of poverty (Meth & Dias, 2004; Nattrass, 2004). Moreover, Bushbuckridge was particularly hard hit by the downsizing of the mining industry. Socio-economic hardship was thus pivotal in driving entry into the trade, highlighting the critical safety net function offered by this activity (Table 3). The majority of women producers (e.g., for mats, brooms, and marula beer) took up trading in response to a shock in the household, such as the death of a breadwinner, illness or retrenchment within the household (usually of their husbands), or because they were ‘‘suffering’’ and ‘‘hungry’’ and unable to find work. Among male woodworkers, 93% commenced woodworking after they had been retrenched from a formal job, usually as a migrant worker in the mining, manufacturing, or agricultural sectors. Of particular interest was that 10% of broom and mat producers were pushed into selling natural products after they had been left with grandchildren to care for following the death of the children’s parents, presumably from AIDS. However, not all reasons for entering the trade were ‘‘supply-push;’’ there was also an element of choice in joining the trade for some. Several women mentioned that they preferred to work from home as they had

Table 2. Duration of participation in the natural product trade at the time of the study Years tradinga <1–5 6–10 11–15 16–20 21–30 >30 a

Year started

Mat producers (%)

Broom producers (%)

Broom traders (%)

Marula beer sellers (%)

Woodworkers (%)

1998–2003 1993–1997 1988–1992 1983–1987 1973–1982 Prior to 1973

30 9 10 22 19 10

38 11 21 13 13 6

30 20 13 10 18 10

100 0 0 0 0 0

Not measured 0 13 12 18 58

2003 was used as the reference year.

510

WORLD DEVELOPMENT

Table 3. Producers’ and traders’ reasons for entering the natural product trade (responses categorized from answers to an open-ended question and rounded up) Reasons for entering the trade Observed others selling the product and needed income Suffering and hunger For income as no jobs To help other earners Retrenchment or job resignation Loss of husband’s incomea Otherb

Mat producers (%)

Broom producers (%)

Broom traders (%)

Marula beer sellers (%)

Woodworkers (%)

10

–

–

53

–

22 6 19 4 30 9

24 19 – – 47 10

22 5 – 16 35 22

14 24 6 4 – –

– – – 93 – 7

a

From either death, or illness or retrenchment. This includes caring for grandchildren, parents died, have baby so work at home, people asked to buy, part of community project, family business. b

young children or ill or elderly family members to care for, or that they themselves were too old or unwell to undertake a full-time job. Woodworkers reported how they preferred to be with their families than to work as migrant laborers far from home. Others described how the trade offered better returns than the other employment available, such as local domestic or farm work. The need to support other income earners and augment meager household incomes was also revealed as grounds for participating in the trade. The growth trend apparent in Bushbuckridge is echoed across South Africa and other regions, where there is evidence that increasing numbers of rural households are turning to natural product trading for income generation under a deteriorating socio-economic climate (e.g., Gyan & Shackleton, 2005; Kepe, 2002; Lowore, 2003; Monela, Kajembe, Kaoneka, & Kowera, 1999). For example, in Zimbabwe, Campbell et al. (2002) found increased numbers of households trading in woodland products in response to a decline in agricultural subsidies and a rise in the incidence of HIV/AIDS. Mead and Liedholm (1998) presented evidence from eight countries to show that under difficult economic times there is increased pressure for people to start new, generally single individual, businesses. In a review of the role of forests in livelihoods, Scherr et al. (2004) reached similar conclusions for regions where population is growing faster than per capita income. Certainly, in these situations, the natural product trade offers an opportunity for self-employment where capital costs are minimal, raw material is relatively freely available, local markets exist, and many people already have the skills required to harvest and process the product.

(b) Who is involved in the natural product trade? A comparison of key variables relating to the profile of producers and traders and their households with the random sample revealed that, on the whole, producer and trader households tended to be significantly worse off and more vulnerable than their counterparts (Table 4). For example, the proportion of female-headed households among mat, broom and marula beer producers and traders was significantly higher than the one third recorded in the random sample and for South Africa as a whole (Table 4; Posel, 2001). Female-headed households are commonly considered to be the ‘‘poorest of the poor.’’ While this has been questioned and qualified elsewhere (Campbell et al., 2002; Chant, 1997), evidence from South Africa shows that these households are indeed a particularly vulnerable group, with their poverty rate, at 60%, being double that of maleheaded households (Gelb, 2003). All producers and traders were significantly less educated than the average adult in the random population, except for marula beer sellers (Table 4). This reflects the age patterns across products, with the latter being, on average, younger than other producers and consequently better educated. Mat producers and woodworkers were significantly older than the other groups. Poor levels of education and high illiteracy among producers, along with their age, were often put forward as reasons precluding them competing successfully in the formal job market. The younger average age of marula beer sellers may reflect (a) the seasonal nature of this activity, with the younger generation using it as a stop-gap while they seek more permanent employment and (b) the fact that some scholars

LINKS BETWEEN THE LOCAL TRADE IN NATURAL PRODUCTS

511

Table 4. Comparison of producers’ and traders’ profiles and household (hh) characteristics across products and with a random household sample Mat Broom Broom Marula Woodworkers Random producers producers traders beer household sellers sample

Significance level for v2 or Kruskal– Wallis testA

% Female % Female-headed hh

100 53

98 47

95 60

100 45

0 0 (excluded from test)

– 35

– v2 = 44.1, df = 4, p < 0.0001

No. of years of education Mean Median SE

2.1e 0 0.3

3.6c 3 0.4

5.6cd 6.5 0.7

7.0ab 8 0.6

5.9bd 5 0.5

8.3a 7.9 0.4

H or v2 = 139.5 df = 5, p < 0.0001

Age Mean Median SE

57b 58 1.0

50c 50 1.1

47c 49 2.0

39d 39 1.8

63a 65 1.8

–

H or v2 = 91.8 df = 4, p < 0.0001

% of hh with pensions

36

21

30

22

44

35

% of hh with one or more formal jobs

24

7

10

22

2

35

% of hh with no regular cash income sources

12

50

48

45

50

15

% producers and traders selling other products

58

40

100

73

13

–

v2 = 25.2, df = 5, p < 0.0001 v2 = 39.1, df = 5, p < 0.0001 v2 = 67.1, df = 5, p < 0.0001 v2 = 78.6, df = 4, p < 0.0001

11,160a 9,900 692

6,883bc 4,715 856

12,144acd 11,040 2,860

11,706ac 8,400 1,087

H or v2 = 22.8 df = 5, p < 0.001

% of hh owning fields

56

11

20

–

4

50

% of hh owning livestock

38

16

6

–

4

39

v2 = 73.0 df = 4, p < 0.0001 v2 = 40.0 df = 4, p < 0.0001

hh size Mean Median SE

7.7 7 0.3

7.0 7 0.3

6.7 6 0.5

6.8 7 0.3

–

7.5 7 0.3

Total annual cash income of hhs (R) Mean Median SE

10,656a 6,205bd 9,840 4,712 1,217 777

H or v2 = 3.8, df = 4, p > 0.05

US$1 = R6.50. A Where Kruskal–Wallis tests were significant, pair-wise comparisons were performed using Mann-Whitney U tests, in which alpha (p level) was decreased by the number of pairs compared, so that unlike superscripts indicate significant differences between groups at p < 0.003 (where all groups are compared) and p < 0.005 (where only product types are compared).

were selling marula beer to pay their school fees. Informal broom traders also tended to

be younger and better educated. Petty trading is hard work with most traders working 10

512

WORLD DEVELOPMENT

hours a day in the market place for six days per week, possibly presenting an entry barrier for older women. Also related to the age pattern was the proportion of households with pensions. Significantly more mat producing and woodworking households had pensions than did other groups. This provided a regular source of cash for these households, decreasing their reliance on the trade. Producer and trader households were largely landless and few possessed livestock, possibly a factor driving them into the trade and contributing to their increased reliance on this activity (Table 4). The exception was among mat producers, where the proportion of households with arable fields and livestock mirrored the random sample at approximately 50% and 40%, respectively. In terms of household income, a higher proportion of producer and trader households (45–50%) were without any regular source of cash income, including welfare grants, than found among the random sample (15%), with the exception of mat producers (12%) (Table 4). Related to this, significantly fewer trading households had household members with formal jobs. Broom producers, marula beer sellers, and woodworkers were thus particularly dependent on the incomes obtained from trading. By contrast, earnings from selling mats were largely supplementary for the majority, but not all, of mat producers. While trading households, except for mat producers,

clearly had fewer sources of regular income than the broader population, differences in mean total household cash income were not as extreme (Table 4). Only marula beer selling and, to a lesser extent, broom producing households had significantly lower incomes than the random population. To explore this further, frequency distributions for four total annual cash income classes, based on quartiles from the random sample, were plotted for each product, firstly including income from trading, and then excluding it in order to observe what difference this made to income profiles (Figure 1). The results indicate that marula beer sellers and broom producers were from among the poorest in the community, with the highest proportion of households in the lowest income class. The pattern for mat producers and broom traders was similar to that of the random sample, both with and without trading income. On the other hand, taking out the income from trading for the woodcraft and broom producer cases resulted in a much higher proportion of households moving into the lowest income class (Figure 1b). This suggests that trading is making a significant contribution to these households helping to lift them into a higher income bracket, and performing an income equalizing role. This ability of natural products sales to reduce income inequality has been demonstrated in other studies (Fisher, 2004).

(b) Without income from the sales of natural products

% of households

(a) With income from the sales of natural products 100 90 80 70 60 50 40 30 20 10 0

1

2

3

1 Mat weavers 2 Broom producers 3 Broom traders

4

5

1

4 Marula beer sellers 5 Woodcrafters

2

3

4

Better off Middle

5

random households

Poor Poorest

Figure 1. Proportion of households (hh) for each product type that fall within four total annual household cash income classes when (a) the net annual income obtained from the sale of natural products is included and (b) when the net annual income obtained from the sale of products is excluded. Classes are based on quartiles for the random household sample, where poorest = 6R3,600, poor = >R3,600–R8,400, middle = >R8,400–R15,090, and better off = >R15,090highest. US$ = R6.50.

LINKS BETWEEN THE LOCAL TRADE IN NATURAL PRODUCTS

The results presented in this section largely support the assertion that it is primarily the most marginalized segments of rural society with few income alternatives that participate in local natural resource trading (Cavendish, 2000; Fisher, 2004; Neumann & Hirsch, 2000). However, as revealed in Figure 1, not all households participating in the trade were equally badly-off; there were exceptions among individual producer and trader households across all product types, and for a large proportion of mat producers who tended to be the most similar in their income profiles to the random population. (c) What are the livelihood benefits from the natural product trade? (i) Incomes from product sales, variation, and comparison with benchmarks Overall, mean incomes per household from trading in natural products were relatively modest even for high value products such as carvings, although in all instances there were ‘‘outlier’’ cases demonstrating significant returns (Table 5 and Figure 2). Pair-wise comparisons showed that average net product incomes for woodworkers and broom producers were significantly higher than for the other groups, at about R7,000 (US$1,077) and R2,000 (US$308) per annum, respectively. Marula beer was the lowest at R520 (US$80). Average net

513

annual product incomes for mat producers and broom traders were approximately R1,000 (US$154). Overall, with the exception of woodworkers, these incomes fell below the stipulated minimum wage, pension rate, and the various poverty line measures (recognizing, however, that some households do obtain cash from other sources), but were higher or comparable to local wage earnings (Table 6). Returns per hour of work were often better than alternatives and approximated the official minimum wage rate, but because earnings were so erratic they generally fell short of the returns from other employment when extrapolated to an annual basis (Shackleton, 2005). When compared to other natural products in different parts of South Africa, mean incomes earned tended to be within similar ranges to many products, but less than those earned by medicinal plant vendors and aloe tappers (Shackleton, Shackleton, Buiten, & Bird, 2007). Relatively modest mean annual cash incomes from product sales tend to be fairly characteristic of the natural product trade worldwide (Schreckenberg, Marshall, & te Velde, 2006). Presenting mean values only, however, provides an extremely limited picture of the incomes earned and the role of these products in household livelihoods, and obscures important patterns. Closer examination of the data revealed considerable heterogeneity in gross and net incomes within households trading

Table 5. Annual cash incomes per household (Rands) derived from product sales Mat producers

Broom producers

Broom traders

Marula beer sellers

Woodworkers

Kruskal–Wallis testA

Gross income Mean Median Maximum Minimum SD SE

1,229cd 800 7,200 75 1,371 139

2,752b 1,792 15,007 356 2,564 313

2,182bc 1,842 19,200 48 3,033 492

687d 512 2,664 84 537 81

14,755a 8,640 48,768 5,008 14,778 4,098

H or v2 = 124.1 df = 4, p < 0.0001

Net incomeB Mean Median Maximum Minimum SD SE

998b 582 5,765 28 1,136 119

2,004a 1,300 14,918 40 2,386 296

1,091b 867 9,600 24 1,516 246

520b 462 2,300 31 469 71

7,427a 6,400 19,328 1,600 5,979 1,658

H or v2 = 52.1 df = 4, p < 0.0001

Income type

US$1 = R6.50. A Unlike superscripts indicate significant differences between mean incomes for product types based on non-parametric (Mann-Whitney U) pair-wise tests at p < 0.005 (reduced by the number of pairs in the comparison). B Net income = gross income minus direct costs (transport, equipment, materials, but not labor).

WORLD DEVELOPMENT

Net annual income (R) from product sales per household

514 20000 17500 15000 12500 10000 7500 5000 2500 0 -2500 mats

broom producers

broom traders

marula beer

woodcraft

Product type median

highest and lowest values

interquartile range containing 50% of values

outliers extremes

Figure 2. Box plots of net annual income from product sales per household for each product type.

Table 6. Benchmarks against which incomes from product sales were compared Benchmark Government stipulated minimum wage—rural—R28.50 per day Regional wage rate—R12.50 per day (Shackleton, 2004) Immediate local wage rate (from survey)—R6 per day Pension Child grant South African poverty line—from Carter and May (1999) and the more recent Bureau of Market Research (2003)—values for the latter are shown in parentheses International poverty line—US$1 and US$2 per person per day

Value per month and per annum R713 per month or R8,556 per annum R312 per month or R3,750 per annum R150 per month or R1 800 per annum R700 per month or R8,400 per annum R150 per month per child or R1,800 per annum R237 (or R398) per person (adult equivalent) per month or R3,150 (R4,777) per person per annum R195 and R390 per month or R2,340 and R4,680 per person per annum

Note: Figures for pensions, etc. are for 2003 when the data were collected. US$1 = R6.50.

the same product (Table 5, Figure 2). Some producers earned incomes of less than R100 (US$15) per month, while others had returns comparable to the benchmarks shown in Table

6. Based on net annual income from product sales, 40% of all producers earned returns of more than R1,000 (US$154) per annum, 22% more than the local wage rate, and 12%

LINKS BETWEEN THE LOCAL TRADE IN NATURAL PRODUCTS

and 7% more than R3,000 (US$462) and R4,000 (US$615) per annum, respectively. Two percent had net incomes above the official minimum wage. Considering woodcraft alone, as a more full-time activity, 23% of producers earned more than the minimum wage, and 54% more than R4,000 (US$615) per annum. For broom producers, some 20% netted more than R3,000 (US$462) per annum. These results suggest that natural product processing and trading, while important as a safety net and in mitigating poverty, may also provide a pathway out of poverty for some individuals and their households, especially considering that, in many cases, income from the product was not the only source of revenue for both producers and their households. Reducing data to statistical averages obscures this finding, sometimes leading to homogenization and potentially misleading conclusions. The need to take a differentiated perspective and to draw out variability has been recognized as central to achieving a more nuanced and in-depth understanding of rural livelihoods and poverty, especially within dynamic dryland systems (Twyman, Sporton, & Thomas, 2004). The above findings raise the question as to why some households achieved higher returns than others. Within case studies, income showed no significant relationships with measured household variables (Shackleton, 2005). Instead earnings appeared to be influenced by participants’ motivations for involvement in the trade, the intensity of involvement (a choice of producers), individual entrepreneurial flair and effort (evident from qualitative life histories), the number of different avenues pursued for marketing products, and access to other income in the household, which helped overcome constraints such as high transport costs. Between product types, seasonality and the degree of value addition were the main factors accounting for differences in income. (ii) Non-financial benefits The livelihoods approach, as well as the broader definitions of poverty, emphasizes the need to move beyond just income to also understanding the social dimensions of rural people’s welfare. This includes notions of well-being, vulnerability and risk, capability, and empowerment. This study revealed many important, and much appreciated, non-financial benefits of natural product trading that were key in reducing vulnerability and improving the quality of life of the individuals and families con-

515

cerned. Independence and self-esteem were important social benefits. Trading reduced producers’ and traders’ reliance on their social safety nets and on donations from other households. This independence is particularly important in the context of declining transfers between rich and poor, or among the poor themselves, due to a deepening of poverty and the impacts of HIV/AIDS (Devereux, 1999). Several producers mentioned how their relatives had encouraged them to trade rather than relying on friends and kin for support. Trading also provided women with their own independent source of income that was invested in the ‘‘good’’ of the household and in paying school fees. This investment in the future may contribute to intergenerational poverty reduction. A few producers used their income to invest in other income generating activities, and in lending clubs and burial societies thus contributing to livelihood diversification and risk reduction. Engagement in the trade was also important in providing an opportunity for participants to work from home; a necessity for some and the quality of life decision for others. Natural product processing and trading provides flexibility in hours, allows producers to do most of their production at or near home, and permits them to set their own pace with their rewards directly related to the effort they put in. This opportunity for home-based employment is likely to become increasingly important as the HIV/AIDS pandemic progresses and the social crisis of care for the ill and orphans worsens (Marcus, 2000). Other benefits included the building of social capital through new and extended social networks, often beyond the immediate family and neighborhood bonds upon which most households rely. These new networks were important in allowing producers to boost their incomes through the sharing of costs; in facilitating diversification into new trading activities; and in providing an additional safety net to draw on in times of crisis. Several women revealed how producing and selling a particular product was important for their psychological well-being by providing them with a way to use their time productively—‘‘it is better to do brooms than sit around at home.’’ Moreover, the fact that producers were seen as mat makers, or woodcarvers or broom producers provided them with an identity, and often their skills, which in many instances helped to perpetuate tradition and culture, were admired. Several other authors have noted the significance of the social

516

WORLD DEVELOPMENT

benefits of, in particular, natural product craft production, for producer’s livelihoods and well-being (e.g., Campbell, Cunningham, & Belcher, 2005; Kru¨ger & Verster, 2001; Marcus, 2000; Rogerson & Sithole, 2001; Terry, 1999). (d) How does the natural product trade fit with the broader household livelihood portfolio? (i) Contribution to total household income and dependence on the trade Woodcraft and broom production, the highest earning activities, tended to contribute, on average, a higher proportion to total cash income (>50%) than marula beer, broom trading, and mat production activities (<50%) (Table 7). Within product types, income shares from trading varied widely between households, from less than 25% to as much as 100% (Table 7). The most destitute households among broom, mat, and marula beer producers generally obtained proportionally more of their income from trading than less-poor households (Figure 3) (the small sample size for woodworkers precluded meaningful disaggregation for this product). Greater dependence on natural product income by more impoverished households has

been revealed in other studies (Cavendish, 2000; Vedeld, Angelsen, Sjaastad, & Kobugabe Berg, 2004), although this is not necessarily universal (Ambrose-Oji, 2003; Schreckenberg et al., 2006). Related to income share was the perceived importance of the trade to households (Table 7). Generally, the higher the product share the more important the producer/trader regarded the income for the household, even if this was low. All woodworkers and 60% of broom producers rated their trading activities as their most important source of cash income, followed by 62% for broom traders, 41% for marula beer sellers, and 13% for mat producers (Table 7). (ii) Exploring the livelihood role of the different products and developing a livelihood strategy typology It is clear from the above that the contribution the trade in natural products makes to the overall household economy can vary widely both within and across product types. In this section we look more closely at the role of natural product trading in relation to the wider household livelihood portfolio and attempt to answer the question: what type of livelihood

Table 7. Income contribution from product sales to total household cash income and the perceived importance of the product to households Product type

Contribution to total cash income (% of households) 0–25% 26–50% 51–75% 76–100% Mean (based on original data)

Mat production

Broom production

Broom trading

Marula beer

Woodcraft

75 11 4 10 21 ± 3

34 25 9 32 51 ± 5

100 0 0 0 11 ± 2

79 21 0 0 14 ± 2

0 23 23 54 77 ± 5

13

60

63 (vending in general)

41 (self-employment in general)

100

14 42 43 1

70 29 1 0

60 40 0 0

– – – –

80 20 0 0

59 41

55 45

80 20

20 80

% of producers/traders rating the trade as their most important source of household cash income Perceived importance of the trade to households (% of producers/traders) Very important Important Slightly important Not important

% of producers/traders who would prefer an alternative source of employment/income Yes 36 No 64

LINKS BETWEEN THE LOCAL TRADE IN NATURAL PRODUCTS

517

Figure 3. Mean plots of the percentage contribution of the sales of three products to total household cash income for households of differing income status from poorest to ‘‘wealthiest’’ (Class 1 = 6R500/month, Class 2 = >R500– R1,000/month, Class 3 = >R1,000–R2,000/month, Class 4 = >R2,000/month). US$1 = R6.50.

strategies are represented by the different products across households and why, and what does this mean in terms of dependence on the resource and the role of the trade in poverty alleviation? As a first step, we undertook an exploratory principal components analysis (PCA), which revealed that the ‘‘number of other sources of income within the household,’’ or in other words income diversification, was an important variable distinguishing individual cases and products (Shackleton, 2005). Based on this, a

‘‘household diversification index’’ was created using Bray–Curtis principal coordinates analysis. Variables included in the analysis related to the number of formal and informal jobs in the household, other self-employment by the producer, number of pensions, number of child grants, and whether the household had any other sources of income at all. The principal coordinate or output scores (or diversification indices) from this analysis were then used to plot cases (Figure 4) and to identify patterns across and within products. While cases were

0.4

Product type

Diversification index (Bray Curtis- axis 2)

woodcraft marula beer broom traders broom producers

0.2

mats

0.0

-0.2

-0.4 -0.4

-0.2

0.0

0.2

0.4

Diversification index (Bray Curtis - axis 1)

Figure 4. Plot of scores from a Bray–Curtis principal coordinates analysis of sources of household income; the output scores represent an index of diversification.

518

WORLD DEVELOPMENT

scattered across the axes, clusters that related to product type were clearly identifiable. Mat producers on the far right-hand side of the figure came from the most diversified households. Woodworkers, marula beer sellers, and broom traders were the least diversified groups, with the least alternative sources of income either because they were specialized in their trading activity or, in the case of marula beer sellers, because they had little else. By contrast, broom producer cases occupied a more central position with processing and trading the main source of income for some households and supplementary for others. These findings were used in developing the livelihood strategy typology described below. Several attempts have been made to group natural product trading activities according to the type of livelihood strategy they represent. Lowore (2003), working in the ‘‘miombo’’ woodlands of Malawi, used the livelihood strategy framework proposed by Devereux (1999) (Table 8) to assess the importance of selling woodland products for household food security. Like this study, she found that the contribution made by trading activities varied among households, even those selling the same product, resulting in the trade representing a range of livelihood strategies from survivalist to accumulative for different poor households. Stack et al. (2003) in an analysis of the mopane worm (Imbrasia belina) trade in southern Africa reached a similar conclusion. Belcher et al. (2005) developed a typology of livelihood strategies based on the groupings that emerged from an analysis of 61 natural product commercialization cases according to the role of the product in the household economy. This typology drew heavily on the degree that trading households were integrated into the cash economy together with the income share from the sales of products. The former criterion is not particularly useful in the South African context as most rural households are well integrated into the cash economy; a result of the systematic undermining of farming livelihoods by past policies (Marcus, 2000). To overcome some of the drawbacks of the existing typologies, a modified typology related directly to our findings is proposed (Table 8, Column 3). This incorporates the definitions provided in the first two columns of Table 8, but refines these in the context of this study and brings in the important notion of choice that has already been mentioned. We then classified the products from this study, using the

understandings gained from the multivariate analyses, against all four strategies, and discussed what this means in terms of their contribution to poverty alleviation (Table 8, Columns 4 and 5). As the analyses illustrated, while there was some association between product type and livelihood strategy, all products were represented by more than one livelihood strategy, often forming a continuum across these. Woodworking was largely a specialization strategy demanding a great deal of producers’ and their families’ time and some reinvestment in the business, although for new entrants it may represent a coping strategy until either other work is found or the producer decides to specialize in woodcraft. It was the producers’ most important source of income (Table 8), and generally provided enough income to meet the basic household livelihood requirements. Although incomes were higher than for most of the other trades, woodworking households were not necessarily well-off. The demands of the trade and the family orientated production mode (Shackleton & Steenkamp, 2004) precluded woodworking households from pursuing other forms of income generation, plus it was a costly business leaving little profit for producers to substantially improve their standards of living. However, woodworking did lift the total income of producer households to levels similar to those of the general population, and several households were able to send their children for tertiary education, an aspiration of many rural households. Woodworking thus has the potential to reduce poverty, although other constraints (such as raw material scarcity and market competition—Shackleton & Steenkamp, 2004) may place limitations on this. It was interesting to note that only 20% of woodworkers expressed a preference for other employment (Table 7). Mat production and broom trading were mainly proactive or ‘‘by choice’’ livelihood diversification strategies. Mat makers generally came from older, established households that tended to have numerous other sources of income within the household, including pensions. Mat making was thus primarily an income supplementing activity, serving to pull household incomes up to higher levels or to provide cash for specific purposes such as school fees, burial society fees, or luxury goods. Incomes from trading were generally small, but together with other cash sources could make a difference to the overall income of households, often easing cash flow constraints and reducing risk and vul-

Table 8. Classification of products using a modified livelihood strategy typology (text in inverted commas in Column 1 refers to the typology used by Stack et al. (2003); product type is in italics where it corresponds to the most important strategy for that product; hh = household) Household livelihood strategy typologies Belcher et al. (2005)

Accumulation Increase in income flows and stocks of assets Proactive and positive

Specialized Intermediate—high hh income > 50% of income in cash NTFP > 50% of income

‘‘Stepping-up’’

Supplementary Intermediate hh income > 50% income in cash NTFP < 50% hh income

‘‘Linking-in’’

Integrated Low hh income Hh income from mixed sources Large % of income in cash NTFP < 50% hh income

Classification of products studied

Specialization Full-time activity Most important source of hh income (few other sources) —>50% share Incomes comparable to low skill wage rates Only sometimes accumulative Some reinvestment in the trading activity May start out as a coping or necessity strategy, but often becomes the activity of choice

Woodcraft Mats (a few producers) Brooms (a few producers)

Diversification ‘‘by choice’’ Supplementary Some choice involved in entry and levels of engagement Seldom most important source of income—< 50% share Numerous other income sources in household Often ad hoc and variable from year to year Low incomes from trade Intermediate to low hh income

Mats Broom trading (petty traders diversified into brooms) Brooms (some producers)

Contribution to poverty alleviation and levels of dependency on the product Ensures basic living requirements Pathway out of poverty for some High dependence

Supplements marginal incomes May be undertaken with a specific objective in mind rather than just survival Reduces risk Diversifies income base With other activities, can assist in raising the standard of living of households Low dependence

519

Adaptive Spread risk through diversification Proactive and positive

Additions and refinements based on this study

LINKS BETWEEN THE LOCAL TRADE IN NATURAL PRODUCTS

Devereux (1999) and Stack et al. (2003)

(continued on next page)

520

Table 8—continued Household livelihood strategy typologies Belcher et al. (2005)

Coping Minimization of the costs of adverse livelihood shocks, such that future livelihood capacity is not seriously impaired Reactive and defensive

Subsistence Low hh income <50% income in cash NTFP < 50% of income

‘‘Hanging-on’’/‘‘Linking-in’’

Survival Prevention of destitution Reactive and defensive

‘‘Hanging-on’’

Additions and refinements based on this study Coping or ‘‘necessity’’ Response to setback Response to few other opportunities Entered through lack of choice Few other income sources in hh and low hh income Often most important source of income to hh—>50% share Can become a long-term activity Can replace other safety nets

Broom production Mats (a few producers) Marula beer Woodcraft

Survival Marula beer No choice—last resort Brooms Low returns and low entry barriers No other income sources Still rely on other safety nets and other similar survival activities <50% share

Contribution to poverty alleviation and levels of dependency on the product Safety net Mitigation of poverty Reduced reliance on other social safety nets Decreased vulnerability Contributes to food security Important for school fees High dependence

Safety net Prevention of destitution Contributes to food security Important for school fees Moderate dependence— still need other safety nets and activities

WORLD DEVELOPMENT

Devereux (1999) and Stack et al. (2003)

Classification of products studied

LINKS BETWEEN THE LOCAL TRADE IN NATURAL PRODUCTS

nerability. Generally, producers only participated in the trade to the extent they felt was necessary to meet their specific cash needs and to fit with other household activities. However, a few producers had turned mat making into their primary livelihood activity and were earning reasonable incomes, even hiring other people to assist, thus moving into a specialization strategy (Shackleton, 2005). At the other extreme, mat weaving also sometimes provided a safety net, or means of coping, for poor households with few other income sources, particularly grandmothers caring for AIDS orphans. Broom trading was an interesting case, in that trading as a whole tended to be a specialized strategy (hence the position of this group shown in Figure 4), while selling brooms was a way of diversifying the type of products offered for sale. Brooms did not earn traders high incomes, except for a few selling in pension markets 3 across the municipality, but provided a reliable source of income at little risk since they did not perish in the same way as many of the other goods traders were selling. Although mat producers and broom traders were not highly dependent on the trade, its role in elevating often otherwise marginal incomes should not be dismissed. Producers’ own perspectives were that these products made an important contribution to their livelihoods, with the majority (64%) expressing satisfaction with this activity probably because it fitted well with their lifestyle (Table 7). Broom production was primarily a coping or ‘‘necessity’’ strategy for the women involved. Most broom producers had little else in the way of household income, and had entered the trade following a setback in the household. Broom production was, thus, often the most important source of income for the household, although net profits were modest. For a few producers, however, brooms provided more substantial income, playing a role in reducing poverty levels in those households. In other cases, it provided a means to diversify income sources particularly among households headed by members other than producers themselves. However, for the most part, the broom trade provided a safety net for producers, sometimes replacing other safety nets such as social transfers between households. Even though broom producers entered the trade due to shocks, the lack of alternatives meant that this safety net role had become long term (or an adaptive strategy) for a large number of producers.

521

The sale of marula beer represented a survival or coping strategy for most households. Marula beer sellers tended to be representative of the poorest households in the community and were typically younger, ‘‘incomplete’’ and female-headed and therefore unable to diversify. They were the group with fewest ‘‘regular’’ sources of alternative income and the lowest cash income (Table 4). Several sellers reported how they depended on relatives for food. Marula beer was one of the easiest trades to enter, and could provide good returns for the three months of the fruiting season. Some women, from better off households, entered this trade with the specific aim of paying their children’s school fees. The timing of the income is critical in this respect. For others, it provided much needed cash, if even for an extremely short period, or was a better source of income than the ad hoc activities they normally depended on. This product, thus, plays a vital role in preventing destitution and enhancing food security, but due to its seasonality can rarely replace other safety nets or result in poverty reduction. Furthermore, the fact that a much younger group of women were involved in beer selling compared to the other products suggests that this constitutes a temporary activity while participants seek more permanent and lucrative sources of income. 5. CONCLUSIONS: THE ROLE OF THE NATURAL PRODUCT TRADE IN POVERTY ALLEVIATION If we accept the new thinking on poverty that places emphasis on income distribution, human capital, vulnerability, and an obligation to protect the poor (Ashley & Maxwell, 2001), then the natural product trade has a role to play. This study has demonstrated that these products clearly make a difference to the welfare of the most marginalized sectors of the community, delivering a range of financial and nonfinancial benefits. Although only in limited instances was the trade able to boost household incomes to any significant extent, it nevertheless played a critical role in (a) providing additional options for income generation in the context of few opportunities, (b) allowing households to diversify and supplement their income base, (c) providing a safety net for those facing shock and hardship, (d) reducing reliance on other safety nets such as inter household transfers and state welfare, and (e)

522

WORLD DEVELOPMENT

meeting specific cash needs such as school fees. Thus, together with the less tangible benefits of self-reliance, identity and respect, new skills and social networks, flexibility, and often a more dignified way of earning a living than some of the alternatives for low skilled people, few could argue that the natural product trade does not have a positive impact on the wellbeing of the individuals and households involved. This is particularly so when the alternative is demoralizing joblessness and greater vulnerability. Producers’ own comments and perspectives regarding the benefits of the trade as well as the measurable proportion entering the trade by choice are clear indicators that these benefits are very real to those involved. However, despite the benefits, the fact is that it is generally the poorest households who depend most on natural product sales and they tend to remain poor due to marginal returns. This has led some commentators to suggest that these products may have little role in poverty alleviation and may indeed constitute a ‘‘poverty trap’’ in the sense that dependence on them undermines possibilities for accumulation and diversification in the household (Neumann & Hirsch, 2000; Wunder, 2001). On the other hand, those with a more optimistic view see natural products as a crucial safety net preventing the poor from descending deeper into poverty (Campbell et al., 2002; Scherr et al., 2004). A similar divergent viewpoint exists within the small enterprise literature. Most commentators concur that ‘‘micro-enterprises can be a positive factor in contributing towards poverty alleviation’’ because of the numbers and types of people involved (e.g., marginalized, women) (UNDP South Africa, 2003). Indeed, Mead and Liedholm (1998) suggested that single person, survivalist micro-enterprises are a particularly appropriate target group for those concerned with poverty alleviation. Others, however, believe that ‘‘necessity’’ entrepreneurial activities cannot provide an effective route to tackling poverty because of their limited capacity for growth (Driver, Wood, Segal, & Herrington, 2001). Before drawing conclusions either way it is important that the role of natural product trading is viewed in the wider socio-economic and poverty context within which this livelihood opportunity is pursued, as well as the suite of reasons producers themselves provide for engaging in the trade. The alternatives available to the rural poor, who are often poorly educated and skilled with limited access to formal

employment opportunities, are seldom considered. Indeed, the unemployment crisis in South Africa undermines even migration as a viable option. So, while the returns from natural product trading are generally modest, they frequently compare favorably with (a) the immediate alternatives of local wage labor, assuming that it is available, (b) income from other sources such as farming, which is also marginal in the South African communal area context, and (c) other self-employment, which often requires upfront investment that people simply do not have. In semi-arid regions like Bushbuckridge, the situation is further exacerbated by the harshness of the climate and the lack of viable, alternative land uses. Moreover, the HIV/AIDS crisis in South Africa adds a chronic and long-term care and financial burden to already stressed and vulnerable households (Marcus, 2000; UNDP South Africa, 2003), and often forces members into homebased income generation activities (Kepe, 2002). The current study has shown how the natural resource base provides a crucial safety net for HIV/AIDS affected households, while it is argued in the South African Human Development report (UNDP South Africa, 2003) that engagement in small enterprise activities has a positive role in assisting households to offset some of the added financial burden caused by HIV/AIDS. Campbell et al. (2002) reached similar conclusions from their livelihood studies in Chivi, Zimbabwe. Given the context outlined above, and the results from this study, which show that the trade in natural products may perform a variety of functions from assisting households to survive and cope, to allowing them to diversify or, at the other end of the spectrum, to specialize and accumulate (Table 8), it would be inappropriate to label this activity as a poverty trap. Indeed, the incomes earned from trading make a real impact in assisting households to deal with the hardship they face. This study has shown that the returns from selling selected natural products can help to lift producer households to income levels approximating those of the broader population. Moreover, some producers, by expending effort, were able to earn significant incomes, whereas others chose a less intensive level of involvement. This points to the need to interpret the contribution of natural products, not just within the broader socio-economic context, but also in the context of the livelihood dynamics and motivations of individual households. Both household and indi-

LINKS BETWEEN THE LOCAL TRADE IN NATURAL PRODUCTS

vidual producer and trader attributes and their reasons for engaging in the trade can influence the magnitude of the benefits derived. As the livelihood strategy classification showed, it is often only the poorest households that rely almost entirely on natural products or those that are already doing well (Table 8), while for others the production and sale of these products constitutes a livelihood diversification strategy. Income from trading should, therefore, not be considered in isolation from how it fits in with and contributes to the broader household economy. The positive contribution that natural product trading can make in easing poverty and hardship is supported by those working in the micro-enterprise sector, where there is evidence, particularly in a supportive policy environment, that ‘‘the poor can be a little less poor’’ through these activities (UNDP South Africa, 2003). A more moderate perspective is thus called for that emphasizes the central role of natural product trading in addressing the more fundamental goal of poverty prevention as well as poverty elimination. In conclusion, there is a need and obligation for the state to underpin the safety net offered by biodiversity, as well as to support those people who have turned to natural product trading as a means to make ends meet under increasingly harsh economic conditions. There is a need to view specific livelihood activities from a more holistic perspective, and to ‘‘favor livelihood-strengthen-

523

ing diversification options that promote risk averse, multi-occupational households’’ (Ashley & Maxwell, 2001) in addition to more specialized activities. Natural resource commercialization activities should not be rejected as unsuccessful or unworthy of support just because they do not always foster high incomes. Households engage in natural product trading for different reasons, with each product having its role and place that varies from household to household. This heterogeneity needs to be appreciated and embraced, and a more differentiated, locally embedded approach adopted. Most importantly there is a need to be cautious about making generalizations, as the variability and complexity in the data from this study, despite a fixed socioeconomic context, clearly demonstrate, and run the risk of underestimating or disregarding the role of the natural product trade as a contributor to rural livelihood security and poverty alleviation. However, at the same time there is a need for realism and to avoid overstating the potential of these products, as tended to be done in the past, but rather see them as one component of a multi-sectoral approach for tackling rural poverty. Thus, natural product trading alone is not the answer, but nor is arable production, livestock rearing, migrancy, or state welfare grants. It is only through the integration of these livelihood sectors that there will be any lasting positive impact on the welfare of the rural poor.

NOTES 1. The Bantustans or homelands were created as labor reserves legitimated by a complex of apartheid ideals and policies that emphasized the importance of separate development. They were the only areas where black people could access land. In the early 1990s, all homelands were reincorporated into South Africa.

2. Only the broom trade had both producers and traders operating within Bushbuckridge (see Table 1). 3. Pension markets are informal markets thatarise around the mobile state pension payout points. They provide an opportunity for entrepreneurs, petty traders, and producers of a range of local products to sell their goods.

REFERENCES Adams, W. M., Aveling, R., Brockington, D., Dickson, B., Elliott, J., Hutton, J., et al. (2004). Biodiversity conservation and the eradication of poverty. Science, 306, 1146–1149. Aliber, M. (2003). Chronic poverty in South Africa: Incidence, causes and policies. World Development, 31(3), 473–490.

Ambrose-Oji, B. (2003). The contribution of NTFPs to the livelihoods of the ‘forest poor’: Evidence from the tropical forest zone of south-west Cameroon. International Forestry Review, 5(2), 106–117. Arnold, J. E. M. (2002a). Clarifying the links between forests and poverty reduction. International Forestry Review, 4(3), 231–234.

524

WORLD DEVELOPMENT

Arnold, J. E. M., (2002b). Socio-economic benefits and issues in non-wood forest products use. Appendix 4.1.1. Non-wood forest products for rural income and sustainable forestry. Rome: Food and Agricultural Organization (FAO). Arnold, J. E. M., & Townson, I. (1998). Assessing the potential of forest product activities to contribute to rural incomes in Africa. Natural Resource Perspectives (No. 37). London: Overseas Development Institute (ODI). Ashley, C., & Maxwell, S. (2001). Theme issue: Rethinking rural development. Development Policy Review, 19(4), 395–425. Beck, T., & Nesmith, C. (2001). Building on poor people’s capacities: The case of common property resources in India and West Africa. World Development, 29(1), 119–133. Belcher, B. (2005). Forest product markets, forests and poverty reduction. International Forestry Review, 7(2), 82–89. Belcher, B., Ruiz-Pe´rez, M., & Achdiawan, R. (2005). Global patterns and trends in the use and management of commercial NTFPs: Implications for livelihoods and conservation. World Development, 33(9), 1435–1452. Bird, N., & Dickson, C. (2005). Poverty reduction strategy papers: Making the case for forestry. ODI Forestry Briefing (No. 7). London: Overseas Development Institute (ODI). Bryceson, D. F. (2002). The scramble for Africa: Reorienting rural livelihoods. World Development, 30(5), 725–739. Bureau of Market Research (2003). Research report 3/19 on minimum living levels. Pretoria: University of Pretoria. Campbell, B., Cunningham, A., & Belcher, B. (2005). Carving out a future: Planning for woodcarving in the 21st century. In A. Cunningham, B. Belcher, & B. Campbell (Eds.). Carving out a future. Forests, livelihoods and the international woodcarving trade (pp. 249–269). London: Earthscan. Campbell, B., Jeffrey, S., Kozanayi, W., Luckert, M., Mutamba, M., & Zindi, C. (2002). Household livelihoods in semi-arid regions. Options and constraints. Bogor, Indonesia: Centre for International Forestry Research (CIFOR). Carter, M. R., & May, J. (1999). Poverty, livelihood and class in rural South Africa. World Development, 27(1), 1–20. Cavendish, W. (2000). Empirical regularities in the poverty–environment relationship in rural households: Evidence from Zimbabwe. World Development, 28(11), 1979–2003. Chant, S. (1997). Women-headed households: Poorest of the poor? Perspectives from Mexico, Costa Rica and the Philippines. IDS Bulletin, 28(3), 26–48, Brighton: Institute of Development Studies (IDS), University of Sussex. Devereux, S. (1999). Making less last longer: Informal safety nets in Malawi. IDS Discussion Paper 373. Brighton: Institute of Development Studies (IDS), University of Sussex.

Driver, A., Wood, E., Segal, N., & Herrington, M. (2001). Global entrepreneurship monitor: 2001 South African executive report. Cape Town: Graduate School of Business, University of Cape Town. FAO (2003). State of the world’s forests 2003. Part II. Selected current issues in the forest sector. Rome: Food and Agricultural Organization (FAO). Fisher, M. (2004). Household welfare and forest dependence in Southern Malawi. Environment and Development Economics, 9, 135–154. Gelb, S. (2003). Inequality in South Africa: Nature, cases and responses. Section 2. An overview of inequality, poverty and growth in South Africa. DFID policy initiative in addressing inequality in middle-income countries. Johannesburg: The Edge Institute.. Godoy, R., Wilkie, D., Overman, H., Cubas, A., Cubus, G., Demmer, J., et al. (2000). Valuation of the consumption and sale of forest goods from a Central American rainforest. Nature, 406, 62–63. Gyan, C., & Shackleton, C. M. (2005). Abundance and commercialization of Phoenix reclinata in the King Williams Town area, South Africa. Journal of Tropical Forest Science, 17, 325–336. Kepe, T. V., (2002). Grassland vegetation and rural livelihoods: A case study of resource value and social dynamics on the Wild Coast, South Africa. Unpublished PhD thesis, Cape Town, South Africa: University of the Western Cape. Koziell, I. (2001). Diversity not adversity. Sustaining livelihoods with biodiversity. London: International Institute for Environment and Development (IIED). Kru¨ger, S., & Verster, R. (2001). An appraisal of the Vulamehlo handcraft project. Development Southern Africa, 18(2), 239–252. Lawrence, A. (2003). No forest without timber? International Forestry Review, 5(2), 87–96. Lowore, J., (2003). Miombo woodlands and rural livelihoods in Malawi. Unpublished report submitted to the Centre for International Forestry Research (CIFOR). Mzuzu, Malawi: Department of Forestry, Mzuzu University. Marcus, T. (2000). Crafting in the context of AIDS and rural poverty: A livelihood strategy with prospects. Transformation, 44, 17–35. Marshall, E., & Newton, A. C. (2003). Non-timber forest products in the community of El Terrero, Sierra de Manantlan Biosphere Reserve, Mexico: Is their use sustainable? Economic Botany, 5(2), 262–278. Marshall, E., Schreckenberg, K., & Newton, A. C. (Eds.) (2006). Commercialization of non-timber forest product: Factors influencing success. Lessons learned from Mexico and Bolivia and policy implications for decision-makers. Cambridge: UNEP World Conservation Monitoring Centre. McSweeny, K. (2004). Tropical forests product sale as natural insurance: The effects of household characteristics and the nature of shock in Eastern Honduras. Society and Natural Resources, 17, 39– 56.

LINKS BETWEEN THE LOCAL TRADE IN NATURAL PRODUCTS Mead, D. C., & Liedholm, C. (1998). The dynamics of micro and small enterprises in developing countries. World Development, 26(1), 61–74. Meth, C., & Dias, R. (2004). Increases in poverty in South Africa, 1999–2002. Development Southern Africa, 21, 59–85. Monela, G. C., Kajembe, G. C., Kaoneka, A. R. S., & Kowera, G. (1999). Household livelihood strategies in the miombo woodlands of Tanzania: Emerging trends. Tanzania Journal of Forestry and Nature Conservation, 73, 17–33. Narendran, K., Murthy, I. K., Suresh, H. S., Dattaraja, H. S., Ravindranath, N. H., & Sukumar, R. (2001). Nontimber forest product extraction, utilization and valuation: A case study from the Nilgiri Biosphere Reserve, Southern India. Economic Botany, 55(4), 528–538. Nattrass, N. (2004). Unemployment and AIDS: The social–democratic challenge for South Africa. Development Southern Africa, 21, 87–108. Ndoye, O., Ruiz-Pe´rez, M. & Eyebe, A., (1997). The markets of non-timber forest products in the humid forest zone of Cameroon. ODI Rural Development Forestry Network Paper No. 22c. London: Overseas Development Institute (ODI). Neumann, R. P., & Hirsch, E. (2000). Commercialization of non-timber forest products: Review and analysis of research. Bogor, Indonesia: Centre for International Forestry Research (CIFOR). Oksanen, T., & Mersmann, C. (2003). Forestry in poverty reduction strategies—An assessment of PRSP processes in sub-Saharan Africa. In T. Oksanen, B. Pajari, & T. Tuomasjukka (Eds.), Forestry in poverty reduction strategies: Capturing the potential. EFI Proceedings (No. 47, pp. 121–158). Joensuu: European Forest Institute. Posel, D. R. (2001). Who are the heads of household, what do they do, and is the concept of headship useful? An analysis of headship in South Africa. Development Southern Africa, 18(5), 651–670. Reddy, S. R. C., & Chakravarty, S. P. (1999). Forest dependence and income distribution in a subsistence economy: Evidence from India. World Development, 27(7), 1141–1149. Roe, D. (2004). The millennium development goals and conservation. Managing nature’s wealth for society’s health. London: International Institute for Environment and Development (IIED). Rogerson, C. M., & Sithole, P. M. (2001). Rural handicraft production in Mpumalanga, South Africa: Organization, problems and support needs. South African Geographic Journal, 83(2), 149–158. Ros-Tonen, M. A. F., & Wiersum, K. F. (2005). The scope for improving rural livelihoods through nontimber forest products: An evolving research agenda. Forests, Trees and Livelihoods, 15, 129–148. Scherr, S. J., White, A., & Kaimowitz, D. (2004). A new agenda for forest conservation and poverty reduction. Making markets work for low income producers. Washington, DC: Forest Trends and Centre for International Forestry Research (CIFOR). Schreckenberg, K., Marshall, E., & te Velde, D. W. (2006). NTFP commercialization and the rural poor.

525

More than a safety net? In E. Marshall, K. Schreckenberg, & A. C. Newton (Eds.), Commercialization of non-timber forest product: Factors influencing success. Lessons learned from Mexico and Bolivia and policy implications for decision-makers (pp. 71–76). Cambridge: UNEP World Conservation Monitoring Centre. Shackleton, S. E. (2004). Livelihood benefits from the local level commercialization of savanna resources: A case study of the new and expanding trade in marula (Sclerocarya birrea) beer in Bushbuckridge, South Africa. South African Journal of Science, 100, 651–657. Shackleton, S. E. (2005). The significance of the local trade in natural resource products for livelihoods and poverty alleviation in South Africa. Unpublished PhD thesis, Rhodes University, Grahamstown, South Africa. Shackleton, C. M., & Shackleton, S. E. (2004). The importance of non-timber forest products in rural livelihood security and as safety nets: A review of evidence from South Africa. South African Journal of Science, 100, 658–664. Shackleton, S. E., & Shackleton, C. M. (2005). The contribution of marula (Sclerocarya birrea) fruit and fruit products to rural livelihoods in the Bushbuckridge district, South Africa: Balancing domestic needs and commercialization. Forest, Trees and Livelihoods, 15(1), 3–24. Shackleton, C. M., Shackleton, S. E., Buiten, E., & Bird, N. (2007). The importance of dry woodlands and forests in rural livelihoods and poverty alleviation in South Africa. Forest Policy and Economics, 9, 558–577. Shackleton, S. E., & Steenkamp, C. (2004). The woodcraft industry in the Mpumlanga–Limpopo Province Lowveld. In M. J. Lawes, H. A. C. Eeley, C. M. Shackleton, & B. G. S. Geach (Eds.), Indigenous forests and woodlands in South Africa: Policy. People and practice (pp. 399–430). Pietermarizburg: University of KwaZulu-Natal Press. Stack, J., Dorward, A., Gondo, T., Frost, P., Taylor, F., & Kurebgaseka, N. (2003). Mopane worm utilization and rural livelihoods in Southern Africa. Proceedings of the international conference on rural livelihoods, forests and biodiversity, 19–23 May 2003, Bonn, Germany. Bogor, Indonesia: Centre for International Forestry Research (CIFOR). Stats SA (Statistics South Africa) (2004). 2001 Census— municipal summaries. . Sunderlin, W. D., Angelsen, A., Belcher, B., Burgers, P., Nasi, R., Santosa, L., et al. (2005). Livelihoods, forests and conservation in developing countries: An overview. World Development, 33(9), 1383–1402. Takasaki, Y., Barham, B. L., & Coomes, O. T. (2004). Risk coping strategies in tropical forests: Floods, illnesses, and resource extraction. Environment and Development Economics, 9, 203–224. Taylor, D. A. (1999). Requisites for thriving rural nonwood forest product enterprises. Unasylva, 198(50), 3–8. Terry, E. M. (1999). The economic and social significance of the handicraft industry in Botswana. Unpublished

526

WORLD DEVELOPMENT

PhD thesis, School for Oriental and African Studies, University of London, London. Twyman, C., Sporton, D., & Thomas, D. S. G. (2004). Where is the life in farming? The viability of smallholder farming on the margins of the Kalahari, Southern Africa. Geoforum, 35, 69–85. UNDP South Africa (2003). South Africa human development report 2003. The challenge of sustainable development: Unlocking people’s creativity. Cape Town: Oxford University Press. United Nations Millennium Declaration (2000). Millenium Development Goals. . Vedeld, P., Angelsen, A., Sjaastad, E. & Kobugabe Berg, G. (2004). Counting on the environment. Forest incomes and the rural poor. Environment Department

Papers, Paper No. 98, Environmental Economics Series. Washington DC: The World Bank Group. World Bank (2002). A revised forest strategy for the World Bank Group. Washington, DC: World Bank Group. Wunder, S. (2001). Poverty alleviation and tropical forests—What scope for synergies? World Development, 29(11), 1817–1833. WRI (2000). World resources report 2000–2001. People and ecosystems: The fraying web of life. World Resources Institute, UNDP, UNEP and World Bank. Washington, DC: World Resources Institute (WRI). WRI (2005). World Resources 2005. The wealth of the poor. Managing ecosystems to fight poverty. World Resources Institute, UNDP, UNEP and World Bank. Washington, DC: World Resources Institute (WRI).

Available online at www.sciencedirect.com