Commercialisation of mopane worm (Imbrasia belina) in rural households in Limpopo Province, South Africa

Forest Policy and Economics 62 (2016) 141–148

Contents lists available at ScienceDirect

Forest Policy and Economics journal homepage: www.elsevier.com/locate/forpol

Commercialisation of mopane worm (Imbrasia belina) in rural households in Limpopo Province, South Africa L.J.S. Baiyegunhi ⁎, B.B. Oppong SAEES — Discipline of Agricultural Economics, University of KwaZulu-Natal, P. Bag X01, Scottsville, 3209 Pietermaritzburg, South Africa

a r t i c l e

i n f o

Article history: Received 8 April 2015 Received in revised form 24 August 2015 Accepted 31 August 2015 Available online 5 September 2015 Keywords: NTFPs Mopane worm Commercialisation Livelihood Income Conservation South Africa

a b s t r a c t Non-Timber Forest Products (NTFPs) have been identified as a key source of income for rural households. In South Africa, mopane worms (Imbrasia belina) have become an important source of food and cash income for rural people living in or near the mopane woodlands. However, the continued over-exploitation and commercialisation has implications for long term management of the mopane woodlands in the northern parts of South Africa. This study used a cross-sectional dataset collected from a survey of 120 households in Limpopo province in an effort to identify socio-economic factors influencing mopane worm commercialisation and intensity of commercialisation. The Household Commercialisation Index (HCI) and the Double Hurdle Model were used for the analysis. Result shows that about 63% of mopane worm harvested is sold within a production year which implies high level of commercialisation. The result of the Double Hurdle Model showed that gender, education, household size, quantity harvested, social capital, distance, transportation and information impact on households’ decision-making to commercialised mopane worm. Furthermore, household age, gender, education, exogenous income, price, quantity of marketable surplus, absence of institution/law and transportation are statistically significant factors influencing the intensity of mopane worm commercialisation in the study area. This therefore suggests that, in order to achieve a balance between sustainable harvesting of mopane worm and improving the livelihoods of the rural poor that depends on it for food and income, there is a need for policy makers to focus on the dynamics in household socio-economic conditions. Implications for policy were discussed. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Non-Timber Forest Products (NTFPs) are biological matter of wild plants and animals other than timber extracted from forests and woodlands (e.g. fruits, nuts, vegetables, game, medicinal plants, resins, bark, fibers, palms, grasses including small wood products, firewood etc), harvested by local households and communities from around homestead, fields, grazing land and relatively intact vegetation for domestic consumption or trade (CIFOR, 2011; Shackleton et al., 2007, 2011). Trade in NTFPs has been recognized as promoting forest conservation, supporting current consumption, longer-term poverty reduction and the safety-net of rural households by enhancing their livelihoods through increased cash income, especially when other sources of income fail to meet household needs (Angelsen et al., 2014; Rizek and Morsello, 2012; Pouliot, 2012; Saha and Sundriyal, 2012; Heubach et al., 2011; Paumgarten and Shackleton, 2011; Areki and Cunningham, 2010; Kamanga et al., 2009; Babulo et al., 2009; Cocks et al., 2008). The contribution of NTFPs trade to livelihoods typically ranges from 5% to over 90% of total household income, depending on the degree of ⁎ Corresponding author. E-mail address: [email protected] (L.J.S. Baiyegunhi).

http://dx.doi.org/10.1016/j.forpol.2015.08.012 1389-9341/© 2015 Elsevier B.V. All rights reserved.

involvement and specialisation, extent of value adding and seasonality of the resource (Asfaw et al., 2013; Moresello et al., 2012; Shackleton et al., 2008), with some cases indicating an income share greater than that from cash crops and informal cash incomes (Dovie, 2003). For example, studies have shown that in Asia and Africa, as much as 20.1% and 21.4% respectively of rural households’ income may be derived from forest incomes (Angelsen et al., 2014). The cash value and directuse value of NTFPs has been calculated to be worth several hundred dollars per annum per household (Dovie et al., 2005; Shackleton and Gumbo, 2010). Therefore, many resource poor households in rural communities all over the world depend on NTFPs for their survival (Shackleton and Gumbo, 2010). NTFPs also are valued for their biological diversity for medicinal, local aesthetic use as well as for their cultural, recreational and spiritual values (Kim et al., 2012; Cocks et al., 2011; Areki and Cunningham, 2010; Muzayen, 2009; Cocksedge, 2006). A majority of South Africa’s rural population lives on communal lands in the former homelands, where poverty is endemic and has remained pervasive (Baiyegunhi and Fraser, 2011, Aliber, 2003). As a result, the consumption and commercialisation of NTFPs is seen as an important potential way for improving households’ living standards and achieving solution to the endemic poverty (Scherr et al., 2004; Shackleton and Shackleton, 2004). Mopane worms, caterpillars of the emperor moth

142

L.J.S. Baiyegunhi, B.B. Oppong / Forest Policy and Economics 62 (2016) 141–148

Imbrasia belina, which feed virtually solely on the mopane tree Colophospermum mopane, are valuable NTFP resource that are used for subsistence and sold to generate income (Makhado et al., 2012; Ghazoul, 2006). In Southern Africa, estimates shows that the mopane woodland covers about 30-35% of the 1.5 million km2 of savannas with an estimated 23,000 km2 of mopane woodland occurring in the warmer Limpopo and Mpumalanga Provinces (Mapaure, 1994). It was previously estimated that from these areas, a total an annual population of 9500 million mopane worms worth US$84 million was harvested, of which approximately 40% was harvested by rural people who are often poor women (Styles, 1994). Recently, it has been estimated that about 16,000 metric tonnes of mopane worms are traded commercially at between US$2.50–US$4.00 per kilogram in South Africa, generating about US$39–US$59 million per annum (Potgieter et al., 2012). However, due to the reduction in availability of mopane worms, the value is currently estimated at between US$30–US$50 million (Makhado et al., 2014). Makhado et al. (2009) found that individual mopane worm traders in the Limpopo Province, South Africa can earn about R20,000 (US$2457) per annum. With an increasing focus on poverty alleviation in most developing countries, NTFPs have been considered for their role in minimizing the impact of crises on rural households and as a possible means to assist households to move out of poverty (Shackleton and Pandey, 2014; Cosyns et al., 2011; Rasul et al., 2008). However, rural communities are increasingly selling products that were previously used for subsistence purposes (Gondo et al., 2010; Dovie et al., 2005). This change is motivated at local level by a greater need for cash as people become more integrated into a market economy and face economic hardship and unemployment (Belcher et al., 2005). The low entry barriers to trade for poor and marginalized people with minimal education and skills make NTFPs an important option (Shackleton and Gumbo, 2010). Furthermore, the areas dominated by mopane trees generally have low agricultural potential where cultivation is risky, with regular failure of staple grain crops leading to a high degree of livelihood vulnerability (Hope et al., 2009). Also, agricultural productivity from tribal lands is inadequate, as it just maintains a household at subsistence level (Papola, 2000). Hence, mopane worms represent an important source of food and income at a time when many households have limited alternatives (Frost, 2005). Despite the potential income benefits associated with the commercialisation of NTFPs especially mopane worm, there is concern that the rising commercialisation in the global market can have a variety of ecological consequences including resource depletion, changes in biodiversity and resource quality as well as on the subsistence use of NTFPs, particularly in areas where there is high dependence on it for daily nutrition and income (Makhado et al., 2012; Makhado, 2008; Belcher et al., 2005). If the abundance or productivity of NTFP species is impaired, then the potential contributions to rural livelihoods will diminish over time (Mutenje et al., 2011; Thang et al., 2010). Therefore, in order to keep the balance between conservation and households’ livelihood, the link between economic benefits of NTFPs and their resource base (including local availability and sustainability) need to be understood. In so doing, the relation between households' socio-economic conditions and commercial collection of mopane worms should be identified first. Within any given community, there is significant socio-economic differentiation arising from a multitude of factors such as levels of education, employment, income, age, family size relationship to elites etc. (Kepe, 2002; Cavendish, 2000). It is important to examine such differentiation when considering policy and management interventions to support rural livelihoods and promote sustainable use of natural resources. This is because it is plausible that different socio-economic groups will perceive and use NTFPs differently. In order to identify the significant relations between households' socio-economic conditions and commercial collection of mopane worms, this study addressed two broad questions. First, how much does mopane worm commercialisation

contribute to rural households’ income portfolio? Second, what household level socio-economic characteristics and contextual and market variables affect mopane worm commercialisation and the intensity of commercialisation among rural households in Limpopo province, South Africa. 2. Materials and methods 2.1. Study area The study was conducted in Greater Giyani and Greater Letaba local municipalities in Mopani district of Limpopo Province, South Africa. The Mopani district is vast with about 90% of the population living in rural areas, with estimated population in 2011 of about 1,061,815 people, with a total of 263,862 households within an area of 20,010 km2. The Greater Giyani local municipality has a population size of 244,217 people, with 63,548 households which makes it the second largest municipality, while Greater Letaba municipality with a population size of 212,701 and 58,261 households is the third largest municipality in terms of population in the Mopani District (StatisticsSA, 2011). Formal employments are limited and much of the population depends on state welfare grants, in particular old age pensions, and migrant remittances for cash income, subsistence agriculture and dependence on NTFPs for nutrition and income. Poverty levels remain very high, resulting in considerably high levels (63%) of food insecurity (D’Hase and Vermeulen, 2011), and with more than half (57.5%) of the population living below the poverty line (LDLGH, 2007). 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 job loss, high costs of living, a lack of arable land and increasing vulnerability and poverty arising from the HIV/AIDS epidemic (Nattrass, 2004; Aliber, 2003). The vegetation of the study area is typical semi-arid savannah, with rainfall ranging from 500 mm in the east to 1000 mm in the west. Intraand inter-seasonal drought is also very frequent occurring 3–4 years; this limits dry land arable production (Shackleton et al., 2007). The subtropical climate in the area gives rise to the growth of vegetation classified under the lowveld mopaneveld savanna (Mucina and Rutherford, 2006) where C. mopane grows in abundance. The map of the study area showing the Greater Giyani and Greater Letaba Municipalities where respondents were surveyed for the study is presented below as Fig. 1. 2.2. Sampling and methods of data collection Exploratory analyses were carried out using a case study data from a random survey of households conducted between August 2012 and May 2013 in four purposively selected villages within the Greater Giyani and Greater Letaba municipalities based on prior knowledge of mopane worm harvesting and commercialisation in these areas. These villages display obvious differences in several aspects of their ecological settings, location, social and economic characteristics. However, at a superficial level they may be regarded as typical of rural villages in any of South Africa’s former homelands, as they have similar characteristic in terms of low levels of development, poor service provision, high unemployment and a reliance on variety of livelihood strategies including arable agriculture, animal husbandry, formal and informal employment, government welfare grants and use and sale of NTFPs and environmental goods and services (Paumgarten and Shackleton, 2009; Frost, 2005). Hence, in this study, they are viewed as a case study rather than a representative samples. The study population of mopane worm harvesting households or the exact number of collecting villages within the district was not known. Therefore, a reconnaissance survey was first conducted with the assistance of the government extension officer in the area to identify households involved in mopane worm harvesting and commercialisation. The

L.J.S. Baiyegunhi, B.B. Oppong / Forest Policy and Economics 62 (2016) 141–148

143

Fig. 1. Map of Mopani district showing the Greater Giyani and Greater Letaba Municipalities.

list of the household obtained served as the sample frame from which a total of 120 households (by randomly selecting 30 respondent households per village) were chosen for the study. The randomly drawn sample size was 25–30% of the total household listed in sample frame from each village. The household head was the unit of analysis. Qualitative data were collected through focus group discussion (FGD), key informant interviews (KII), village meetings, and personal field observations. Eight households in each of the four villages were purposively selected to attend the focus group discussions. Key informants such as community leaders were also approached to identify current challenges and opportunities regarding mopane worm harvesting in the area. The focus group discussion, village meetings and key informants interviews were conducted by a moderator and an assistant who prepared the list of discussion questions that guided the interviews, from which background information on households' adaptations and mopane worm collection and commercialisation were obtained. This information was then used to design a semi-structured questionnaire administered to respondents during the interview to obtain quantitative data. The questions focused on household demographic and socio-economic characteristics and institutions related to mopane worm collection and commercialisation. All respondents were asked to recall quantities harvested from the woodland and the respective amounts consumed/sold/bartered. To assist with field work and translate the English questionnaire to the respondents, local interviewers who had in-depth knowledge about mopane worm and were confident with the ethnic language (Northern Sotho and Tsonga) were used for the data collection. 2.3. Analytical methods 2.3.1. Determination of intensity of commercialisation Strasberg et al. (1999) suggested the intensity of participation in the output market as a conventional way to measure commercialisation and

thus measured commercialisation using the household commercialisation index (HCI). The HCI is defined as the ratio of the gross value of all crop (mopane worm) sales to gross value of all production (mopane worm harvested). HCI can take any value from zero to one and can also be expressed in percentage terms as: HCI ¼

GCS  100 GCP

ð1Þ

where GCS GCP

Gross value of crop sale Gross value of crop produced

2.3.2. Econometric estimation In empirical studies, econometric models applied to market participation in general typically adopt a two-step analytical approach. The reason being that commercialisation is seen to embody two decision processes: the unobservable decision to sell and the observed degree or extent (intensity) of commercialisation. The estimation of intensity of commercialisation is therefore dependent on whether the households do engage in mopane worms marketing or not. Given that not all of the sampled households will participate in markets in the reference period, this implies that when only such households are considered in estimating the intensity of participation, the parameter estimates will be inconsistent due to the sample selection bias, therefore making estimation by ordinary least squares inappropriate (Wooldridge, 2003). To address this concern, Heckman sample selection model (two-step version) is employed to correct for the possibility of bias due to sample selection as applied by Benfica et al. (2006) and Goetz (1992). It consists of two steps. First, a selection equation is estimated using a probit model of participation in the relevant market as a function of both those

144

L.J.S. Baiyegunhi, B.B. Oppong / Forest Policy and Economics 62 (2016) 141–148

variables that likely also determine crop sales volumes, conditional on market participation, as well as one or more exclusion restriction variables (Wooldridge, 2003). The second step is an OLS regression equation of the mopane worm sales plus the Inverse Mills Ratio (IMR) as a regressor derived from the first-stage probit regression. The probability of mopane worm commercialisation by a household is assumed to be determined by an underlying response variable that captures the true households' socio-economic characteristics. The underlying response variable y⁎ is defined by the regression equation: yi ¼ α 0 zi þ ei

ð2Þ

yi  is a latent level of utility the household gets from commercialisation, ei ~ N(0,1) and, yi ¼ 1 yi ¼ 0

if if

yi N0 yi ≤0

The variable yi takes the value of 1 if the marginal utility the household i get from mopane worm sale (commercialisation) is greater than zero, and zero otherwise. A probit model of mopane worms’ commercialisation which follows random utility (Wooldridge, 2002) can be specified as: Prðyi ¼ 1jzi ; α Þ ¼ Φðhðzi ; α ÞÞ þ ei

ð3Þ

where, yi is an indicator variable equal to 1 for households that sells mopane worms; zi is a vector of explanatory variables; α is a vector of parameters to be estimated; Φ is the standard normal cumulative distribution function; ei is a random error term assumed to be distributed normally with zero mean and unit variance σ 2. In the second step, the Inverse of Mills Ratio (IMR) is added as a regressor in the sales function regarding level of participation in order to correct for potential selection bias. The second-stage (sales) equation is then given by: EðGi jy ¼ 1Þ ¼ f ðxi ; βÞ þ γλ

ð4Þ

where, E is the expectation operator; G is the level of mopane worm commercialisation (the household commercialisation index - HCI); x is a vector of explanatory variables affecting sales; β is the vector of the corresponding coefficients to be estimated; λ is the inverse mills’ ratio to account for sample selection in supply analysis; γ is the associated parameter to be estimated. The inverse mills’ratio can be calculated as λ ¼




Variable

Description

Dependent COMMERCZ

Expected sign

Decision to sell mopane worm or not (1 if yes; 0 otherwise) Percentage of total output sold

HCI Explanatory variables Household socio-economic characteristics Age Gender

Age of the household head (years) Gender of the head (1 if head is female; 0 otherwise) Education of the head (years) Household size (adult equivalent) Household exogenous income (wages, pension, grants and remittances)

EDUC HHSIZE HHINC Output attributes Price Quantity Institutional variables INSTLAW SOCAPT Transaction costs variables DIST TRANS

+ +/− +/−

Average price of mopane worm (Rands/kg) Quantity of mopane worm harvested (kg)

+ +

Absence of institutional regulation/law (1 if yes; 0 otherwise) Number of groups/associations household head is a member

+

Distance to nearest market (km) Transport assets (1 if the head owns bicycles/motor cycles; 0 otherwise) Access to market information (1 if yes; 0 otherwise)

INFOR

− +

+

− + +

Based on a priori expectations.

attributes, institutional and transaction costs-related variables. The definitions for the variables used in the analysis are presented in Table 1.

3. Empirical results 3.1. Description of determinants of commercialisation decision and intensity of commercialisation The determinants of the decision to market and intensity of commercialisation, as well as the significance level of tests of difference Table 2 Description of determinants of commercialisation decision and intensity of commercialisation (n = 120).

ϕðhðzi ; α 0 ÞÞ Φðzi α 0 Þ

where, φ(⋅) is the normal probability density function and Φ is the cumulative distribution function. Therefore, Gi can be expressed as following: Gi ¼ β0 xi þ γ λi þ ui ; ui  N 0; σ 2

Table 1 Definitions of the variables used in the analysis.

ð5Þ

where ui is a random error term with zero mean and variance σ 2. Gi⁎ is only observed for market participants (yi = 1), in which case Gi = Gi⁎. OLS estimation of Eq. (4), inclusive of λ, produces consistent estimates, eliminating selectivity bias (Greene, 2003). 2.4. Explanatory variables used in the model The explanatory variables hypothesized to explain mopane worms' commercialisation and level of commercialisation were identified based on the theory of market participation and on past empirical work on market participation. The explanatory variables were classified into four categories: household socio-economic characteristics, output

Variables

Units

Seller Mean

Age Gender EDUC HHSIZE HHINC Price Quantity INSTLAW SOCAPT Distance Trans Infor Number of households

Years (1 = female; 0 = Male) Years Adult equivalent Rands Rands/kg Kilogram (1 = Yes; 0 = No) Numbers Kilometres (1 = Yes; 0 = No) (1 = Yes; 0 = No)

Non-sellers S.D

Mean

41.0 10.65 57.0 0.79 0.026 0.19 4.53 3.45 6.45 2.10 260.53 44.17 35.00 15.00 225 56.23 0.21 0.41 2.00 1.92 4.71 6.16 0.67 0.47 0.97 0.32 78

5.28 3.55 667.92 – 49.65 0.23 2.29 6.40 0.38 0.45 42

MT

S.D 12.10 ⁎⁎ 0.041 ⁎⁎⁎ 2.70 1.63 67.31 – 12.34 0.42 2.24 6.91 0.49 0.18

⁎⁎ ⁎⁎⁎ ⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎

Source: Calculated from field survey data (2013). Notes: MT1 = Test of difference between means of sellers and non-sellers: ⁎⁎⁎, ⁎⁎ and ⁎ denote variables are statistical significant at the 1%, 5% and 10% probability levels respectively. The exchange rate at the time of the survey was about 1US$ = 11.00 Rands.

L.J.S. Baiyegunhi, B.B. Oppong / Forest Policy and Economics 62 (2016) 141–148

between means for each determinant for sellers and non-sellers are presented in Table 2. The results presented in Table 2 shows that the average age of mopane worm seller is about 41 years. This means household engaged in mopane worm commercialisation in the area are relatively young and within the economically active population. However, there is a statistically significant difference in age of mopane worm sellers and nonsellers. Female-headed households participated more in mopane worm markets. It is also found that non-participants are more educated than participants. The average household size for the sample was 5; with mopane worm sellers having significantly larger household size per capita than non-sellers. Social grants, pension and remittances are the major source of household exogenous income with a monthly average of R1 755 ($160) per household. The average price received per kilogram of mopane worm sold was R35 ($3.18), while household average annual income from mopane worm commercialisation was about R21 060 ($1, 915)1; constituting about 30% of a household’s total monthly income. This value translates into $5.24 per day, which is above the $2.00 per day international poverty line. This implies that mopane worm commercialisation can make significant impact on household and national poverty statistics, preventing a deepening of poverty for millions, and alleviation of poverty for a large proportion of households involved in the mopane worm markets. The share of households having a membership in association was significantly higher for mopane worm sellers than non-sellers. About 86% of the market participants are members of associations or groups. A household belongs to an average of three groups, with two types predominating i.e. financial groups, including stokvels (saving clubs) and burial societies (which provide insurance for funeral costs) and religious groups. In general, a high percent of mopane worm non-sellers were located farther away from nearest market than sellers; thus, the average distance to the nearest market was significantly higher for non-sellers than sellers. The statistics also shows that majority (67%) of mopane worm sellers had their own transport assets for regular access to nearby towns/markets. Similarly, majority (97%) of mopane worm sellers also owned communication assets to gain access to market information. A majority of the respondents indicated that there are no institutions and laws regulating the use and management of the woodland resources and they were not aware of any current or future management strategies either at village or government level. As a result they considered the woodlands as a free gift of nature thereby treating it as an open access resource. Open-access resources are those that can be accessed by anyone at any time without restraint, which implies that the rules limiting individual use of the common pool resource are either absent or are not enforced (Tietenberg, 2006). About 65% of the sampled households participated in mopane worm market, with market participation level (commercialisation) of about 63% indicating a high rate of mopane worm commercialisation in the area.

3.2. Result of the econometric estimation 3.2.1. Specification tests Prior to estimating the selection model, the model was corrected for possible multicollinearity problems using the VIF (Variance Inflation Factor) and correlation matrix. The VIF was less than the critical value of 10 (Gujarati and Porter, 2009), confirming that multicollinearity was not a problem. The correlation coefficient is less than 0.5, indicating weak correlations, which suggest that multicollinearity is not a concern. To account for possible heteroscedasticity, a common specification error for cross-sectional data, the z-statistics and t-statistics reported in Table 3 are based on heteroscedasticity robust standard errors (White, 1980). 1

(US$1 = ZAR11 in 2014).

145

3.2.2. Mopane worm commercialisation decision and the intensity of commercialisation The results of the selection regression, which involved the probit analysis of the commercialisation decision and also the results of the underlying regression, which establishes the determinants of intensity of commercialisation are estimated jointly and are presented in Table 3. In order to explain the differential impact of explanatory variables on dependent variable, the coefficient estimates as well as the marginal effects (which represent changes in probability of participation) of the probit estimates are also presented in Table 3. The model used fits the data well, as 76% of the participation outcomes in mopane worm market is correctly predicted. The Wald test of the hypothesis that all regression coefficients are jointly equal to zero is rejected at the 1% significance level. This shows that jointly the independent variables included in the probit regression model explain the variations in the household’s probability to sell mopane worm. The coefficient estimates for second-stage Heckman selection estimation for the intensity of commercialisation are presented in Table 3 (Hurdle 2). The coefficients in Table 3 are elasticities conditional on participation because the double-log specification was used. The coefficient of the Inverse Mills Ratio (λ) is significantly different from zero (pb 0.05) for mopane worm market supply equations, indicating that sample selection bias would have resulted if the mopane worm supply equations had been estimated without consideration of the market participation decision. This means estimating jointly the equations for selection or decision to sell and the level of sales as it is indicated earlier is correct. In the analysis of the decision to market (Hurdle 1), eight variables were significant: gender, education, household size, quantity of mopane worm harvested, social capital, distance, transportation and information. The coefficients have expected signs. Gender, household size, quantity of mopane worm harvested, social capital and information have a statistically significant positive effect on the likelihood of market participation, whereas education and distance have a statistically significant negative effect on the likelihood of market participation. In the analysis of the intensity of commercialisation (Hurdle 2), eight variables were significant: age, gender, education, household income, price of mopane, quantity of mopane worm harvested, distance and information. The coefficients have expected signs. Gender, household income, price of mopane, quantity of mopane worm harvested and the absence of institution/law have a statistically significant positive effect on the intensity of commercialisation, whereas, age, education and transport have a statistically significant negative effect on the likelihood of market participation,

4. Discussion This study found that female-headed households were more likely to participate in mopane worm markets and supply more marketable surplus compared to male-headed households. This is probably because female-headed households are resource constrained, lacking access to productive assets (land, labour, capital) which limits their agricultural production capabilities, hence they rely more on forest products for subsistence and cash income, suggesting that mopane worm market is more accessible and/or attractive to female-headed households. Kipkemboi et al. (2007) found that female-headed households collect greater quantities of wetland products than their male counterparts. All being equal, increasing age of the household head reduces the intensity of commercialisation. A possible explanation for this could be that older household heads have accumulated more assets and tends to have higher reliance on crop and livestock income (Angelsen et al., 2014). In addition, older people may be less able physically to access forest resources, due to hardship associated with collection such as long distances to forests (Mujawamariya and Karimov, 2014; Mulenga et al., 2011).

146

L.J.S. Baiyegunhi, B.B. Oppong / Forest Policy and Economics 62 (2016) 141–148

Table 3 Estimates of the determinants of market participation and intensity of commercialisation of mopane worm. Double hurdle regression estimates Hurdle 1: Probability of market participation (probit regression)

Hurdle 2: Intensity of commercialisation (OLS regression with the IMR)

Variable

Coefficient

Change in probability

Coefficient

t-Statistics

Age Gender Educ HHSIZE HHINC Price Quantity Instlaw SOCAPT Distance TRANS INFOR IMR (λ) Constant

0.038 (0.751) 0.497⁎⁎⁎ (3.120) −0.075⁎⁎ (0.035) 0.166⁎⁎⁎ (3.130) 0.102 (0.920) 0.001 (2.242) 0.078⁎⁎ (0.035) 0.035 (0.282) 0.232⁎ (1.841) −0.019⁎⁎ (-2.361) 0.571⁎⁎ (2.501) 0.526⁎⁎ (0.268) – 1.315⁎⁎ (2.702)

0.015 0.160 −0.025 0.044 0.027 0.001 0.126 0.016 0.090 −0.007 0.139 0.090 –

−2.282⁎ (1.205) 0.192⁎⁎ (1.370) −0.723⁎⁎ (3.152)

−1.89 −0.14 −0.23 −0.08 2.28 0.15 1.97 2.63 0.36 0.84 −1.72 0.41 0.16 −2.01 2.33

No of Observation Wald χ2 (12) Pseudo R2 Log pseudo-likelihood Outcomes correctly predicted

0.030 (0.380) 0.458⁎⁎ (0.202) 0.385⁎⁎⁎ (2.63) 1.395⁎⁎ (0.706) 1.824⁎⁎⁎ (0.693) 0.292 (0.805) 0.214 (0.281) −0.185⁎ (0.092) 0.339 (0.834) 0.090 (0.569) −0.250⁎⁎ (0.124) 12.554⁎⁎ (4.452)

120 82.54 0.1320 −234.67 76%

Note: Robust standard error (z-statistics are in parentheses). ⁎⁎⁎, ⁎⁎ and ⁎ denote variables are statistical significant at the 1%, 5% and 10% probability levels respectively.

As expected, households with more years of education tend to have lower dependency of income from forest resources. This might reflect better opportunities for the households in the off-farm labor market. Advancement in education increases the opportunities for gainful employment and therefore improved household income and welfare. Education is expected to lead to increased earning potential and improve occupational and geographical mobility of labour (Adhikari et al., 2004) and it could also means better access to new information on sustainable natural resource use and management (Dolisca et al., 2006; Obua et al., 1998). Increases in education are associated with increased concern about the environment; it alters attitudes and behaviour towards pro-environmental conservation and protection (Ignatow, 2006; Dunlap et al., 2000). Larger households are also more likely to participate in mopane worm markets. Larger households are also likely to have higher consumer to worker ratios, and income per adult equivalent is therefore likely to be lower. Household size influences the demand for forest resources and hence increases dependency on forests (Agarwal and Angelsen, 2010). Considering the high labour demand for mopane worm harvesting, processing and marketing, large household size influences the quantity of labour available, allowing household to have a surplus production above the subsistence needs and for sale which influences market participation (Bahta and Bauer, 2012; Ur-Rehman and Chisholm, 2007). Although, households’ exogenous income (wages, pension, social grants and remittances) is expected to make mopane worm commercialisation less attractive activity, households with exogenous incomes were found to collect larger quantities and offer marketable surplus, because they could afford collection and processing costs and even the cost of transportation to better markets as shown by studies such as Getachew et al. (2007) and Muzayen (2009). Also increased household income in terms of wages, pension, social grants and remittances relaxes household liquidity constraints thereby enhancing large scale collection and processing of mopane worms. It increases households’ ability to hire labour, cover transaction and transportation costs (Abu et al., 2014). Studies in South Africa have shown that wealth influence the procurement and sale, with greater proportion of households procuring NTFPs through self-collection and sales either on full-time or on ad hoc basis (Cocks et al., 2008; Shackleton and Shackleton, 2006). Other studies from developing countries have also shown that middle to upper income

households with access to capital, transport and market, and with alternative ‘fallback’ options earn more from commercialisation of NTFPs as a primary livelihood activity (Ambrose-Oji, 2003; Angelsen and Wunder, 2003; Kepe, 2002). The collection of mopane worm, as like any other the collection of any other NTFPs in particular or any agricultural product in general, is attractive when collectors/producers expect to obtain a positive return for their effort. As a result, the price received from the previous season makes collection in the current season attractive because it creates the expectations for high profits. This means that marketed supply increases with price, once participation decisions are made. Higher mopane worm prices can be expected to lead to a higher volume of mopane worm sales because harvesting effort responds positively to expected prices. The high conditional elasticity of mopane worm suggests that its price is indeed a motivation for increasing marketed surplus among sellers. This is important as it should be sufficiently large to justify the trade-off between collection and other economic activities such as pastoralism or livestock rearing common in the study area. Studies have shown that output price is an incentive for sellers to market their surplus production (Omiti et al., 2009; Olwande and Mathenge, 2012). Form the discussion thus far, it is evident that the decision to participate in market and the intensity of participation depends on incentives and trade-offs. However, some factors reinforce the household’s decision either positively or negatively. One factor is the quantity of mopane worm harvested (Output). The study found that households that collect higher quantities participate in market and offer marketable surplus. A possible explanation could be since mopane worms are harvested freely in the wild, high quantities harvested could motivates households to sell surplus after subsistence needs have been met. It has been documented that the lack of responsibility in the management of woodlands, the diminished roles of tribal authorities in woodland management and lack of resources to effectively enforce conservation regulations has led to overharvesting and exploitation. It is unclear who among the traditional leaders, villagers and conservation officials at local level have the responsibility to manage the woodlands (Gugushe et al., 2008, Von Maltitz and Shackleton, 2004). Effectively, rural household continued to view woodland resource as a free-gift of nature. As expected, association can be good platforms for exchanging information. Group participation facilitates the spread of knowledge, information and innovations. At local level, insecurity caused by conflicts

L.J.S. Baiyegunhi, B.B. Oppong / Forest Policy and Economics 62 (2016) 141–148

over resources is not uncommon (Messer and Cohen, 2006). Insecurity implies that collectors cannot go into the forest unless they have some form of protection or are in groups. Hence, households that are members of group/association participate in mopane worm markets. Variables capturing transaction costs are proxy by distance to nearest market, ownership of communication assets and transport assets. As expected, there is a decline in volume of mopane worm sold with increasing distance further away from the market. A possible explanation is that the farther away a household is from the market, the less likely it will participate in the market, as it becomes more difficult and costly to sell large quantities of mopane worms compared those living closer to the local markets (Ouma et al., 2010). As expected, ownership of transportation and communication assets increases market participation. This suggests that better access to information and transportation is likely to result in increased mopane worm market participation (Heltberg and Tarp, 2002; Alene et al., 2008; Boughton et al., 2007).

5. Conclusions and policy implications Commercialisation of mopane worms is valuable to rural people in Limpopo Province, South Africa, considering the prevailing socioeconomic conditions in the area. Cash generated from sale of mopane worms helps alleviate poverty, improve livelihoods and allows participation of the rural people in a growing cash economy. However, high rate of unstainable commercial harvesting practices are common leading to the depletion of mopane woodland. It is expected that in view of increased rural population and market pressures, demand for and commercialisation of mopane worm will increase further, which in turn will degrade the status of the woodland in the near future; accordingly the rural households will face shortage of mopane worm supply which will have adverse effect on their livelihoods. Therefore, it is imperative to link rural development and NTFPs security strategies with forest management. Rights of access to the resource are not clearly defined and as the resource grows in value, the lack thereof may jeopardize current benefits to the rural people, thus a workable regulatory framework need to be put in place to secure these rights. Rural communities need to be supported to establish indigenous natural resource management systems/projects, this will increase rural households’ knowledge and perception of the benefit of sustainable resource use and management practices through better access to technical information, extension and training. This could be a step forward towards ensuring sustainable use of natural resources in rural areas and contribute to a better life for the majority of the unemployed rural dwellers. It is imperative for government and private stakeholders to appropriately find a better understanding of the nature and objectives of the existing social groups/networks in rural areas and use them for project designs and delivery. Conflict of interest The authors declare that they have no conflict of interest. Ethical clearance for this study (Ref No: HSS/0422/013M) was obtained from the University of KwaZulu-Natal Research Office.

Acknowledgments The College of Agriculture, Engineering and Science, University of KwaZulu-Natal (JP5956-48901) is acknowledged for funding this study through its postgraduate student bursary. We also thank the all the households and the key informants in the study areas that participated in the survey and focus group discussions.

References Abu, B.M., Osei-Asare, Y.B., Wayo, S., 2014. Market participation of smallholder maize farmers in the upper west region of Ghana. Afr. J. Agric. Res. 9 (31), 2427–2435.

147

Adhikari, B., Di Falco, S., Lovett, J.C., 2004. Household characteristics and forest dependency: evidence from common property forest management in Nepal. Ecol. Econ. 48 (2), 245–257. Agarwal, B., Angelsen, A., 2010. Using community forest management to achieve. REDD+ goals in realizing REDD (Available at:b National….sitemaker.umich.edu/ifri/files/ifri_ may_2010_newsletter.pdfN [accessed on 23 September 2014]). Alene, D.A., Manyong, V.M., Omanya, G., Mignouna, H.D., Bokanga, M., Odhiambo, G.D., 2008. Smallholder marketed surplus and input use under transactions costs: maize supply and fertilizer demand in Kenya. Food Policy 32 (4), 318–328. Aliber, M., 2003. Chronic poverty in South Africa: incidence, causes and policies. World Dev. 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. Int. For. Rev. 5 (2), 106–117. Angelsen, A., Wunder, S., 2003. Exploring the forest-poverty link: key concepts, issues and research implications. CIFOR Occasional Paper No. 40. Centre for International Forestry Research (CIFOR), Bogor, Indonesia. Angelsen, A., Jagger, P., Babigumira, R., Belcher, B., Hogarth, N.J., Bauch, S., Borner, J., Smith-Hall, C., Wunder, S., 2014. Environmental income and rural livelihoods: a global-comparative analysis. World Dev. 64, S13–S28. Areki, F., Cunningham, A.B., 2010. Fiji: commerce, carving and customary tenure. In: Laird, S.A., McLain, R.J., Wynberg, R.P. (Eds.), Wild Product Governance: Finding Policies That Work for Non-timber Forest Products. Earthscan, London, pp. 229–242. Asfaw, A., Lemenih, M., Kassa, H., Ewnetu, Z., 2013. Importance, determinants and gender dimensions of forest income in eastern highlands of Ethiopia: the case of communities around Jelo Afromontane forest. Forest Policy Econ. 28, 1–7. Babulo, B., Muys, B., Nega, F., Tollens, E., Nyssen, J., Deckers, J., Mathijs, E., 2009. The economic contribution of forest resource use to rural livelihoods in Tigray, Northern Ethiopia. Forest Policy Econ. 11, 109–117. Bahta, S., Bauer, S., 2012. Policy options for improving market participation and sales of smallholder crop producers: a case study of the Free State Province of South Africa. Afr. J. Agric. Res. 7 (24), 3525–3533. Baiyegunhi, L.J.S., Fraser, G.C.G., 2011. Vulnerability and poverty dynamics in rural areas of Eastern Cape Province, South Africa. Ghana J. Dev. Stud. 8 (2), 84–100. Belcher, B., Ruiz-Perez, M., Achidiawan, R., 2005. Global patterns and trends in the use and management of commercial NTFPs: implications for livelihood and conservation. World Dev. 33 (9), 1435–1452. Benfica, R., Tschirley, D., Boughton, D., 2006. interlinked transactions in cash cropping economies: the determinants of farmer participation and performance in the Zambezi River Valley of Mozambique. Paper Presented at the 26th International Association of Agricultural Economists Conference, Gold Coast, Australia, August 12-18, 2006. Boughton, D., Mather, D., Barrett, C.B., Benfica, R., Abdula, D., Tschirley, D., Cunguara, B., 2007. Market participation by rural households in a low-income country: an assetbased approach applied to Mozambique. Faith Econ. 50, 64–101. Cavendish, W., 2000. Empirical regularities in the poverty–environment relationship of rural households: evidence from Zimbabwe. World Dev. 28, 1979–2003. CIFOR, 2011. Forests and non-timber forest products. CIFOR fact sheets (Available at: bhttp://www.cifor.org/publications/corporate/factSheet/NTFP.htmN. [assessed on 24 August 2015]). Cocks, M.L., Bangay, L., Shackleton, C.M., Wiersum, K.F., 2008. ‘Rich man poor man’ — interhousehold and community factors influencing the use of wild plant resources amongst rural households in South Africa. Int. J. Sustain. Dev. World Ecol. 15, 198–210. Cocks, M., López, C., Dold, T., 2011. Cultural importance of non-timber forest products: opportunities they pose for bio-cultural diversity in dynamic societies. In: Shackleton, S.E., Shackleton, C.M., Shanley, P. (Eds.), Non-timber Forest Products in the Global Context. Springer, Heidelberg, pp. 107–128. Cocksedge, W., 2006. Incorporating non-timber forest products into sustainable forest management: an overview for forest managers. Royal Roads University, Victoria (Available at: bhttp://cntr.royalroads.caN. [accessed on 24 August 2015]). Cosyns, H., Degrande, A., de Wulf, R., van Damme, P., Tchoundjeu, Z., 2011. Can commercialisation of NTFPs alleviate poverty? A case study of Ricinodendron heudelotii (Baill.) Pierre ex Pax. Kernel marketing in Cameroon. J. Agric. Rural. Dev. Trop. Subtrop. 112, 45–56. D’Hase L, Vermeulen H. 2011. The food security status of Limpopo Province: archive of interdisciplinary discussion groups and past events. Available at: bhttp://blogs.sun.ac. za/fsi/files/2010/10/Presentation-food-security-limpopofinal1.pdfN (accessed on 15 August 2012). Dolisca, F., Carter, D.R., McDaniel, J.M., Shannon, D.A., Jolly, C.M., 2006. Factors influencing farmers' participation in forestry management programs: a case study from Haiti. For. Ecol. Manag. 236 (2), 324–331. Dovie, D.B.K., 2003. Rural economy and livelihoods from the non-timber forest products trade. Compromising sustainability in southern Africa? Int. J. Sustain. Dev. World Ecol. 10 (3), 247–262. Dovie, D.B.K., Witkowski, E.T.F., Shackleton, C.M., 2005. Monetary valuation of livelihoods for understanding the composition and complexity of rural households. Agric. Hum. Values 22, 87–103. Dunlap, R., Van Liere, K., Mertig, A., Jones, R., 2000. Measuring endorsement of the new ecological paradigm: a revised NEP scale. Promoting Environ. 56 (3), 425–442. Frost, P.G.H., 2005. A Guide to Sustainable use of Mopane Worms. Institute of Environment Studies, Harare: University of Zimbabwe. Getachew, M., Sjaastad, E., Vedelbd, P., 2007. Economic dependence on forest resources: a case study from Dendi District, Ethiopia. Forest Econ. Policy 9, 916–927. Ghazoul, J., 2006. Mopane woodlands and the Mopane worm: enhancing Rural Livelihoods and Resource Sustainability. Final Technical Report. DFID, London, pp. 1–119. Goetz, S.J., 1992. A selectivity model of household food marketing behavior in SubSaharan Africa. Am. J. Agric. Econ. 74 (2), 444–452.

148

L.J.S. Baiyegunhi, B.B. Oppong / Forest Policy and Economics 62 (2016) 141–148

Gondo, T., Frost, P., Kozanayi, W., Stack, J., Mushongahande, M., 2010. Linking knowledge and practice: assessing options for sustainable use of mopane worms (Imbrasia belina) in Southern Zimbabwe. J. Sustain. Dev. Afr. 12 (4), 127–145. Greene, W., 2003. Econometric Analysis. 5th ed. Prentice Hall, Upper Saddle River, New Jersey. Gugushe, N.M., Grundy, I.M., Theron, F., Chirwa, P.W., 2008. Perceptions of forest resource use and management in two village communities in the Eastern Cape Province, South Africa. Southern Forests: a Journal of Forest Science 70(3), pp. 247–254. Gujarati, D.N., Porter, D.C., 2009. Basic Econometrics. 5th ed. McGraw-Hill/Irwin, New York. Heltberg, R., Tarp, F., 2002. Agricultural supply response and poverty in Mozambique. Food Policy 27 (1), 103–124. Heubach, K., Wittig, R., Nuppenau, E.-A., Hahn, K., 2011. The economic importance of nontimber forest products (NTFPs) for livelihood maintenance of rural West African communities: a case study from northern Benin. Ecol. Econ. 70, 1991–2001. Hope, R.A., Frost, P.G.H., Gardiner, A., Ghazoul, J., 2009. Experimental analysis of adoption of domestic mopane worm farming technology in Zimbabwe. Dev. South. Afr. 26 (10), 29–46. Ignatow, G., 2006. Cultural models of nature and society: reconsidering environmental attitudes and concern. Environ. Behav. 38 (4), 441–461. Kamanga, P., Vedeld, P., Sjaastad, E., 2009. Forest incomes and rural livelihoods in Chiradzulu District, Malawi. Ecol. Econ. 68, 613–624. Kepe, T., 2002. Grassland vegetation and rural livelihoods: a case study of resource value and social dynamics on the wild coast, South Africa PhD. Thesis University of the Western Cape, Cape Town. Kim, I.A., Trosper, R.L., Mohs, G., 2012. Cultural uses of non-timber forest products among the Sts'ailes, British Columbia, Canada. Forest Policy Econ. 22, 40–46. Kipkemboi, J., Van Dam, A.A., Ikiara, M.M.D., 2007. Integration of smallholder wetland aquaculture-agriculture systems (fingerponds) into riparian farming systems on the shores of Lake Victoria, Kenya: socio-economics and livelihoods. Geogr. J. 173 (3), 257–272. LDLGH, 2007. Districts and Municipalities. (Available online at: bhttp://www.limpopodlgh.gov.zaN [accessed on 8 May 2012]). Makhado RA. 2008. Mopane wood utilization and management perceptions of rural inhabitants in the Greater Giyani municipality, Limpopo Province. Unpublished MSc Thesis. University of Limpopo, Polokwane, South Africa. Makhado, R.A., Von Maltitz, G.P., Potgieter, M.J., Wessels, D.C., 2009. Contribution of woodland products to rural livelihoods: in the northeast of Limpopo Province, South Africa. S. Afr. Geogr. J. 91 (1), 46–53. Makhado, R.A., Potgieter, M.J., Wessels, D.C., Saidi, A.T., Masehela, K.K., 2012. Use of mopane woodland resources and associated woodland management challenges in rural areas of South Africa. Ethnobot. Res. Appl. 10, 369–379. Makhado, R.A., Potgieter, M.J., Timberlake, J., Gumbo, D., 2014. A review of the significance of mopane products to rural people's livelihoods in southern Africa. Trans. R. Soc. S. Afr. 69 (2), 117–122. Mapaure, I., 1994. The distribution of Colophospermum mopane (Leguminosaecaesalpinioideae) in Africa. Kirkia 15, 1–15. Messer, E., Cohen, M.J., 2006. Conflict, food insecurity and globalization. FCND Discussion Paper 206. IFPRI, Washington DC: USA. Moresello, C., Ruiz-Mallén, I., Diaz, M.D.M., Reyes-Garcıá, V., 2012. The effects of processing non-timber forest products and trade partnerships on people's well-being and forest conservation in Amazonian societies. PLoS One 7, e43055. Mucina, L., Rutherford, M.C., 2006. The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19. Pretoria, South African Biodiversity Institute. Mujawamariya, G., Karimov, A.A., 2014. Importance of socio-economic factors in the collection of NTFPs: the case of gum Arabic in Keya. Forest Policy Econ. 42, 24–29. Mulenga, B.P., Richardson, R.B., Mapemba, L., Tembo, G., 2011. The contribution of nontimber forest products to rural livelihood in Zambia. FSRP Working Paper No.54. Mutenje, M.J., Ortmann, G.F., Ferrer, S.R.D., 2011. Management of non-timber forestry products extraction: local institutions, ecological knowledge and market structure in south-eastern Zimbabwe. Ecol. Econ. 70, 454–461. Muzayen SF. 2009. The role of non-timber forest products to rural livelihoods and forest conservation: a case study at Harana Bulluk District Oromia National Regional State. Ethiopia. Unpublished MSc Thesis. Wondo Genet College of Forestry and Natural Resources: Wondo Genet. Nattrass, N., 2004. Unemployment and AIDS: the social-democratic challenge for South Africa. Dev. South. Afr. 21, 87–108. Obua, J., Banana, A.Y., Turyahabwe, N., 1998. Attitudes of local communities towards forest management practices in Uganda: the case of Budongo forest reserve. Commonwealth Forestry Review 77(2), pp. 113–118. Olwande, J., Mathenge, M., 2012. Market participation among poor rural households in Kenya. The International Association of Agricultural Economists (IAAE) Triennial Conference, Foz do Iguaçu, Brazil, 18-24 August 2012. Omiti, J.M., Otieno, D.J., Nyanamba, T.O., McCullough, E., 2009. Factors influencing the intensity of market participation by smallholder farmers: a case study of rural and periurban areas of Kenya. Afr. J. Resour. Econ. 3 (1), 57–82.

Ouma, E., Jagwe, J., Aiko, G.O., Abele, S., 2010. Determinants of smallholder farmers' participation in banana markets in Central Africa: the role of transaction costs. Agric. Econ. 41 (1), 111–122. Papola, T.S., 2000. Growth, poverty alleviation and sustainable resource management in the mountain areas of South Asia. Proceedings of the international Conference, Kathmandu, Nepal, 31 January–4 February, 2000, p. 590. Paumgarten, F., Shackleton, C.M., 2009. Wealth differentiation in household use and trade in non-timber forest products in South Africa. Ecol. Econ. 68, 2950–2959. Paumgarten, F., Shackleton, C.M., 2011. The role of non-timber forest products in household coping strategies in South Africa: the influence of household wealth and gender. Popul. Environ. 33, 108–131. Potgieter, M.J., Makhado, R.A., Potgieter, A., 2012. Mopane worms: assessing the potential of insects as food and feed in assuring food security. Technical Consultation Meeting, 23–25 January 2012. FAO, Rome, Italy. Pouliot, M., 2012. Contribution of “women’s gold” to West African livelihoods: the case of shea (Vitellaria paradoxa) in Burkina Faso. Econ. Bot. 19, 491–501. Rasul, G., Karki, M., Sah, R., 2008. The role of non-timber forest products in poverty reduction in India: prospects and problems. Dev. Pract. 18, 779–788. Rizek, M.B., Morsello, C., 2012. Impacts of trade in non-timber forest products on cooperation among Caboclo households of the Brazilian Amazon. Hum. Ecol. 40, 707–719. Saha, D., Sundriyal, R.C., 2012. Utilisation of non-timber forest products in humid tropics: implications for management and livelihood. Forest Policy Econ. 14, 28–40. Scherr, S.J., White, A., Kaimowitz, D., 2004. A new agenda for forest conservation and poverty reduction. Making markets work for low income producers. Forest Trends and Center for International Forestry Research (CIFOR), Washington, D.C. Shackleton, S.E., Gumbo, D.J., 2010. Contribution of non-wood forest products to livelihoods and poverty alleviation. In: Chidumayo, E.N., Gumbo, D.J. (Eds.), The dry Forests and Woodlands of Africa: Managing for Products and Services. Earthscan, London, pp. 63–92. Shackleton, C.M., Pandey, A.K., 2014. Positioning non-timber forest products on the development agenda. Forest Policy Econ. 38, 1–7. 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. S. Afr. J. Sci. 100 (11&12), 658–664. Shackleton, C.M., Shackleton, S.E., 2006. Household wealth status and natural resource use in the Kat River valley, South Africa. Ecol. Econ. 57, 306–317. Shackleton, C.M., Shackleton, S.E., Buiten, E., Bird, N., 2007. Importance of dry woodlands and forest in rural livelihoods and poverty alleviation in South Africa. Forest Policy Econ. 9, 558–577. Shackleton, S.E., Campbell, B., Lotz-Sisitka, H., Shackleton, C.M., 2008. Links between the local trade in natural products, livelihoods and poverty alleviation in a semi-arid region of South Africa. World Dev. 36, 505–526. Shackleton, C.M., Delang, C., Shackleton, S.E., Shanley, P., 2011. Non-timber forest products: concept and definition. In: Shackleton, S.E., Shackleton, C.M., Shanley, P. (Eds.), Non-timber Forest Products in the Global Context. Springer, Heidelberg, pp. 3–21. StatisticsSA. 2011. The South Africa I Know, the Home I Understand. Available at: b http:// beta2.statssa.gov.zaN (accessed on 19 January 2015). Strasberg, P.J., Jayne, T.S., Yamano, T., Nyoro, J., Karanja, D., Strauss, J., 1999. Effects of agricultural commercialisation on food crop input use and productivity in Kenya. Michigan State University, International Development Working Papers No. 71. Caxton Magazines. Johannesburg, South Africa, Michigan, USA. Styles, C.V., 1994. The big Value in Mopane Worms: Farmer's Weekly 22 July. pp. 20–22. Thang, T.N., Shivakoti, G.P., Inoue, M., 2010. Changes in property rights, forest use and forest dependency of Katu communities in Nam Dong District, Thua Thien Hue Province, Vietnam. Int. For. Rev. 12, 307–319. Tietenberg T. 2006. Open access resource. Available online at bhttp://www.eoearth.org/ view/article/155050N (accessed on 8 January 2015). Ur-Rehman, M., Chisholm, N., 2007. Factors affecting households' participation in the natural resource management. Activities in district Abbottabad, Pakistan: a multivariate analysis. J. Asian Afr. Stud. 42, 495–516. Von Maltitz, G.P., Shackleton, S.E., 2004. Use and management of forests and woodlands in South Africa. In: Lawes, M.J., Eeley, H.A.C., Shackleton, C.M., Geach, B.G.S. (Eds.), Indigenous Forests and Woodlands in South Africa: Policy, People and Practice. University of KwaZulu-Natal Press, Pietermaritzburg, pp. 109–138. White, H., 1980. A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica 48 (4), 817–838. Wooldridge, J.W., 2002. Econometric Analysis of Cross Section and Panel Data. MIT Press, Cambridge, Massachusetts. Wooldridge, J.W., 2003. Econometric Analysis of Cross Section and Panel Data. MIT Press, USA.