Influence of Social Capital on the Poverty Reduction of Rangeland Communities in the Bamyan Province, Afghanistan

Role of Rangelands and Forests in Afghanistan

Rangeland in the most part of the globe has economic (Knight, 2007), environmental, and cultural values (McLeod, 2011; Cameron et al., 2014). About 5-billion-hectare rangeland cover (Joshi et al., 2013) support 1 to 2 billion people (Sayre et al., 2013; Hoekstra et al., 2005). Variant product of rangeland is providing wildlife habitat (Christensen et al., 2008; Hobbs et al., 2008), water supply (Havstad et al., 2007), carbon sequestration (Derner and Schuman 2007; Henderson et al., 2015), forage and fodder, storage of precipitation, decrease water flow, declining drought impact, and goods and various ecosystem services (Joshi et al., 2013). Australia continent is prevailing by rangeland, both arid and semi-arid rangeland cover 70%, compose 25% of landscape of earth and have a backing of more than 20 million people’s subsistence (Chen and Gillieson, 2009). Climate variability, global warming, land degradation and carbon cycle can cause change vegetation cover, terrestrial ecosystem and environmental activity (Liu and Kafatos, 2005; Mansour et al., 2012). The native grazing land as Savanna and grassland has been decreased from 3,200 million ha to 1,700–2,700 million ha in 1990, globally (Lambin et al., 2003). The great economic value of rangeland and rainforest support millions of people’s livelihood and provide mankind’s good and services in developing countries. Both ecosystems encompass medicinal crops, timber, pasture crops, recreational opportunities and rangeland ecosystem is major source of rearing livestock and animal’s food product as food, milk and blood, manure product such as fuel and fertilizer, wool, hides (Williams and Allred, 1968; Mannetje, 2013). Rangeland is occupying 65% of total of land five Central Asian countries i.e. Kazakhstan, Kyrgyzstan, Tajikistan, Turkmenistan and Uzbekistan (Gintzburger, 2005), grazed land is the main source of subsistence’s pastoral and agro-pastoral communities (Mirzabaev et al., 2016). Grazing supports feeding animal husbandry, that provide 25–40% meat production, 30% milk production of small ruminant, 22% beef production in Latin America and 55% in Oceania (Boone et al., 2018).

The rangeland cover of Hindu Kush Himalaya (HKH) of Afghanistan, Bhutan, India, China, Nepal and Pakistan contains 54% of 4.2 million km² HKH region, and those lands are used as the main source of livelihood for mountain communities through agroforestry practices (Zakharenka and Rasuli, 2018). Rangelands have a usefulness for rearing animal husbandry during warm season, as fodder on cold season and fuel wood whole of year mostly in rural areas of Afghanistan (Ali and Shaoliang, 2013). Afghanistan’s land surface predominant by rangeland ecosystem (Bedunah et al., 2010), encompassing 60 to 75% of 63,000,000 ha land territory (Robinett et al., 2008; Ali and Shaoliang, 2013), and various services such as firewood, building material, medicinal crops, wildlife habitat and livestock products is supplying by grazing ecosystem (Stanfield et al., 2008; Miller, 2012). Rangeland regulates water resources, nourishes approximately 4 million ha irrigated land, and also nomadic and rural communities’ livelihood entirely or partially depending on rangeland (Ismail et al., 2009). Bamyan is a mountainous province that is hotbed of great ecological conditions with various vegetation types. Rangeland vegetation exhibiting rich biodiversity provides economic value to residents, ecological and culture services to native dwellers and tourists.

Rangelands and forests support Afghanistan’s rural economy by providing essential resources, including forage, water, fuel, timber, and non-timber forest products, whereas those rangelands in Afghanistan are not managed well and degraded (Zakharenka and Rasuli, 2018). Degraded rangelands can be rehabilitated either by grazing management or plantation of drought tolerant fodder species with suitable technologies and sites. A long-time conflict and turmoil in Afghanistan since the early 1970s have caused loss of life, insecurity, poverty, ethnic division, and damage to the environment and natural resources, which contributes to the livelihoods of most Afghanistan rural population. The rate of poverty have dramatically risen from 38 % (2011–2012) to 55 % (2016–2017) in Afghanistan. In particular, most people in rangeland areas are poor, and have the lowest per-capita consumption (Zakharenka and Rasuli, 2018).

Sustainable management of rangeland natural resources get a priority for the Government of Afghanistan and is an integral part of its poverty reduction and sustainable development. Because rangeland degradation is caused by socio-economic factors, rangeland management is deeply associated with national development programs, such as private sector development, justice sector reform, agriculture development, mineral and resource development, and human capital development (Roudgarmi, 2013). A well-designed rangeland policy and implementation strategies are necessary for conservation and utilization of the rangeland resources with a sustainable manner (Ahmad et al., 2012).

Study area

Bamyan province is located in the central of Afghanistan, with Hindu Kush throughout Afghanistan (Fig. 1), lies on 33.91° to 35.48° N latitude, 66.28° to 68.28°E longitude and covers 180,292 square kilometers. Eight provinces sieged Bamyan province; Samangan in the North, Baghlan in the northeast, Sari-Pul in the northwest, Parvan and Wardak in the east, Ghor in the West, Daykundi in the southwest and Ghazni in the southeast. The province is separated into 8 districts, and Bamyan town is the center of Bamyan province. Bamyan center and Yakawlang with ancient heritage and beauty landscapes attract tourists during warm seasons, and hotbeds of rich natural resources are the first priorities for researchers and scholars. Topographically, Bamyan with high mountains experienced harsh winter and warm and arid season in summer (Cook, 2011), and mountain covered 93.7% area (77.5% mountainous region and 16.1% area is semi mountainous) is used as rangeland, the 14% area suitable for cultivation with mild slopes (MRRD, 2013; Mohibbi, 2018).

Fig. 1

Study area.

Mostly, the annual income of residents is under 500 USD per capita, and approximately 90% of the population live in rural areas, where they are engaged in agriculture and agroforestry (e.g., agro-pastoralists). Dwellers are obtaining their revenues both agriculture and animal husbandry, and agriculture in Bamyan province is prevailing in traditionally with lowest productivity among other provinces of Afghanistan. Based on demand market values of animals’ production, the livestock rearing has been raising, resulting in the farmers alter cultivation farm into the pasture and rangeland is being degraded due to overgrazing and collected fuel wood in both study areas. The study was conducted in Band-E-Amir National Park (BANP) and Shah Foladi protected areas.

Band-E-Amir National Park (BANP) is located in Yakawlang district (Fig. 1), is sieged by beauty perspective were known first National Park by government of Afghanistan in 2009. Vegetation cover spread in the whole area of BANP, and most subsistence of residents was agro-pastoralists (Mohibbi and Cochard, 2014). BANP covers 61,330 ha, 55 km west of Bmayan town (BAPAC, 2011). The characterized flora of BANP rely on high altitude steppe plains and deserts (Shank and Larsson, 1977). The vegetation cover mainly is predominant by Acantholimon Spp., Artemisia Spp., Cousinia Spp (MAIL, 2008; Shank and Larsson, 1977)., Astragalus Spp. and Haloxylon Spp. (Bedunah et al., 2010).

Shah Foladi in central Bamyan province (Fig. 1), located in Hindu Kush with landscape outstanding natural beauty. Shah Foladi covers 700 square kilometer, form parts of Koh-i-Baba, in 2020 was known as a new protected area by the government of Afghanistan (UNEP, 2020). The North and South are separated by Koh-i-Baba mountain range with peak point 5050 m a.s.l. Shah Foladi is glaciated landscape area, glaciated lake, Pyramid peaks, and U-shaped valleys. The protected area is the most important watershed of Afghanistan that originates Amu Darya River catchment, Balkhab River catchment and Helmand River catchment (Ritchie and Fitzherbert, 2008). Mostly, there is extensive grazing land, desert, semi-desert and Steppe Mountain in the study area. The dominant vegetation cover type is a cushion bush-land with diversity of species of Acantholimon, Artemisia, Astragalus, Cousinia, Ephedra, and Onobrychi in Koh-i-Baba range (Breckle, 2007; Hassanyar, 1995). Rangeland plays a major role in the economic value of Afghanistan (Zakharenka and Rasuli, 2018). Particularly in the Bamyan province, rangeland ecosystem directly supports animal husbandry, wildlife and serves as major biodiversity conservation. The herders and farmers consume animals’ products as milk, yogurt, curt, wool, meat and mostly they sell livestock and provide food for their families. Animal waste is used as fertilizer in the agriculture system. Rural residents used charcoal as fuel wood and heating stove, also they collected the fodder for feeding animals during cold seasons, and collected bush and shrubs are utilized as fuel wood.

Data collection

Band-E-Amir National Park (BANP) in Yakawlang district and Shah Foladi in Bamyan Center were chosen as study areas, purposively due to high reliance on rangeland values (Mohibbi and Cochard, 2014), where steppe vegetation is degraded because of intensive grazing and collected firewood (Mohibbi, 2018). Based on objectives of economic values, rangeland species, pastoral and agro-pastoralist communities were selected as respondents in both study areas. To find out the assessment of rangeland-based economics, a cross-sectional study was conducted by a questionnaire at household level. This study was done in November and December of 2020. According to consultation the Council Development Association (CDA) and lists were prepared, the respondents were selected randomly. A total of respondents was 186 (93 interviewees from BANP and 93 interviewees from Shah Foladi protected area), after excluding missing data and outliers.

According to objectives of research, the semi-structured questionnaire provided reliance on indicators consisting direct values of dominant growing rangeland species, selling animals, selling animals’ products and non-use value of rangeland growing species. These indicators determine target herder-farmers who have attached to rangeland species and livestock rearing and relied on the cover of vegetation. A structured questionnaire was improved and reformed as driver of data collection. Pre-tested questionnaire survey has been done during modification and amendment and raised the validation and dependability. Text of the questionnaire consisted general information (gender, age, education, marital status, occupation, household size and annual income), value of rangeland species and social capital. The questionnaire panel was experts and professors from Seoul National University (SNU) and Yeungnam University of South Korea under Forest Restoration to Enhance Ecosystem Services in Afghanistan (FREESIA) project, fully funded by the government of South Korea. The final structured questionnaires have been used with suggestions and approval of the panel team.

Research variables

The explanatory variables used in this study consist of social capital socioeconomic characteristics, and forest conservation activities (Table 1). Social capital is a multi-dimensional and pluralistic complex entity (So, 2004). The World Bank has developed the Integrated Questionnaire for the Measurement of Social Capital (SC-IQ) to measure the multilateral aspects of social capital (Grootaert and Van Bastelar, 2002; Grootaert et al., 2004). The SC-IQ model is a useful survey tool for microscopic analysis of social capital, targeting individuals or households in the community; furthermore, it is theoretically robust with cases undertaken in various countries (Narayan and Pritchett, 1997; Grootaert, 1999; Diawara, 2013; Ahmad et al., 2012; Lee et al., 2017).

Table 1

Research variables

In this study, the four social capital variables comprised network, trust, cooperation, and social inclusion, based on the SC-IQ model (Grootaert et al., 2004). Network was measured by the number of organizations which the respondents belonged to, and the frequency of activities in these meetings and organizations. Trust was measured by the levels of trust on general people and among neighbors. Cooperation was measured by the experiences on collective activities and cooperative attitude of neighbors. Social inclusion was measured by the perceived feeling safe from crime in village and perceived feeling happiness in general days These four social capital indicators were synthesized to measure the total level of social capital. Socioeconomic characteristics were measured by a period of residence in the village, the number of members in a household (HH), education background of HH leader. Rangeland management activities were measured by income of rangeland species, income of livestock rearing, other income sources. The dependent variable was set by a household’s level of poverty represented by household consumption levels, considering the challenge of accurately measuring income levels by household (Grootaert, 1999; Ahmad et al., 2012). The average GDP per capita in Afghanistan was 513.11 USD (183th of the World) in 2019. In this study, households with a daily income below 1.08 USD were categorized as “poor” (World Bank, 2019).

Data Analysis and Measurement

When it comes to measuring a composite index, variables, which are measured in different units, need to be normalized (Jeong and Kim, 2019). Prior to measurement, we normalized those variables to resolve the inconsistency in terms of the unit of measurement and set a value between 0 and 1. An exploratory factor analysis was conducted to verify our model. We used principal component analysis to extract constituent factors, and a Varimax to simplify factor loading. The eigenvalue was over 1.0, and the factor loading was over 0.60. Some measured variables were eliminated through the scale purification process.

Table 2 shows the result of factor analysis of social capital. The total variance explained was 71.14%, and social capital was extracted from four factors: network, trust, cooperation, and social inclusion. We conducted a reliability analysis through sub-factors to examine the precision of the measurement tool, and Cronbach’s alpha (α) to measure reliability. Generally Cronbach’s alpha score 0.4–0.6 is fair, 0.6–0.75 is acceptable, 0.75–1.00 is excellent (Rosner, 2006). The results show that the reliability of exploratory factors such as trust and social inclusion are excellent, and other exploratory factors are also reliable fairly.

Table 2

Results of exploratory factor analysis

For regression analyses, we built a logit regression model to investigate the impact of social capital, socioeconomic characteristics, and rangeland production activities on the poverty of local communities. The logit model links the cause or stimuli of action with the result or response and quantitatively measures the effect of the cause or stimulus in the process of measuring the extent or result of a reaction. This logit model is easy to statistically process and can produce statistically significant results with very few samples. When the reaction has binary results, the binary logit model can be applied (Yu and Lym, 1992). Using the binary logit model, we estimated the parameters of an empirical model (Equation 1), where the dependent variable, poverty, is determined by independent variables such as social capital, education, household size, land tenure, and income from rangelands (Fig. 2) (Grootaert, 1999; Ahmad et al., 2012).

Fig. 2

Research framework.

where:

P(Y) = poverty of household, SC_i = social capital, X_i = socioeconomic characteristics by household, Z_i= rangeland production activities, μ = error term.

Social capital based on general social characteristics

The amount of social capital in the study area was estimated according to socioeconomic characteristics that include residence area, educational background, rangeland income activities, and poverty status (Table 3). Poor households (.728) were found to have lower social capital than non-poor households (.782). This result is in line with the existing theory that the higher the social capital level, the higher the living standard and social stability, confirming a significantly positive relationship between social capital and household income levels (Zack and Knack, 2001).

Table 3

Level of social capital by household characteristics

Also, depending on residential areas, respondents have a significant difference in terms of the level of social capital. People who live in the Shah Foladi protected area (.687) have lower social capital than those of the Band-E-Amir national park area (.800). Regional features and residential places affect the potential transition of communities’ social capital, because social ties or norms formed by organizational activities in the region establish network and trust within the group and enhance social capital levels (Pretty, 2003; Lee et al., 2017).

The type of rangeland tenure has a significant effect on the difference in social capital. Respondents with a government tenure (.801) were more likely to have higher social capital than those with a public tenure (.608), which implies that people with a government tenure may have a strong social capital because they manage their rangeland collectively with a stable and consistent system. On the other hand, people with a public tenure system may not cooperate with each other for managing rangelands due to instable tenure system. The results provide a consistent perspective with Hardin (1974) who suggest that public lands can be devastated with a public ownership, so privatization and government coercion are required to protect and manage those public lands. To maintain the rangeland systems sustainable, government tenure system may work better than public tenure system in the study area.

Social capital and poverty

Social capital contributes to the overall productivity of the economy by reducing transaction costs (Knack and Keefer, 1997; Zak and Knack, 2001; Putnam, 2000; Jeong and Kim, 2019). Higher levels of social capital may lead to lower transaction costs and a more competitive society. Furthermore, with more social capital, countries can achieve higher standards of living and social stability. Based on these hypotheses, we examined the relationships among social capital, household characteristics, rangeland production activities, and poverty, where poverty represent the livelihood of households with the minimum level of income per capita in Afghanistan (i.e., 1$(USD)/day), to identify social capital’s influence on economic activities.

To estimate the impact of social capital on the poverty level of local communities, we conducted a regression analysis using a logit model. Results in Table 4 show that social capital has a negative and statistically significant impact on the poverty of rangeland households, which is consistent with previous studies that suggest that social capital affect the poverty reduction of local communities (Ahmad et al., 2012). The analysis on marginal effects of social capital revealed that a unit increase in the social capital of local households reduce the likelihood of poverty by 6%. This indicates that those who participate more in community activities and have more neighbors to rely on are less likely to be poor.

Table 4

Effects of social capital and socio-economic characteristics on poverty

The number of household members showed a positive relationship with poverty at a statistically significant level. The number of household members is highly associated with poverty in most developing countries, as noted in previous studies (Grootaert, 1999). Because Afghanistan has a limited social infrastructure such as education and public health, a policy for population control may help to enhance the social cohesion by moving from a family-centered community to a society-centered community. As a result, the society works efficiently for managing rangelands and generating incomes, which helps to eliminate the severe poverty of communities.

Rangeland income sources have a statistically significant impact on poverty, and marginal effects analysis reveals that an increase of rangeland income affects more likely to be under poverty. These results indicate socially marginalized poor people, with no means of income, such as from agriculture, or without any agricultural land in the Afghanistan more likely to visit rangelands to collect non-timber forest products or utilize them. However, relative to livestock income, rangeland income has limited effects in generating income and contributes on keeping people’s livelihood just above subsistence level. Agroforestry practices may provide an alternative way to enhance the increase capacity of rangeland dwellers, while rangelands and their species are managed sustainably, because agroforestry provides a system whereby a deliberate attempt is made to integrate and manage both rangelands and livestock resources on the same landscape (Kiyani et al., 2017).

When it comes to the effects of poverty alleviation, the income contribution of rangeland species income, livestock management and livestock products on poverty showed a small impact relative to socio-economic factors such as social capital, education, and household size. The result implies that the effect of direct income-generation activities may be limited, and the contribution of improving rangeland management and developing new income sources is not enough in achieving poverty reduction. Community development programs should pay more attention on improving social connection, education program, birth control programs, while the program provides other income generating opportunities through its community driven rangeland management policies in Afghanistan.