Determinants of Income Generating Activities of Rural Households

Abstract

This study aims to identify and analyse the determinants of income generating activities of rural households in the vicinity of the Lore-Lindu National Park in Central Sulawesi, Indonesia. It helps to identify factors which are essential for the design of policies promoting alternative income strategies. Data was collected through a standardised, formal questionnaire from 301 randomly selected households. In the analysis, the following income generating activities are differentiated: agricultural self-employment, agricultural wage labour, non-agricultural self-employment, and non-agricultural wage labour. Because of its importance the agricultural self-employment category is further divided into annual and perennial crop production, livestock production, and the sale of forest products. Various econometric models are used to analyse the influence of socio-economic factors on income generating activi-ties. A linear model is applied in the case of total household income. Probit models are used to investigate factors influencing activity choice. In the analysis of activity incomes, a simultaneous equation model with correction for the endogeneity of activ-ity choice is applied. Finally, the determinants of income diversification are analysed using Tobit models.Agricultural activities are the most important source of income for rural households in the region and make up 70% of total household income. Within this category the most important source of income is crop production, with the most important crops being irrigated rice and cocoa. The remaining 30% of the total household income originates from non-agricultural activities. However, only around 18% of the households gain income from the latter activities. In contrast, 96% participate in agricultural activities. Differentiation between different wealth groups shows that activities outside the agricultural sector are particularly important for the less-poor households.The econometric analysis shows that access to physical and human capital has a significant influence on total household income. The area owned, the value of other assets possessed, as well as the number of livestock and family labourers positively influence household income. The possession of land also has a strong positive influence on the participation in crop production, whereas the possession of irrigated land reduces the likelihood of participation in agricultural wage labour activities and in the sale of forest products. Richer and non-indigenous households are more likely to participate in non-agricultural self-employment. In contrast, non-indigenous households are less likely to participate in the sale of forest products and in agricultural wage labour activities. Participation in formal credit markets discourages participation in annual crop production, but encourages participation in the production of perennial crops and non-agricultural wage labour activities. The access to roads has a strong influence on participation in almost all activities. The analysis of activity incomes shows again that the possession of land has a strong positive influence on the income gained from crop production, while the possession of irrigated land reduces the income gained from agricultural wage labour and the sale of forest products. Similar to its effect on participation, the value of other assets owned has a positive influence on the income from non-agricultural self-employment. The analysis also shows the importance of education in non-agricultural wage labour activities and in annual crop production. Ethnicity has a strong influence on perennial crop produc-tion as well as on income from non-agricultural self-employment. The access to tarmac roads has a positive effect on the income from agricultural wage labour and the sale of forest products.The results of the analysis are used to draw policy conclusions with respect to poverty alleviation, deforestation and rural development.