Income inequality of tribal households in Bangladesh: A decomposition analysis

This paper uses field survey data to examine farm and non-farm income to reduce income inequality among tribal households in Sylhet district, Bangladesh. It uses a decomposition method and a regression-based decomposition method to measure income and expenditure inequality by source of income as well as by household and family characteristics. For this, non-farm income was considered as an external transfer of total income to decompose the Gini index and as a potential substitute for farm income, taking into account household preferences. The results indicate that income differences exist between farm and non-farm income, thereby increasing income inequality. The endowment effect on household characteristics was positive and significant, directly supporting the increase in non-farm income per capita daily household expenditure. On the other hand, in terms of its components, non-farm self-employment income reduces income inequality, while non-farm wage income increases income inequality. At the 50th quantile results, an additional year of age of the household head increases the per capita expenditure of farm and non-farm households by BDT 300 to BDT 400, respectively. Moreover, the education of household heads also positively and significantly affects the per capita expenditure of non-farm households. As estimated by the 50th quantile, farm and non-farm households with graduation had higher per capita daily expenditure than household heads with primary education. Raising household education level and total farm household income to a better level, adopting farm-income enhancement programs, and enacting policies to increase alternative income opportunities that will shift income from non-agricultural to agricultural tribal households, reduce income inequality, and bring about the well-being of tribal society.