Households' Income-Generating Activities and Marginal Returns to Labour in Rural Tanzania

This study uses detailed household-level data to assess whether rural Tanzanian households seem able to allocate labour so as to maximise their incomes, and what factors determine if they do. In contrast to much earlier work on income diversification I use crop-level data to explicitly evaluate marginal returns within agriculture. The integrated household survey used allows me to then link these returns to household characteristics and broader labour supply decisions and consumption behaviour. In line with expectations agricultural wage work seems to be a last resort option, as agricultural wage labourers have lower marginal returns than others due to a higher labour allocation to own agricultural production. Furthermore, wage rates are much higher than the agricultural shadow wages, implying that there are gains to be made from expanding the non-farm side of the rural economy. However, there is no evidence that households are stuck in agriculture due to being constrained from entering the existing labour market, and neither do I find preferences for own crops being important for labour allocation. Work preferences seem to play a role though, and the findings are consistent with both credit and social networks being important determinants of a household's marginal productivity in agriculture.