Socio-economic determinants of rural household food expenditure: A quantile regression analysis

This paper provides an analysis of factors that influence rural household expenditure on food using a quantile regression analysis. The objective was to investigate if there are any relationships to discern between household expenditure on food and a number of other socio-economic factors in addition to household income and household size as stipulated in Engel�s law. The results indicate that indeed there are relationships that could be discerned between household food expenditure and gender, education, occupation, household income, number of people depending on household income, gender distribution by age groups and number of livestock sales per annum. However, household income has a relatively smaller effect in magnitude as compared to the rest of these factors. Furthermore, the results show that the relationships differ along the quantiles. In other words, some factors had a significant and higher effect in the lower quantile compared to the higher quantile. On the basis of such results, the study suggests that tailor-made interventions should be considered in the development initiatives that are targeting rural households as different factors affect these households differently.