Commercialisation of Agriculture in Kenya: Case Study of Policy Bias and Food Purchases by Farm Households

This study considers the effect of cash cropping on food availability and investigates the determinants of household food expenditure as a proportion of gross income relying on a sample of rural households in the Nyeri district of Kenya. Results from an application of a Tobit model suggest that household food purchases and food availability may suffer as a consequence of increasing cash cropping in Kenya. Husbands favour commercial crops and, it seems, favour non-food purchases. Married women living with their husbands use proportionately less of their gross income to purchase food compared to unmarried women and to those women not living with their husbands. Male bias in food purchased is present, and is exacerbated when payment for cash crops is lumpy. Lumpy cash payments tend to reduce proportionate food purchases by households. We also find that remittances and family size are positively associated with food purchases as a proportion of gross income. Because of household food availability issues, there is a case in many developing countries for reducing policy emphasis on expansion of production of non-food cash crops and a case for greater encouragement of subsistence food production. Increased commercialisation of agriculture can result in reduced availability of food to women and children.