Effect of Farm Income Risk on Smallholder Production Decisions in the Highlands of Machakos District, Kenya:The Case of Kauti Irrigation Scheme
Raising agricultural productivity and incomes in the densely populated highlands of Machakos District is constrained by various risks which make farm income uncertain. Farmers diversify their activities to cushion themselves against risks but there is no empirical recommendation to guide them in selecting optimal enterprise combinations (farm plans). The farmers adopt own-preferred farm plans but it is not known whether such plans minimize risk. This study assessed the effect of farm income risks on production decisions in Kauti Irrigation Scheme, with a view to identifying farm plans that earn farmers higher incomes at existing or lower levels of income risk. The study used crop and livestock production and marketing data covering 2007/08 and 2008 rain seasons, collected in March 2009 through a farm household survey involving 113 households. The data were analysed using linear and quadratic risk programming techniques. Results show that a typical farm plan in the study area features nine enterprises consisting of 0.96 acres of maize intercrop, 0.13 acres of French beans, 0.03 acres of kales and 0.04 acres of tomatoes in short rains season; 0.95 acres of maize intercrop, 0.02 acres of kales and 0.02 acres of tomatoes in long rains season; and 0.25 acres of coffee and 0.93 livestock units in both seasons. This farm plan is highly diversified compared to optimal farm plans developed in this study, but it is risk-inefficient. However, by adopting the optimal farm plans, a typical household can improve its income from the current KSh 36,049 to KSh 63,913, representing a 77 percent increase, under the current technological and resource constraints. This income can be further increased to KSh 75,339 - which is more than double and 31 percent less risky than the current income, if households' access to working capital is increased. The study recommends that farmers abandon production of coffee and kales; allocate 1.018 acres to maize intercrop, 0.131 acres to French beans, and 0.119 acres to tomatoes in short rains season; 0.761 acres to maize intercrop and 0.164 acres to tomatoes in long rains season; and keep 2.5 livestock units throughout the year. This farm plan earns farmers KSh 75,339, with a standard deviation of KSh 30,790. For this to happen, the government and development partners should develop farm input financing programmes that increase farmers' access to working capital by 67 percent, from the current KSh 9,915 to KSh 16,565. With this additional capital, farmers will particularly bring most of the idle irrigable land into production, reducing the proportion of income risks attributable to production risks. Notwithstanding these potential increases in farm incomes, the farm sizes in the study area are too small such that the optimal farm income is not sufficient to lift households out of poverty. This calls for policy makers to find ways of increasing off-farm employment.