Crop production and soil nutrient management : an economic analysis of households in Western and Central Kenya

The study examines how a combination of socio-economic and household factors influences farm household decisions on soil nutrient management and on crop production in two regions of Kenya (Kiambu and Vihiga). It further examines how these decisions impact on household objectives and on productivity. Emphasis is given to the use of inorganic fertilisers because this is one of the major sources of nutrients that hold considerable potential for increased productivity. A cross-sectional household level data set collected through a formal survey using pre-tested structured questionnaires is used for the investigations. Various methodological approaches are employed including descriptive statistics, cluster analysis, the dual and primal approach to analysing the farm production, and mathematical programming. Cluster analysis yielded three household types in each of the two regions, and these household types form the basis for further analysis. The most important crops to the households in Kiambu are maize, potatoes and kale, as identified through farmer perception and restricted activity profit ranking. In Vihiga maize and beans remains the sole most important combination. Demand for fertiliser is price elastic with own price elasticity of -2.44 in Kiambu and -1.88 in Vihiga. Output prices of maize and of kale in Kiambu, and of maize in Vihiga also significantly influence the demand for fertiliser with elasticities of 0.90, 0.28 and 1.58 respectively. In Kiambu, fertiliser price has very low elasticities on crop outputs, particularly of kale, which would seem to imply that fertiliser does not have a big impact on crop production. However, results fiom the primal approach show that currently labour is the constraining factor, particularly in kale production, making the impact of fertiliser obscure. In fact, increasing the level of fertiliser use will have a substantial impact on the output of all the crops analysed, in both the regions. The results confirm previous observations that households are using much lower levels of fertiliser than is optimal. If the market price of fertiliser and of the outputs is considered, the results show that households could profitably increase their level of fertiliser use on all the crops examined. However, the households' shadow price of fertiliser is much higher than the market price implying that households consider other costs that we do not observe e.g. credit constraints. ResuIts fiom mathematical programming confirm that households are liquidity constrained. The level of the credit constraint varies across the different household types based on the level of their extemal remittances relative to farm size. However, households in all the farm types require credit to enable them increase their level of fertiliser use.