| dcterms.abstract | This research investigated the feasibility of generating higher income and employment in Kenyan small-scale agriculture for alleviating rural poverty. It also assessed the possible conflict and trade-offs between increasing both income and employment. The probable impacts of efficient resource allocation, relaxation of subsistence food constraints, higher profitability, higher off-farm employment, land rental market, and a composite of these policy options were analyzed. Production data for the 1978-79 crop year were obtained from a random sample of 38 farms in Mbiuni location of Machakos District. Using a Land-Adult Equivalent Ratio (LAER), the sample was divided into three groups: low (.80 ha), medium (1.31ha) and high (2.35 ha). For each group, a representative farm was developed. Three linear programming models were constructed for each representative farm. The first assumed a profit- maximizing objective. The second represented an employment-maximizing objective. The third adopted a multiobjective programming approach. Each model included land, labor, capital and subsistence food constraints, and production, labor and capital activities. For each model, results were weighted by the number of farms in each group and aggregated over the sample. Under existing farming systems, enterprise intensity increased, and food crops substituted for cash crops as the LAER declined. The average income per household person was Kenya shillings (Ksh) 700. This was only 37 percent of the then per capita national income of Ksh 1,875 (US $255, 1978). The average income decreased as the LAER fell. On average of the entire sample, 53 percent of the labor supply was employed. As the LAER declined, the rate of employment decreased, increasing underemployment; and the rate of off-farm employment and off-farm income increased. Generally, all farms utilized resources efficiently. In particular, low LAER farms were relatively more efficient than high LAER farms. There was capacity to increase income and employment up to 54 and 31 percent respectively under the income- maximizing strategy. The employment-maximizing strategy generated a substantial (20% or more) 65 percent increase in employment, but a modest 11 percent increase in income. For each strategy, the increases were likely if all the policy options were pursued concurrently. On average of all options, the income strategy generated 36 percent more income but 20 percent less employment than the employment strategy. Conversely, the employment strategy generated 26 percent more employment but 27 percent less income than the income strategy. The average income-employment trade-off was 2.07 man-days of employment sacrificed for each pound (Ksh 20) of income gained by the income model. Equivalently, Ksh 9.68 of income were forfeited per man-day of employment gained by the employment model. The Multiobjective Linear Programming (MOLP) approach maximized both income and employment for each option, taking into account the relative value of each objective. On average of all options, MOLP generated 23 percent more employment but only 7 percent less income than the uniobjective income model. Alternatively, MOLP generated 26 percent more income with merely 2 percent less employment than the uniobjective employment model. The MOLP composite option yielded the best results for the whole analysis. It generated 45 and 60 percent increase in income and employment respectively. Two overall conclusions can be made. First, farm population becomes poorer as the land base per worker declines. A variety of farm policies should be pursued concurrently in order to increase income and employment to alleviate the poverty. These should emphasize food crop production, labor-using and land-saving technologies. Second, a multiobjective programming approach should be adopted in order to maximize both income and employment in agriculture. | en_US |
