TRANSACTION COSTS IN SMALL HOLDER AGRICULTURE: THE CASE OF SOIL CONSERVATION IN SEMI-ARID AREAS OF KENYA

Marginal areas in Kenya that comprise about 80% of the total land area are ecologically vulnerable with very serious problems of soil erosion. A decline in agricultural productivity is the result, accompanied by serious household food insecurity. As more people immigrate to these areas coupled with births, the problem is bound to get worse.

Nevertheless, these areas can be productive iffarmers make investments on their land Investments into soil and water conservation include terraces, trees, cattle, manure, fertiliser, equipment, wells, dams and other infrastructure. Investments into soil and water conservation may be undertaken when sufficient returns are expected_ These returns, in particular monetary returns, can be related to many factors, but are always influenced by transaction costs of market exchange which subsequently determine the level of access to input and output markets.

Essentially the research seeks to determine the influence of transaction costs on soil conservation in smallholder agriculture and their role and impact on sustainable resource management and agricultural productivity.
A multi-stage random sampling was used to collect cross-sectional data from farming households using a structured questionnaire in Machakos and Kitui Districts. Besides descriptive statistics, econometric analysis using Three Stage Least Squares (3SLS) estimated with the help ofHeckman Two Stage procedure was used to test whether transaction costs to the market was a binding constraint to soi! and water conservation, resource use and agricultural productivity. A Cobb-Douglas type of regression function was also used to investigate how farmers respond to the net benefits of soil conservation measures. Further, a dynamic simultaneous agricultural household mode! was used to model households as both production and consumption centres.
The study findings show clearly that transaction costs reduce manure and fertiliser use of farming households as well as and more importantly soi! conservation investments including net benefits of soil conservation investments. The results further show that transaction costs increase labour use. This apparent anomaly can be explained as household' s response to increased transaction costs and the need to meet subsistence needs (i.e. food security). Simulation of the agricultural household mode] of a 10% reduction in transaction costs show that soi! conservation investments increase though with a lower magnitude depending on resource endowments of farming households.
Thus generic measures that can significantly reduce transaction costs such as improvement of road infrastructure; formation of co-operatives, marketing producer groups, and self-help groups serve as viable policy areas in order to induce investments in soil conservation measures on a large scale, with consequent sustainability of farming systems. Other likely policy measures include revamping extension service, improving property rights, and taking into account mobilization of social capital as part of the policy package towards sustainable agriculture.