Throughout the world, public provision of key goods and services outside markets is being replaced by private delivery in increasingly liberalised markets (Florkowski et al. 1997; Ariyo and Jerome 1999; Kuczynski 1999; Nolan and Xiaoqiang 1999). This trend is particularly significant in Africa where privatisation and liberalisation often involve public withdrawal from production and delivery of agricultural extension services (FAO 1994; Umali et al. 1994; Goletti and Govindan 1995; Tambi et al. 1999).
Yet extension systems in Africa continue to flounder. And while proposals continue to emerge as to what African governments should do in order to improve efficiency and effectiveness (e.g. Beynon 1996; Neuchatel 1999; Rivera et al. 2000; World Bank 2000), few propositions have appeared as to how precisely to do so. This is not surprising, for the literature still offers little guidance as to specific factors and processes that likely influence development and diffusion of agricultural technologies in given circumstances. Especially lacking are analyses of the actual and potential roles played by, and obstacles and opportunities facing, rural traders, households, and communities—i.e. the rural 'private sector', on which so much depends, and of which so much is expected, in extension reform efforts.
This paper addresses these gaps in the literature by analysing the design, implementation, and outcome of an intervention to control trypanosomosis (a devastating livestock disease transmitted by tsetse flies) undertaken by the International Livestock Research Institute (ILRI) in South-West Ethiopia. While not conceived as such at the time, this intervention emerges, in retrospect, as a real-world experiment in decentralised private provision of a traditional public extension activity, namely delivery of animal health inputs and services. Key characteristics of the disease and its vector, of the intervention technology, of the overall intervention strategy, of the region selected for intervention, and of participating farmers suggest insights into a range of issues that arise when analysts and decision makers take fully on-board the question of how to reform agricultural extension services in Africa.
The next section describes the intervention. This is followed by a discussion of interactions among key regional features and household characteristics, which, in unison, implied a self-reinforcing process key to the success of the initiative. Implications for extension reform are then drawn.
