Abstract/Summary

Development and dissemination of hybrid rice technology can be one of the key strategies to meet growing demand for staple food in Bangladesh. A number of initiatives for hybrid rice research and development have been undertaken for the last couple of decades, but the adoption rate of hybrid rice is not encouraging. In recent years, although some studies have investigated the underlying socio-economic factors of hybrid rice adoption, research on estimating production efficiency of hybrid rice is relatively few. In addition, there is a bias in sample selection inherent in most of the technology adoption studies because rational farmers will always choose between hybrid and inbred rice depending on a host of price and non-price factors. Furthermore, the production function that uses observations from a single variety only may produce biased estimates of production efficiency. Given this backdrop, this paper adopted the recently developed method that corrects for sample selection bias in a stochastic frontier framework and simultaneously identifies the determinants of hybrid rice adoption and its farm-level productivity and efficiency while relaxing the assumption of perfect efficiency of the individual producers. A survey was conducted to collect primary data on inputs and outputs of hybrid and inbred rice production. Some key information on the socio-economic characteristics of the sample farmers was also collected. Using pre-tested structured questionnaire, a total of 336 and 180 rice farmers were selected and interviewed at six districts of Bangladesh for the crop year 2004 and 2005, respectively. Model diagnostics reveal that serious selection bias exists, justifying use of our sample selection approach. The results reveal that the producer's decision to choose hybrid rice is positively influenced by gross return per unit of land but negatively affected by a rise in relative prices of labour and phosphate fertilizer. Adoption of hybrid rice is significantly higher for pesticide and organic manure users. Land, irrigation and mechanical power are the significant determinants of hybrid rice productivity. Productivity of hybrid rice varies across regions. The mean technical efficiency of the hybrid rice producers is estimated at 0.86 with a range of 0.55 to 0.97, implying substantial scope to improve production by eliminating inefficiency. Policy implications include price policies to improve hybrid rice price, tenurial reform aimed at consolidating farm size and smooth operation of the mechanical power services and investment in irrigation to expand hybrid rice adoption and productivity in Bangladesh.