Abstract
This paper provides empirical evidence regarding the impact of agricultural technologies on smallholders’ output market participation. The analysis is based on Farmer Innovation Fund impact evaluation survey collected by the World Bank in 2010-2012 covering 2,675 households in Ethiopia. Endogenous treatment effect and sample selection models are employed to account for the self-selection bias in technology adoption and market participation. Regressions based on matching techniques are employed for robustness check. The estimation results show that the use of improved agricultural inputs significantly affects farm households marketable surplus production. We found evidence that application of high-yielding varieties increases surplus crop production by 7.39 percent per year, whereas chemical fertilizer use increases surplus by 2.32 percent. When farmers apply the two inputs jointly, marketed surplus increases by 6 percent which establish the complementarity of the two technologies. Marketable surplus crop production and market participation of farmers are determined by access to modern inputs, crop price, farm size, availability of labor, and infrastructure. Access to credit and training fosters technology adoption. Therefore, agriculture and rural development policy need to focus on supporting agricultural technology adoption.
Original language | English |
---|---|
Publisher | UNU-MERIT working papers |
Volume | 2017 |
Edition | 008 |
Publication status | Published - 6 Feb 2017 |
JEL classifications
- d04 - "Microeconomic Policy: Formulation; Implementation; Evaluation"
- o12 - Microeconomic Analyses of Economic Development
- o13 - "Economic Development: Agriculture; Natural Resources; Energy; Environment; Other Primary Products"
- o33 - "Technological Change: Choices and Consequences; Diffusion Processes"
- q13 - "Agricultural Markets and Marketing; Cooperatives; Agribusiness"
Keywords
- Surplus production
- technologies
- social network
- Ethiopia
- agricultural innovation
- endogenous treatment effect model