Modeling interrelated inputs adoption in rainfed agriculture in Senegal

The adoption of certified seeds and chemical fertilizers is central for African agri- culture which is characterized by very low productivity. This paper analyzes the technology adoption of certified seeds and inorganic fertilizers for two central crops in Senegal: rice and groundnut. The joint adoption of these two technologies is modeled in the presence of production risk using a flexible bivariate probit model. A recent agricultural survey, representative country-wide, collected in 2017 is used for our application. Descriptive statistics confirm the low rate of agricultural technology adoption. In the rainfed system, the average inorganic fertilizer used is about 28 kg/ha. The analysis reveals that in rural Senegal de- cisions to adopt certified seeds and inorganic fertilizers are interrelated for both rice and groundnut systems. For the rice system, a heterogeneous dependency is revealed, while for groundnut technology adoptions, a homogeneous correlation is found. Production risk is found to have a significant impact on technology adoption. We also found that determinants of individual technologies and their joint adoption vary widely across crops. However, the main determinants of technology adoption in rainfed agriculture in Senegal include cooperative membership, access to extension services, access to credit, education, family size, and farm size.