Sustainable intensification among smallholder maize farmers in Ethiopia: Adoption and impacts under rainfall and unobserved heterogeneity

In addition to experimental data, using information collected from farmers is vital to assess the performance of candidate farming technologies. The latter provides the basis for farmers’ assessments and adoption decisions. Using detailed georeferenced plot and household level data from Ethiopia, this paper estimates the impact of a set of sustainability-enhancing agronomic practices on maize yields and implied net maize income with the aid of multinomial endogenous regime-switching regression that corrects for unobserved heterogeneity. The treatments’ effects on yield ranged from 29.5% for the fertilizer-dominant strategy to 88% in the case of fertilizer, maize-legume diversification, and soil and water conservation combinations. Second, while the impacts of the agronomic packages are significant in almost all states of the world, there were considerable heterogeneity effects conditioned on observed rainfall deviations from the historical average with levels effects of up to 92% and heterogeneity effects of 33–64%. The heterogeneous and rainfall-conditioned impacts point to the role of local biophysical modifiers of the impacts of agricultural practices. Investments in context-appropriate adaptation in agricultural research and extension is suggested.