This GitHub repository hosts the complete codebase used in the research paper "Behavioural Factors Matter for the Adoption of Climate-Smart Agriculture."
Increasing agricultural productivity while maintaining environmental sustainability are two important targets in achieving the sustainable development goals under climatic shocks. In this regard, different climate-smart agricultural (CSA) practices have been recommended and promoted to meet these goals. However, the adoption of these practices remains low and variable. For the most part, low adoption has been attributed to external factors. Behavioural and psychological factors also matter but have received little empirical and policy attention. In this study, we examine the relationship between aspirations, aspiration gaps, and the adoption of CSA practices such as crop rotation, intercropping, fallowing, and the use of organic soil amendments. Employing parametric and non-parametric estimation techniques on a pooled farm household survey from Cameroon and Kenya, we show that aspirations are associated with the use of crop rotation and organic soil amendments. We also investigate the theorized non-monotonic inverse U-shaped relationship between aspiration gaps and investments. We find evidence of this relationship for the adoption of CSA practices, suggesting an aspiration failure for smallholder farmers. These results imply that aspirations that are ahead but not too far ahead of the current state serve as the best incentives for stimulating the adoption of CSA practices. Employing the multivariate probit model, we further highlight interdependencies in the use of these CSA practices. Specifically, we underscore significant complementarities, suggesting that the bundled use of these practices is recommended. Overall, the analysis demonstrates that aspirations matter for farmer decision-making with many implications for both agricultural, food, and environmental policy.
The Stata script contains our statistical analysis and estimation processes. The script is thoroughly annotated to elucidate every step, from data manipulation to econometric modeling, ensuring that our analysis is transparent and reproducible.
Included are the code for recreating the paper's graphs and charts, along with adaptable templates for similar visualizations in other research contexts.
Here, you'll find additional resources that complement the main findings of our study. This include extended data tables, additional statistical analyses, and robustness checks that provide further context and support for our conclusions.