Predicting farm investment success in semi-arid regions Algeria: An asset-based policy tool

The aim of this research was to develop a predictive model for agricultural investment success in Algeria’s semi-arid cereal farming systems, with the view to guiding governmental policies on loan and subsidy allocation. The research focused on semi-arid regions, where cereal farming was fundamental for food security and strategic agricultural investment was crucial for enhancing outcomes and farm resilience. Utilising survey data from 198 farms in Setif-Algeria, it was built seven composite asset dimensions and established farm typologies via multiple variate analysis. A binary logistic regression model was developed to predict agricultural investment success. The logistic regression model demonstrated a practical overall classification accuracy of 79.8%, with a R² of 0.358 and R² of 0.486. Results indicated that “Technical Asset” was the strongest positive predictor of success (B = 0.728, OR = 2.072, p = 0.002). Some composite assets such as “Diversity of Farming”, “Connectivity”, and “Performance” were found to negatively predict investment success, suggesting complex dynamics such as over-diversification or higher-risk schemes by well-resourced farms. The results confirmed substantial structural differences between farm types, with large-scale farms demonstrating significantly higher levels of biophysical, human, and diversification assets, reflecting stronger resource capacity and adaptive potential. The developed predictive model showed strong explanatory and classification performance, confirming the key role of technical capacity as a decisive factor in achieving successful investment outcomes. The research gave a practical predictive framework for policymakers, enabling data-driven decisions to optimise resource distribution, reduce the risks associated with public financing and ultimately foster more successful agricultural investments, thereby strengthening sustainable cereal cropping and food security. These findings provided policymakers with a data-driven screening tool to optimise the allocation of agricultural subsidies and loans, directly strengthening food security in semi-arid regions