Climate Anomalies, Seasonal Interactions, and Input Responses in Cocoa Production: Evidence from Côte d’Ivoire

This paper analyzes the nonlinear and stage-specific impacts of rainfall variability on cocoa yields in Côte d’Ivoire, the world’s leading cocoa producer. We combine a georeferenced Standardized Precipitation Index (SPI) dataset (1981–2025) with farm-level survey data collected during two contrasting climate years: a drought year (2022) and a near-optimal year (2023). Results reveal three key findings. First, the rainfall–yield relationship is highly nonlinear, exhibiting both Ushaped and inverted U-shaped patterns across seasons, suggesting that linear specifications may misestimate climate impacts. Second, climate sensitivity varies by production stage and baseline conditions: moisture availability during the growing season is the primary constraint under drought, while pre-season water recharge becomes more critical in wetter conditions. Third, climate effects are asymmetric, with yield losses from moisture deficits exceeding gains under favorable rainfall, indicating substantial downside risk for farmers operating near optimal conditions. Herbicide expenditure per hectare positively affects yields in the favorable year, underscoring the importance of weed management, while plantation age is negatively associated with productivity, consistent with declining yields in older cocoa stands.