Using Cubic Splines in Crop Insurance Models – A Replication Study

This study replicates the work of Bucheli et al. (2022) to evaluate the reliability and generalizability of their findings on the use of cubic spline methods in weather index insurance design in Eastern Germany. The study consists of two parts: a direct replication and an extended replication using different crops and another regional focus of yield data from Saxony, Germany. The direct replication confirms the original findings, while the extended replication reveals limitations in the model's risk reduction capacities. The gain in model fit, compared with daily temperature data, seems to be modest. It is questionable if an index insurance, which is solely based on temperature, will be considered as an attractive risk management tool, given the substantial basis risk that remains with farmers. The study highlights the importance of considering crop-specific risk profiles, regional climate, and agricultural conditions in insurance design. Moreover, our analysis highlights the relevance of FAIR (findable, accessible, interoperable, reusable) data for the performance of replication studies.