LOSS INSURANCE MODEL OF RISK FOR AGRICULTURAL COMMODITY BASED ON MAXIMUM DAILY RAINFALL INDEX CONSIDERATION

Agricultural commodities in rainfed areas face significant risks of yield loss and crop failure due to uncertain rainfall patterns and intensities. Index-based crop insurance has been introduced as an adaptive strategy to simplify loss assessment using climate indicators. However, most existing schemes cover only a single peril, such as drought. This study aims to develop a loss model of risk for agricultural commodity using maximum daily rainfall index that accounts for both drought and flood risks. The model consists of two components: rainfall modelling and insurance modelling. Rainfall modelling identifies the appropriate probability distribution to define rainfall index parameters—trigger and exit—which represent thresholds for yield reduction and total crop failure, respectively. These parameters are derived through numerical integration and can be approximated using percentiles when crop-specific water requirement data are unavailable. Insurance modelling determines a benefit claim model based on rainfall probability and parameters of rainfall index, with three possible benefit claim conditions: full, partial, and none. A case study using maximum daily rainfall data (September–December, 1984–2014) for paddy in Dramaga, Bogor, indicates that the Burr Type XII distribution fits the data better than the GEV distribution. The estimated premium ranges from IDR 300000 to 300822.9 per hectare. In high-rainfall areas like Dramaga, premiums are primarily influenced by the probability of excess rainfall, while drought risk is negligible. Analysis over a 10-year actual maximum daily rainfall data (September–December, 2015–2024) shows that lower insured percentiles result in lower premiums. To improve accuracy, trigger and exit should ideally be determined based on the specific crop's water requirements. Despite data limitations, this model provides a conceptual model for developing more representative and actuarially fair loss model for agricultural commodity risk.