Search results

Agricultural weather index insurance (WII) has been introduced in pilot or experimental form in many countries. However, the effective demand for WII is often limited by the impact of the basis risk. Thus, the purpose of this paper is to propose a new type of double trigger product, named “supplement” type, to reduce basis risk and improve the performance of the standalone WII.

Two measures of performance are introduced by the certainty equivalent income of expected utility theory. Through the Monte Carlo experiments and empirical study, this paper compares the performance of three types of double trigger products.

The findings indicate that the supplement type can significantly improve the performance of the single weather index product. First, it covers the downside basis risk and the catastrophic basis risk when the standalone WII fails to do so, especially in case of extreme losses. Second, it is superior when the correlation between the weather index and the yield index is not so strong, and can further enhance the performance of insurance even when the weather index and the yield index are highly correlated, for which the standalone WII could perform well.

The supplement type double trigger product proposed in this paper as an enhancement version finds a more preferable way to improve the standalone WII with relative lower complexity.

While crop insurance ratemaking has been studied for many decades, it is still faced with many challenges. Crop insurance premium rates (PRs) are traditionally determined only by point estimation, and this approach may lead to uncertainty because it is sensitive to the underwriter’s assumptions regarding the trend, yield distribution, and other issues such as data scarcity and credibility. Thus, the purpose of this paper is to obtain the interval estimate for the PR, which can provide additional information about the accuracy of the point estimate.

A bootstrap method based on the loss cost ratio ratemaking approach is proposed. Using Monte Carlo experiments, the performance of this method is tested against several popular methods. To measure the efficiency of the confidence interval (CI) estimators, the actual coverage probabilities and the average widths of these intervals are calculated.

The proposed method is shown to be as efficient as the non-parametric kernel method, and has the features of flexibility and robustness, and can provide insight for underwriters regarding uncertainty based on the width of the CI.

Comprehensive comparisons are conducted to show the advantage and the efficiency of the proposed method. In addition, a significant empirical example is given to show how to use the CIs to support ratemaking.