High liabilities or heavy subsidies: Farmers’ preferences for crop insurance attributes in Hunan, China

Understanding farmers’ preferences for crop insurance attributes is crucial in designing better insurance products and guiding government policies but such research is lacking, particularly in developing countries. The paper aims to discuss these issues.

This study uses a survey featuring a discrete choice experiment and policy simulation.

Overall, crop insurance has positive values to farmers, although preference is heterogeneous based on socioeconomic characteristics and risk position. Policy simulation confirms the roles of liability in strengthening insurance participants’ welfare and premium subsidy in encouraging participation. Introducing one more product into the market can accommodate farmers’ diverse needs and lead to increases in both aggregated social welfare and participation while maintaining the current level of government expense in subsidy – a potential Pareto improvement.

Methodology employed is not the most novel in the choice experiment literature as many of the advances in choice experiment design could not be applied due to the actual condition in rural China and Chinese farmers’ capability in understanding the experiment.

The results indicate that the current single-product market structure using “low liability with high premium subsidies” cannot accommodate the diverse needs among farmers. Providing more varieties of liability-subsidy combinations, e.g. a high liability with low premium subsidy insurance product, can substantially improve participants’ welfare with little impact to the probability of participation.

The authors believe that this is one of the very few studies that that analyze farmers’ preferences and willingness to pay for the attributes of crop insurance products. It also shows how crop insurance product design can build upon farmers’ choices to achieve a potential Pareto improvement in aggregated social welfare in the context of a fast-developing crop insurance market.