A sustainable area-yield insurance program with optimal risk pooling: A behavior-based machine learning approach

Abstract

Area-yield insurance is a promising alternative to address challenges that hinder the sustainability of traditional individual-loss insurance programs, including moral hazard, high administration costs, and data sparsity. However, the presence of basis risk diminishes the efficiency of area-yield insurance contracts. In this paper, we propose a behavior-based machine learning approach to optimally determine the risk pools, leading to a more sustainable area-yield insurance program. We first select the optimal number of risk pools by studying the producers' farming behavior under the protection of area-yield insurance contracts via a utility maximization framework. Then we utilize an unsupervised spectral clustering model to group producers into these risk pools. Such machine learning technique facilitates pooling producers with similar production history together, which guarantees the efficiency of the area-yield insurance contract, and moreover, addresses the challenges of high-dimensionality and computational complexity. The proposed optimal risk pooling procedure is empirically tested and cross-compared with the data from major corn production counties in the U.S. heartland region. Empirical results show that the proposed method reduces contract basis risk and mitigates producers' tail risk significantly. Moreover, compared to alternative statistical methods, the proposed framework produces risk pooling results that are more effective in risk reduction, and contain geographical and economic meanings.

Keywords: sustainability, area yield, basis risk, crop insurance, machine learning

JEL Classification: C10, C50, G22, G52, Q16, Q18