Weather index insurance design: a novel approach for crop insurance in Brazil

• Autor(a) principal: Miquelluti, Daniel Lima
• Data de Publicação: 2019
• Tipo de documento: Tese
• Idioma: eng
• Título da fonte: Biblioteca Digital de Teses e Dissertações da USP
• Texto Completo: http://www.teses.usp.br/teses/disponiveis/11/11132/tde-02082019-100224/
• topic: Crop insurance; GWQLASSO; GWQLASSO; Risco sistêmico; Seguro agrícola; Systemic risk

Crop insurance is recognized as one of the most efficient mechanisms of income protection in agriculture, transferring risk from agriculture to other agents and economic sectors. Insurance tends to stimulate the increase of cultivated area and the use of technology, especially as it acts as an additional guarantee for access to credit. In Brazil, however, the massification of rural insurance is limited due to the restricted budget to fund government subsidization. Also, the lack of predictability and guarantee of resources prevents the long-term planning of investments by the private sector, imposes costs on the beneficiaries and generates dissatisfaction of the target public. This thesis aims to contribute to the expansion of crop insurance in Brazil through the research of index insurance, which has lower administrative and claim adjustment costs when compared to traditional insurance. The absence of in situ claim adjustment and moral hazard monitoring reduces the administrative costs of this type of insurance, permitting a subsidy free crop insurance. In the first of two articles, we explore the availability and quality of public databases for soybean yields and daily rainfall in the state of Paraná in Brazil in order to verify the feasibility of an index insurance product. We use multiple imputation by chained equations (MICE) to fill missing values in the rainfall dataset and study the existence of spatial and temporal patterns in the data by means of hierarchical clustering. Our results indicate that Paraná fulfills data requirements for a scalable weather index insurance with MICE and hierarchical clustering being effective tools in the pre-processing of data. The second article studies the efficiency of a novel regression approach, the geographically weighted quantile LASSO (GWQLASSO) in the modelling of yield-index relationship for weather index insurance products. GWQLASSO allows regression coefficients to vary spatially, while using the information from neighboring locations to derive robust estimates. The LASSO component of the model facilitates the selection of relevant explanatory variables. A weather index insurance (WII) product is developed based on 1-month SPI derived from a daily precipitation dataset for 41 weather stations in the State of Paraná (Brazil) for the period of 1979 through 2015. Soybean yield data are also used for the 41 municipalities from 1980 through 2015. The effectiveness of the GWQLASSO product is evaluated against a classic quantile regression approach and a traditional yield insurance product using the Spectral Risk Measure (SRM) and the Mean Semi-deviation. While GWQLASSO proved as effective as quantile regression it outperformed the yield insurance product, thus proving an alternative to the crop insurance market in Brazil and other locations with limited data.