Purpose -The purpose of this paper is to analyze the extent to which weather index-based insurances can contribute to reducing shortfall risks of revenues of a representative average farm that produces corn or wheat in the North China Plain (NCP). The geographical basis risk is quantified to analyze the spatial dependency of weather patterns between established weather stations in the area and locations where the local weather patterns are unknown. Design/methodology/approach -Data are based on the Statistical Yearbook of China and the Chinese Meteorological Administration. Methods of insurance valuation are burn analysis and index value simulation. Risk reduction is measured non-parametrically and parametrically by the change of the standard deviation and the value at risk of revenues. The geographical basis risk is quantified by setting up a decorrelation function. Findings -Results suggest significant differences in the potential risk reduction between corn and wheat when using insurance based on a precipitation index. The spatial analysis suggests a potential to expand the insurance around a reference weather station up to community level. Research limitations/implications -Findings are limited by a weak database in China and, in particular, by the unavailability of individual farm data. Moreover, the low density of weather stations currently limits the examination of the approach in a broader context. Practical implications -The risk reduction potential of the proposed insurance is encouraging. From a policy point of view, the approach used here can support the adjustment of insurers towards different crops. Originality/value -This paper is believed to be the first that investigates a weather index-based insurance designed for an average farm in the NCP and the quantification of geographical basis risk.

Purpose -Since the 1990s, there has been a discussion about the use of weather index-based insurance, also called weather derivatives, as a new instrument to hedge against volumetric risks in agriculture. It particularly differs from other insurance schemes by pay-offs being related to objectively measurable weather variables. Due to the absence of individual farm yield time series, the hedging effectiveness of weather index-based insurance is often estimated on the basis of aggregated farm data. The authors expect that there are differences in the hedging effectiveness of insurance on the aggregated level and on the individual farm-level. The purpose of this paper is to estimate the magnitude of bias which occurs if the hedging effectiveness of weather index-based insurance is estimated on aggregated yield data. Design/methodology/approach -The study is based on yield time series from individual farms in central Germany and weather data provided by the German Meteorological Service. Insurance is structured as put-option on a cumulated precipitation index. The analysis includes the estimation of the hedging effectiveness of insurance on aggregated level and on individual farm-level. The hedging effectiveness is measured non-parametrically regarding the relative reduction of the standard deviation and the value at risk of wheat revenues. Findings -Findings indicate that the hedging effectiveness of a weather index-based insurance estimated on aggregated level is considerably higher than the realizable hedging effectiveness on the individual farm-level. This refers to: hedging effectiveness estimated on the aggregated level is higher than the mean of realized hedging effectiveness on the individual farm-level and almost every evaluated individual farm in the analysis realizes a lower hedging effectiveness than estimated on the aggregated level of the study area. Nevertheless, weather index-based insurance designed on the aggregated level can lead to a notable risk reduction for individual farms. Originality/value -To the authors' knowledge, this paper is the first that analyzes the influence of crop yield aggregation with regard to the hedging effectiveness of weather index-based insurance.