Actuarial Research On Area Yield Rice Insurance In Hubei Province

As an effective mechanism to transfer risks and compensate losses, agricultural insurance is not only an important part of the agricultural security system, but also one of the internationally important non-price protection tools. Since the 16th National Congress, the central Party and State Council have been attaching great importance to the development of policy-oriented agricultural insurance, which is manifested in six consecutively released pieces of No.1 central document from the year of 2004 to 2009, the "tenth-five year" planning and the document "State Council Guidance on Insurance Reform and Development" issued in June 2006. With a new round of pilot project carried out all through the country since 2004, it is seen that agricultural insurance develops well, for the premium income from 2004 to 2006 was 396 million, 729 million and 846 million respectively. Supported by national finance, it has continually increased to 5.18 billion in 2007 and 11.07 billion in 2008.However, agricultural insurance has been developing well in China in the recent two years, there is still one question that the security provided by current material cost crop insurance is insufficient. Then it is a very critical issue to study how to improve security level under the condition that national finance and farm households' income are limited. Compared with crop income insurance and farm-level yield crop insurance, it is undoubted that area yield crop insurance which has been promoted in the developing countries like India and Brazil in recent years is the better choice.It is affordable and accessible to farm households for the premium rate of area yield crop insurance is low while of great help for them to resist system risk. It avoids the moral hazard and adverse selection problems for the insurer by eliminating system risk to some extent. In addition, the data of such an insurance plan is easily available and credible, reducing transaction cost sharply. It is favorable for the government to lighten financial burden providing only a small amount of premium subsidy. Nevertheless, area yield crop insurance cannot prevent basic risk; if packaged with income insurance, it can reduce basic risk.Nowadays, there is scarce domestic research on area yield crop insurance, let alone relative actuarial research. Therefore, with rice production in Hubei Province as an example, this dissertation makes a deep and detailed empirical study on risk evaluation and division, rate making and contract design of county-level yield crop insurance actuarial.The main content and conclusion of the dissertation are as follows. First of all, it analyzes the main agricultural meteorological disasters in Hubei, focusing on the spatial and time distribution, intensity and range and annual change of drought and flood, as well as ranking drought and flood in different regions.Secondly, the dissertation evaluates the disaster risks confronted with in rice production in Hubei Province. Since non-parameter information diffusion model is superior to common parameter and non-parameter methods when evaluating losses from disasters with a small sample size, the dissertation makes use of non-parameter density function to evaluate the probability of early rice, midseason rice and late rice with a loss rate ranging from 0 to 1.Thirdly, it makes risk division of area yield crop insurance in Hubei Province. Climate, agricultural disaster, topography, water conservancy facilities and crop rotation considered, the dissertation selects twelve indexes, such as altitude, ranking of drought and flood and so on, to make risk division of county-level yield rice insurance in Hubei Province by factor analytic method and fuzzy clustering method.Fourthly is the making of pure premium rate of area yield rice insurance in Hubei Province. Comparing the result calculated respectively by experience rate method, non-parameter density function and hierarchical Bayesian' model under ARIMA model, it is found that the result from the former two methods are much close while the last one is more reasonable theoretically. There is significant difference of the rate made respectively by non-parameter density function and hierarchical Bayesian' model. However, if 8 regions selected and similar yield fitting model used, 100% coverage yield basically coincides with corresponding pure premium rate. Accordingly, the dissertation selects non-parameter density function under ARIMA model to make the pure premium rate of early rice, midseason rice and late rice in each county (city or district) of Hubei Province on 100% coverage level.At last, the dissertation discusses the design of an area yield rice insurance contract from five aspects, which are selection of insured area, forecast of central tendency of county-level yield, indemnity payout rules, determination of the range of franchise clause and coverage level and rate-making.The dissertation has made breakthrough in risk division, rate-making and contract design of county-level yield crop insurance and enriched the accomplishment of actuarial research on area yield crop insurance to a certain degree, which will provide reference for similar research and offer a scientific basis for developing area yield rice insurance in Hubei Province.