【作者】 李文娟；

【导师】 魏华林；

【作者基本信息】 武汉大学 ， 金融学， 2009， 博士

【摘要】 巨灾保险是一国保险体系的重要组成部分,也是长期备受保险理论界关注的一个议题。本世纪频发的巨灾事件带动了国内外巨灾保险理论研究的热潮,但在巨灾保险的研究领域,研究供给的多,研究需求的少。本文致力于从一个较新的思路和视角,即需求视角,来分析和探索巨灾保险。本文之所以研究巨灾保险需求及其影响因素,理由有三：其一,在已经建立起比较完善的巨灾保险制度的国家,制度本身并没有解决参保率的问题,巨灾保险需求的问题、参保率低的难题始终是巨灾保险制度中的一大困扰。其二,需求在巨灾保险体系中有着举足轻重的地位,无论政府在其中扮演的角色如何,巨灾保险都是在与市场结合的运作过程中最终得到检验的,在这一检验过程中,需求主体不仅扮演着市场始发者的角色,而且扮演着制度检验者的角色。其三,价格杠杆不能完全指挥巨灾保险需求,依靠需求自发调节的思路往往由于居民风险意识不足而受阻,这也使得巨灾保险需求问题具有相当的挑战性。围绕巨灾保险需求及其影响因素,本文进行了五个部分的研究,第一部分是巨灾保险需求的理论分析,从风险角度,微观角度,宏观角度探讨巨灾保险需求产生的现实原因,以不确定性经济学作为巨灾保险需求的理论基础,并提出了巨灾保险需求不足的一个较新的理论解释：消费不可逆假说的悖论。第二、三、四部分分别对洪水保险需求、地震保险需求、风灾保险需求进行了实证分析,在了解各类风险及其造成的经济损失和保险损失的情况下,分别运用时间序列和横截面数据分析这三类典型的巨灾保险需求及其影响因素,得出了各个因素的作用力方向和影响力大小。第五部分是结论和政策建议,对二至四章的实证分析进行了汇总和对比分析,得出了以风险意识创造需求,以有效供给启发需求,以合理制度引导需求的政策建议。关于巨灾保险需求的现实动因和理论基础,论文认为产生巨灾保险需求的现实动因源于三个方面：一是巨灾风险严重威胁人类的生命财产,二是家庭和企业风险管理体系的需要,三是国家安定和可持续发展的需要。对于巨灾保险需求的理论基础,论文分析了不确定性经济理论,认为外生不确定性是巨灾保险潜在需求的抽象,内生不确定性是巨灾保险现实需求的描述。值得一提的是,论文在分析中发现了一个特有现象：在巨灾发生的情况下,杜森贝利的“消费不可逆假说”存在悖论,这一发现构成了巨灾保险需求不足的解释原因。因为需求与消费是密切相关的,潜在需求催动了主体的消费动机,而有效需求必须以消费的形式体现出来。消费不可逆理论的主要思想是“由俭入奢易,由奢入俭难”,但在巨大的变故或灾难面前,人们的消费会迅速作出调整。我们将这种情况称之为消费不可逆性理论的“沸水悖论”,尤如青蛙能跳离沸水,人也能适应巨灾后的恶劣的生存条件,从而大大减弱了人们通过巨灾保险来保持原有生活水平的动力。洪水保险需求的实证分析部分,论文在总结洪水风险及其造成的经济损失和保险损失的基础上,选取美国三十年间的洪水保险保费收入、政府提供的家庭救助额度、居民储蓄、居民可支配收入、雇员参与社会保险的额度数据。进行了协整分析和格兰特因果检验。实证分析的结论是：政府救助、居民可支配收入、居民参与社会保险的额度对洪水保险需求的长期影响强于短期影响,社会保险的参保额度对洪水保险需求具有Granger因果性。对于储蓄变量,论文认为,以中国的国情看,储蓄的确可以代表居民手中的零风险资产,而对美国居民来说则未必如此,因而造成了变量的解释力不强。较为特殊的是,居民可支配收入与洪水保险需求互为Granger原因,当洪水保险需求增加时,居民可能付出额外的保险费用于管理洪水风险,从而导致可支配收入的减少；居民可支配收入增加时,导致洪水保险需求减少。这一点与传统认知是相悖的。一般来说,我们认为居民可支配收入的增加代表购买能力的增强,会刺激保险需求。但该实证结果促使我们从另一个角度考虑该问题：即把居民可支配收入作为“自保能力”的衡量指标,当居民的自保能力增强时,对洪水保险的需求会减弱。地震保险需求的实证分析部分,论文在总结地震风险及其损失特点,以及地震活动发生的的时间和空间规律的基础上,选取美国五十一个州的地震保险保费收入、政府提供的救助额度、居民可支配收入的横截面数据进行加权最小二乘法回归估计,并对汶川地震进行了案例分析。实证结果表明：政府救助会负面分化地震保险需求——需要说明的是,这里的地震保险需求是自愿的、市场化运作的需求,由商业保险公司提供的产品来满足。值得一提的是,实证分析中引入了一个虚拟变量,即该州近年是否发生过七级以上地震,实证分析的结果也表明,虚拟变量“近年是否发生过大地震”比较显著,表明人们的地震保险需求受非理性因素的影响,历史上没有发生大地震,人们不自觉的会低估该事件发生的概率；历史上发生过大地震,人们的巨灾保险需求则会极大提升。风灾保险需求的实证分析部分,论文在总结风灾在世界上造成的经济损失和保险损失,以及风灾在中国的危害的基础上,选取居民的购买力、政府的救济额度作为解释变量,以保持实证的一致性,对美国十七州的风灾保险金额进行加权最小二乘法回归估计。与前两章不同的,本章的实证分析选取保险金额作为被解释变量,而不是保险费作变量；同时,本章还增加了一个解释变量,即风险暴露的总金额,用以衡量风险与需求之间的关系。实证的结论是,风险暴露的财产金额对风灾保险需求有正解释力,风险暴露的财产的价值提升时,风灾保险需求增加；而政府救济则正向的影响风灾保险需求,这意味着政府救济增加,风灾保险需求也随之增加。与前两章的实证结果相反,也与我们的一般认知相悖。通常认为,在面对巨灾时,政府的救助和保险供给是此消彼长的关系,但从另一个角度讲,政府救助会提高低收入人群的购买能力,从而间接造成巨灾保险购买能力的增强。可见,各个变量的影响力及其作用方向都不是绝对的。在全文研究的结论部分,论文对三类巨灾风险实证的十五个指标进行了描述和对比。定性描述如下：‘是经济发展和人们购买力的增强,并不自然带动巨灾保险需求增加。居民可支配收入的增加,同时意味着自保能力的增强,在替代效应大于收入效应时,收入增加有可能成为分化和削弱巨灾保险需求的因素。二是政府救济通常会抑制巨灾保险需求,但论文的三个实证中有一个并不支持此结论。论文将此现象解释为：政府救济会提高低收入人群的购买能力,有可能间接造成巨灾保险现实需求的增加。这也说明,巨灾保险需求的各个影响因素及其作用力方向没有一定之规,而要与不同的巨灾保险险种、运作模式相结合。三是在巨大的变故或灾难面前,人们的消费会迅速作出调整,从而减弱了人们通过巨灾保险来保持原有生活水平的动力。四是历史上遭遇过大灾的地区,人们对巨灾保险的需求明显高于其它地区,而且这种影响是持续性的。五是风险暴露与巨灾保险需求呈正相关关系,即风险越大,需求越大。对中国来说,以风险意识创造需求,以有效供给启发需求,以合理制度保障需求,才能逐步构建供需平衡的巨灾保险市场,才能形成真正意义上的、包容了供给与需求双重和谐的巨灾保险体系,打造合理、可行的巨灾风险管理体系。本文的创新之处主要表现在三个方面,一是发现在巨灾发生的情况下,杜森贝利的“消费不可逆假说”存在悖论,这一发现构成了巨灾保险需求不足的解释原因。因为需求与消费是密切相关的,潜在需求催动了主体的消费动机,而有效需求必须以消费的形式体现出来。消费不可逆理论的主要思想是“由俭入奢易,由奢入俭难”,但在巨大的变故或灾难面前,人们的消费会迅速作出调整。我们将这种情况称之为消费不可逆性理论的“沸水悖论”,尤如青蛙能跳离沸水,人也能适应巨灾后的恶劣的生存条件,从而大大减弱了人们通过巨灾保险来保持原有生活水平的动力。二是本文的实证结果发现居民购买力的增强并不必然正向的影响巨灾保险需求,相反,还可能因为随之而来的自保能力的增强而抑制巨灾保险需求。因此,与传统观点不同,在拉动巨灾保险需求的措施中,我们主张淡化居民购买能力的作用,而取而代之以提升风险意识、提供有效供给、提出合理制度来促进巨灾保险需求。三是本文的实证结果发现,政府救助、社会保险等因素对巨灾保险需求的影响力及其作用方向都不是绝对的。传统观点认为政府救助与巨灾保险需求是此消彼长的关系,但从另一个角度讲,政府救助会提高低收入人群的购买能力,从而间接造成巨灾保险购买能力的增强。巨灾保险需求的各个影响因素及其作用力方向,并没有放之四海而皆准的定律,要与不同的巨灾保险险种、运作模式相结合。本文的不足之处有三点,一是因为数据获取的限制,本文实证使用的是美国的数据,虽然也结合中国的实际情况进行了理论上的拓展分析,但无法运用中国的数据进行实证,是本文最大的遗憾。二是对于消费不可逆假说的悖论,本文只进行了理论分析,没有深入进行实证检验。三是本文对巨灾的研究相对集中在自然灾害,而缺乏对人为灾害的研究。更多还原

【Abstract】 Catastrophe insurance is an important component of a country’s insurance system, and always has been a topic of concern. Frequency of catastrophe events in this century led to hot theoretical research about catastrophe insurance. However, most of the research in the field of catastrophe insurance is about supply, only little of them about demand. In this paper, from a new perspectives, we focus on the demand of catastrophe insurance.The paper studies the demand of catastrophe insurance and its influencing factors for three reasons. First, even in countries which have established a relatively perfect catastrophe insurance system, the system itself did not resolve the issue of catastrophe insurance participation. Inadequate demand for catastrophe insurance issues, and low catastrophe insurance participation rate have always been problems difficult to solve. Second, the demand for catastrophe insurance is of great importance in catastrophe insurance system design. No matter what role the government plays in the system, the operation of the process is in combination with the catastrophic insurance market, and will ultimately be tested in the market. Demand not only plays a role in the market of origin, and also play a role of testing. Third, the price can not fully leverage catastrophe insurance demand, which due to lack of risk awareness among residents blocked.Around the needs of Catastrophe Insurance and its influencing factors, this paper conducts a study of five parts:the first part is the theory analysis of catastrophe insurance needs from a risk point of view, the micro-perspective, a broad perspective. On basis of the Economics of uncertainty the paper puts forward a relatively new theoretical explanation on the lack of catastrophe insurance needs:the hypothesis of irreversible consumption paradox. Second, third and fourth parts of the paper gives an empirical analysis of the needs for flood insurance, earthquake insurance and typhoon insurance respectively in understanding the various types of risk and the resulting economic losses and insured losses. Using time series and cross-sectional means to analyze three typical catastrophe insurance demand and its impact factors. The fifth part is the conclusions and policy recommendations. It summarizes two to four part’s empirical analysis and gives comparative analysis,which come to the recommendations that creates demand by risk awareness, stimulates the demand by effective supply, introduce demand by reasonable system.Chapter1 explores the reality motives and theoretical basis of the demand for catastrophe insurance. The reality motives of the demand for catastrophe insurance are in three aspects. From the risk point of view, the catastrophic risk is of a serious threat to human life and property; from the microscopic point of view, catastrophe insurance is a important part in the family and the enterprise risk management system, from a macro vision, the catastrophic insurance is an important cornerstone of sustainable development. The theoretical foundation of demand for catastrophe insurance is uncertainty in the economy, in which exogenous uncertainty is the cause of demand for catastrophe insurance, and endogenous uncertainty is the description of demand for catastrophe insurance. It is worth mentioning that, this paper put forward in the theoretical analysis:in the catastrophic cases, the hypothesis of the existence of irreversible consumption paradox, which is one of the important reasons to explain the insufficient demand for catastrophe insurance. Because demand and consumption is closely related, the potential demand is the motivation of consumption, and effective demand must be reflected in the form of consumption. The main idea of irreversible consumption theory is that "extravagant easy entry by the thrifty, hard by the extravagance frugal", but in the great changes and even before the disaster, income consumers adjusts quickly. We have called the theory "The Paradox of boiling water." As the frog can jumped off the boiling water, people are able to adapt to the poor living conditions after catastrophe, which greatly weakened the people’s desire of catastrophic insurance, which can help to maintain their standard of living.Chapter2 is empirical analysis for flood insurance. Based on the character of flood risk and its economic losses and insured losses, the paper select data of three decades of flood insurance premium in the United States, the amount of government provided family assistance, household savings, disposable income of residents, employees participating in social insurance. The paper carries out cointegration analysis and Granger causality test. Empirical analysis concludes that:on the demand for flood insurance, the Government aid, the disposable income of residents, the residents participate in social insurance, long-term impact of the three are bigger than short-term impact. Social insurance influences the demand for flood insurance with Granger causality. The variable "savings" can reprsent zero-risk assets in China, while the condition in United States is diffrent, resulting in the explanatory power of the variables is not strong. More special is that the disposable income of residents and flood insurance requirements are Granger causes for each other. When the demand for flood insurance is increased, residents may pay an additional premium for flood risk management, resulting in the reduction of disposable income; when the disposable income of residents is increased, demand for flood insurance reduced. This is inconsistent with the traditional understanding. In general, we believe that the the increase of residents’ disposable income will enhance the purchasing power of residents and will enhance insurance demand along. However, the empirical results prompted us to consider from another point of view of the problem:that is, the disposable income of the residents can be regarded as "self-protection capacity" measure. When residents are more powerful in protecting themselves, the demand for flood insurance will be reduced.Chapter3 is empirical analysis for earthquake insurance. Based on characteristics of seismic risk and loss as well as the seismic activity occurred in the laws of time and space, the paper selects the cross-sectional data of the 51 U.S. states, including the earthquake insurance premium, the amount of assistance provided by the government, the disposable income of residents of the data. By weighted least squares estimation, empirical results show that:demnad for earthquake insurance is negatively correlated with the government aid, which should be noted is that here the demand for earthquake insurance is a voluntary, market-oriented and provided by commercial insurance products. It is worth mentioning that the introduction of a dummy variable in empirical analysis, that is, whether or not the state have occurred in recent years, more than 7.0 earthquakes. The results of empirical analysis also shows that the dummy variable "in recent years, a major earthquake has occurred or not" relatively significant, indicating that people’s the demand for earthquake insurance is influenced by irrational factors, such as the history of the earthquake. Where there is no earthquake before, people unconsciously underestimate the probability of the incident. By contrast, where earthquake occurred in history, people’s demand for catastrophe insurance is bigger.Chapter4 is empirical analysis for hurricane. Based on characteristics of typhoon and hurricane, as well as the economic losses and insured losses of them, the paper selects the data of U.S.17 states and do the weighted least squares regression estimation. Different from the previous two chapters, the empirical analysis in this chapter selected coverage as interpreted variable, instead of premium. At the same time, this chapter also added an addition variable, that is, the total amount of exposure. Empirical conclusion is as following. The demand of hurricane is positively correlated with the exposure of the property, which measures the risk of exposure; and also positively correlated with the Government assistance, which means that when the government assistance increases, the demand for hurricane insurance increases. This finding is contrast to the empirical results in previous two chapters, and is counter to our general knowledge. But from another perspective, we see that the government will help to improve the purchasing power of low-income people, which indirectly led to the enhance of purchasing power of Catastrophe Insurance. To conclued, all variables and their role in the direction of influence is not absolute.Chapter5 of the paper is conclusions and policy suggestions. The author gives a summary in intensive of positive analysis results, and draws a conclusion that:1. The development of economy and enhancement of residents’ purchasing power will not incentive Catastrophe insurance demand increase. On the contrary, the enhancement of residents’purchasing power will probably weaken Catastrophe insurance demand. 2. Social insurance promotes Catastrophe insurance demand which led by government, and not followed by the trade-off relationship between commercial insurance and social insurance traditionally.3. Government reliefs will weaken Catastrophe insurance demand to some extent and exist substitution relationship with Catastrophe insurance demand.4.Risk consciousness plays an important role in Catastrophe insurance demand.5.People’ decision on Catastrophe insurance is irrational (short-sighted), will apparently effected by whether Catastrophe occur or not. According to China, creating demand by Risk consciousness, inducting demand by effective supply, ensuring demand by sound system will build clear market of Catastrophe insurance step by step and form demand-supply harmonious Catastrophe insurance system.The innovations of the paper are as following. Firstly the paper finds the existence of paradox of Duesenberry "Consumer hypothesis irreversible", this discovery contributes to explain why the demand of catastrophe insurance is inadequate. Because demand and consumption is closely related, effective demand must be reflected in the form of consumption. In great misfortune or disaster, people’s consumption will be adjusted quickly. We will call this situation "The Paradox of boiling water". Like frogs jumped off in boiling water, people is also able to adapt to the harsh living conditions after the catastrophe, thus greatly weakened the driving force to maintain their original standard of living. Secondly the paper brings forward definitely that the enhancement of residents’ purchasing power does not surely give a positive impact on Catastrophe insurance demand, on the contrary, it may reduce demand for catastrophe insurance because of the increase of residents’self-insurance ability. Maintaining weaken the effect of the purchasing power of residents and substitute with enhancement of risk consciousness, effective supply and reasonable insurance institution to promote Catastrophe insurance demand in stimulating insurance demand measurements. Thirdly, the paper finds that government assistance and demand for catastrophe insurance do not follow trade-off relationship traditionally. All variables and their role in the direction of influence is not absolute.The limitation of the paper is as following. First, because of restrictions on data acquisition, China’s data is not available, so the paper used the US data, which only gives a general explanation of Catastrophe insurance demand. Lack of empirical data in China is the biggest regret in this article. The second point is the irreversible consumption paradox hypothesis, this paper only did theoretical analysis without empirical analysis. Meanwhile, the study of Catastrophe insurance comparatively focuses on natural disaster, and lacks the research of man-made disaster.更多还原