【作者】 左斐；

【导师】 田玲；

【作者基本信息】 武汉大学 ， 金融学， 2009， 博士

【副题名】Special Attention to Catastrophe Risk

【摘要】 本文以“中国财产保险业承保能力研究”为题,关注的是中国财产保险业供给层面的问题。中国社会的巨大变迁已使其步入一个风险高度累积的历史阶段,此种环境中,商业保险在中国社会重大损失事件中发挥的作用却远远低于人们的期望值。而从其他面临巨灾损失威胁的国家的实践看,商业保险适度突破风险可保性要件的范围,主动承保巨灾风险已不仅仅是保险业自身持续发展的需要,更是保险业承担更多社会责任,为社会重大损失事件提供必要补偿渠道的必然结果。顺此逻辑,中国保险人尤其是财产保险人本身的供给能力能否满足巨灾损失补偿的现实需要就是一个必须要回答的问题。2008年下半年开始席卷全球的金融危机给保险业带来了不小的冲击,这场危机及其引发的新一轮退保潮让世界范围内的保险人重新认识到了荒废保险保障功能的风险,一些大型企业公司开始致力于摆脱对投资型保险产品的依赖,将注意力从保费规模的账面增长转移到高收益保障型产品的开发上,行业整体出现了回归保障的潮流。在这样的现实中,保险业究竟可以提供多少保障,尤其针对巨型损失可以提供多少保障,也是一个不能回避的问题。巨灾保险的可获得性和可负担性问题的回答,以及保险保障程度的提升和改善,都必须首先明确保险供给的决定及衡量。而保险供给与保险人的风险承担能力正相关,风险承担能力取决于保险人资本和盈余的承保能力,因而承保能力是保险人或者保险业供给能力的直接体现。所以,在(巨灾)保险需求存在且稳定的基本假定下,本文研究的目的就在于通过对中国财产保险业总体承保能力,特别是巨灾风险承保能力的实证分析,得出对行业供给能力充足情况,以及行业内各种因素对供给能力状况影响及其程度的客观认识,并在此基础上,提出改善和提升中国财产保险业供给能力的建议。全文的主要内容和结构安排如下：第一章是对保险承保能力研究成果的一个综述。这一章系统梳理了保险人、尤其是财产保险人承保能力的相关研究,介绍了多个方法和模型的背景、主要思想、相比之前研究的意义和局限,不同方法和模型之间的相互联系和区别,以及本文实证分析选用模型Nielson (1984)和Cummins等(2002)的理由和适用性。第二章以“中国财产保险业承保能力的总体性分析”为题,是从中国财产保险业所处的国际和国内环境入手对行业承保能力的概括分析。在这一章中首先总结了国际金融危机和中国当前的经济形势对中国财产保险业承保能力的影响,随后主要以肯尼系数和自留承保能力等静态财务指标总体衡量了财产保险行业的承保能力,得出了行业承保能力不足的初步结论。第三章以“中国财产保险业承保能力模型度量”为题,采用Nielson (1984)模型,在财产保险人保险和投资业务一体化的框架内,对2008年中国财产保险业的理论承保能力进行了测度。结果发现,在破产概率不大于0.01,投资回报率不低于4%的假设前提下,中国财产保险行业在2008年理论上可承保的保费约为146.19亿元。扩大破产概率和投资回报率要求参数的取值范围之后发现,破产概率要求越高(破产概率越小),行业总体在理论上可以承保的保费越少；行业投资回报率要求越高,行业总体所能承保的保费越少。通过对比该理论度量结果与实际巨灾损失得出了财产保险业承保能力尤其是巨灾风险承保能力极度短缺的结论；通过对比该结果与2008年行业实际总保费发现,全行业有超承保能力承保的倾向,从而佐证了本文第二章从财务指标分析得出的结论。第四章以“中国财产保险业巨灾风险承保能力度量”为题,是专门对行业所能承受的极限损失幅度和应对极端损失的能力的评估。实证分析选用Cummins等(2002)模型,首先在损失正态分布假定下对2008年年初时点上中国财产保险业巨灾风险承保能力进行了度量,结果发现在800到2000亿元的给定巨灾损失幅度下,全行业巨灾风险承保能力比率在57.7%到87.7%之间；随后,本文针对此度量中损失分布假设的简化处理,对1987-2007年中国大陆洪水和地震损失数据进行了实际拟合,发现巨灾损失数据更加接近对数正态分布,于是基于引入损失对数正态分布假设后改进的模型再次考察了巨灾风险承保能力,结果发现行业巨灾风险承保能力比率上升至65.45%之上,整个行业应对巨灾损失的能力缺口在200-700亿元之间。第五章以“中国财产保险业承保能力不足的影响因素实证分析”为题,是在之前实证分析得出的承保能力总体不足结论的基础上,对造成这一结果的各种因素的影响方向及程度的实证分析。分析采用线性回归模型,从财产保险行业和企业两个层面进行了考察。结果发现,从行业层面看,保险深度、市场集中度、行业投资业务情况、行业费用情况、行业供给主体数量、财产保险的产品结构以及再保险的发展对行业整体承保能力不足这一结果的影响总体是显著的,但各个因素影响的方向和程度不一：市场化程度较低是中国财产保险市场承保能力不足的最主要原因；承保业务与投资业务结构的不合理是导致行业总体承保能力短缺的又一重要诱因,但影响程度较市场化程度较低；行业集中度过高仍然是中国财产保险行业的痼疾；中国财产保险市场险种结构的多元化以及保险业投资渠道的逐渐放宽对扩大行业总体承保能力是有益的。从企业层面看,行业投资、费用、再保险的运用以及资本实力的绝对量指标对行业承保能力影响的显著程度要高于相对量指标。第六章以“对中国财产保险业承保能力调整与改善的建议”为题,是对前文实证研究的过程以及结果的总结和应用。本文认为,在当前中国财产保险行业一般承保能力和巨灾风险承保能力总体不足的情况下,比单纯增加资本金更重要的,且从长远来说对行业总体承保能力贡献更大的是对财产保险行业进行结构调整。相比通过上市、发行次级债和单纯增资,科学的结构调整和有效的管理更能培养行业供给可持续增长的能力。具体应在财产保险的产品结构、财产保险企业的业务结构以及财产保险市场结构三个层面进行调整。此外,应当加快再保险的发展步伐,提高再保险的运用程度,并积极创造条件,引入以巨灾风险债券为代表的保险风险证券化产品提高财产保险业巨灾风险承保能力。本文研究的贡献主要体现在实证分析中：第一,在国内同主题研究中首度在一个同时考虑保险人投资业务与承保业务关系的一体化模型框架中得出了保险业承保能力度量的结果；第二,在巨灾风险承保能力的专门度量中,在更长的时间区间和更大的样本范围内,以及改进的损失分布假定下得到了巨灾风险承保能力及其缺口大小更精确的度量结果；第三,基于实证分析得出的承保能力不足的结论,运用回归分析方法,考察了多个因素对承保能力不足这一结果影响的方向及程度的差异。更多还原

【Abstract】 This dissertation, titled as "Study on Capacity of China Property and Liability Insurance Industry", is a research on the supply side of China Property and Liability Insurance Industry.China has turned into a high-risk society that is characterized by swiftly increasing loss volume and complicated loss circumstances. In this kind of environment, however, insurance industry as a whole is far below expectation in the compensation of significant loss events in this country. The practices of other nations that suffer from natural or manmade disasters similar to China indicate that insurance industry has been playing an ever increasing part in loss finance. If the insurance industry is to play a more vital roal in China, then it is non-avoidable a question that whether the supply capacity of China insurance industry can be accounted on to deal with catastrophic losses.The global financial crisis began in the second half of 2008 has brought fundamental influence to world insurance industry, one of the most significant outcome is to make the importance of coverage feature reoccurred to insurers and reinsurers worldwide. After the break of the crisis, many big insurance firms switched their attention from investment instrument to the inventing of high-coverage insurance product; the back to coverage trend becomes more and more evident. As for insurance market of China, after the third quarter of 2008, all listed firm started to make change to their investment structure, lowering their stock and fund holdings and increasing the bond and deposit portion. It can be seen that raising the coverage level has became a trend in the future development of insurance industry. Under circumstance like this, it became urgent to answer the question that how much coverage on earth insurance industry can actually provide, especially for the needs of catastrophe loss compensation.Therefore, it is clear that the answer to the question of insurance availability and affordability, as well as the enhancing of insurance coverage both calls for the precise measuring of insurance supply capacity. Insurance supply is positively related to insurers’ capability to underwrite risk, which in turn, is dependent on the capacity of insurer’s capital and surplus, then it can be seen that capacity functions as the direct reflection of insurance supply.The purpose of this dissertation, then, is to investigate on the capacity amplitude of China Property and Liability insurance industry through rigid empirical analysis, and bring forth according advices on the improving and enhancing of it. The main assumption of this research is that the insurance or catastrophe insurance demand exsits and is stable. Empirical analysis is the main research method employed in this dissertation. The content and structure of this dissertation is organized as follows:ChapterⅠof this dissertation is a review on insurance capacity literature. In this chapter, the author summarized, compared and evaluated the methods and models on capacity measuring, and explained the reasons why Nielson (1984) and Cummins et.al (2002) models are chosen as the measuring models in this dissertation.ChapterⅡ, titled as "General analysis on Capacity of China Property and Liability Insurance Industry", is a preliminary investigation on this issue in a wide foreign and domestic economic environment. The impact of the Financial Crisis on global insurance industry as well as on domestic market are brought forward in the first place, and then the measuring of capacity using Kenney Ratio and other financial index are conducted, general conclusion in this chapter’s empirical analysis is that the capacity of China Property and Liability insurance industry is not sufficient, and also there is overrun of capacity in the industry.ChapterⅢof this dissertation, titled as "Modeled Measuring of the Capacity of China Property and Liability insurance industry", is the first part of modeled analysis on industry capacity. The model utilized in this part is Nielson (1984), which is a comprehensive capacity measuring model taking into account both the investment and insurance practice of property insurer. The results turn out that under the assumption that industry bankruptcy probability below 0.01 and investment return above 4%, the theoretical premium volume that the industry can write in year 2008 is around 14.62 billon RMB. Through the comparison of this theoretical result to actual premium and loss volume in 2008, conclusions can be drawn that the capacity of the industry is limited, especially for the underwriting of catastrophe risk, and again, the industry overruns its capacity limit.ChapⅣ," Modeled Measuring of the capacity of China Property and Liability Insurance Industry to Rspond to Catastrophic Loss" is a specialized empirical analysis on the maximum capacity the industry can have in year end 2007. The model utilized in this Chapter is Cummins et.al (2002), which is a capacity measuring model from the claim perspective. The main output is that, under loss volume in the rage of 80 to 200 billon RMB, the industry can pay up 57.7% to 87.7% of the total loss. Then based on the simulationresult of 1987-2007 earthquake and flood loss data of China, the lognormal loss assumption is applied to the measuring model, and the new output using lognormal assumption turns out that the industry can pay up65.45% to 100% of the total loss, but as the loss volume increases, the capacity shortage increase disproportionately.Based on the result of empirical analysis conducted in previous chapters, Chapter V, titled as "Empirical Analysis on Factors Influencing Capacity Shortage", is the exploration on the causes and their extent of impact on Capacity Shortage in the industry. Linearity model is assumed and regressed form both the industry and the corporation aspect in this part. The main conclusions include:in the insurance industry level, the density, depth, concentration, investment, expenditure, supply number, product structure and the development of reinsurance as a whole significantly caused the capacity shortage, yet the extent of the impact are different; In the corporation level, investment, expenditure, reinsurance and capital in their absolute values influences the capacity more significantly than in their relative values.The final chapter, Chapter VI, titled as "Suggestions on the Adjusting and Enhancing of Property and Liability Insurance Industry Capacity " is the summary and application of the empirical analysis conducted in previous parts. Research in this dissertation suggest that the for the enhancing of industry capacity, structural adjusting is far more essential than the increase of capital and surplus. Specifically, property and liability insurance product structure adjusting, operation structure adjusting and market structure adjusting all need to be carried out. And also, reinsurance as well as insurance securitization can potentially function more to solve the problem of capacity shortage, so various conditions should be satisfied to stimulate the progress in these practice.The main innovation in this research include:firstly, this dissertation conducts a capacity empirical analysis considering the relationship between underwriting and investment operation of China property insuers, this is rare in domestic literature; Secondly, in the part of catastrophe insuanrce capacity empirical analysis, this dissertation improved the loss distribution assumption and conducts a through analysis under this newly improved assumption in a wider time and sample rage; Thirdly, Based on the limited capacity conclusion drawn in empirical analysis, this dissertation conducts a factor analysis using regression model.更多还原