【作者】 张颖；

【导师】 刘金全；

【作者基本信息】 吉林大学 ， 数量经济学， 2010， 博士

【摘要】 保险业作为金融体系重要的组成部分,认识并揭示我国保险的发展规律已经成为保险经济学研究的核心内容。我国保险发展是否有规律可循、是否存在周期性的波动、是什么原因导致了保险周期性波动以及保险发展与宏观经济之间存在怎样的关联性是目前保险经济学研究的前沿问题。为此,本论文以现代保险相关理论为基础,以我国保险发展实践为出发点,对我国保险发展规律的周期性与影响因素进行了系统研究。本论文所做的主要工作体现在以下几个方面:一、利用谱分析方法检验我国财产-责任保险是否存在承保周期、度量保险承保周期的周期长度,并进一步利用ARFIMA模型、FIGARCH模型和ARFIMA-FIGARCH模型测度我国财产-责任保险价格序列的均值过程及其波动过程的长记忆性。二、利用协整检验、VAR模型、Granger因果关系检验和方差分解方法采用我国非寿险行业和险种数据,测度我国保费与损失之间短期与长期交互影响的动态传导机制,检验保费、赔款支出、利率、利润和不确定性变量等变量之间的相互作用关系,揭示我国保险承保周期波动的原因。三、采用3区制马尔科夫区制转移模型分别检验我国保险业保费收入增长序列以及分别比较了我国财产-责任保险业和人身保险业保费收入增长序列的动态波动路径,检验发现我国保险增长过程存在内生性结构转变,可以划分为“高速增长区制”、“低速增长区制”和“适速增长区制”,考察了我国保险承保周期在不同区制水平下表现的波动特征。四、利用各种非对称ARCH模型检验外部冲击对我国保险增长的影响,检验结果发现,外部冲击对我国保险增长波动的影响呈现明显的非对称性,具有杠杆效应,其中正向冲击对于我国保险增长波动的影响大于负向冲击对于我国保险增长波动的影响。五、利用协整检验、因果关系检验和脉冲响应函数分析方法测度我国保险发展与经济增长之间短期与长期交互影响的动态关联性。结果表明,保险发展与经济增长之间存在长期均衡关系,二者之间具有一定程度的顺周期性质,当经济运行处于稳定增长时,保险发展与经济增长的扩张方向基本一致。保险发展与经济增长之间具有相互促进作用,保险业对经济的促进作用以供给导向模式为主。更多还原

【Abstract】 Insurance industry is one important component of financial system, and the underwriting cycle of insurance industry development has been the subject of much recent discussion of insurance academics. The frontier on insurance theories are to study the presence of the underwriting cyclical ,the causes and the affecting factors of cyclical fluctuations as well as how insurance industry intact with macroeconomic. This dissertation’s aim is to systematically study on the business cycle and the affecting factors in China’s insurance industry development. The main contents and conclusions are as following:1. Spectral analysis method is applied to test the existence of underwriting cycle and measure the cycle length in property-liability insurance for China. Then ARFIMA model, FIGARCH model and ARFIMA-FIGARCH model are applied to examine whether there is long memory property in unit price and its volatility by using of the monthly unit price which is calculated by using the inverse of the loss ratio from January 1999 to November 2009. Empirical results shows that China’s property-liability market has 4.95 years of underwriting cycle. Moreover, there are many frequent fluctuations between in underwriting cycles, with volatility length from 2.4 months to 10.8 months. It is also found that unit price level of China’s property-liability is equivalent to the 1950s level of many other countries such as USA, Switzerland and Germany. This means profit margins of China’s property-liability insurance industry should be relatively large. However, the persistent losses indicate that operations of China’s insurance industry are not standardized, the market is imperfect and operation costs are relatively high. Therefore, we must enhance the core competitiveness of insurance enterprises and promote China’s insurance market to a rational and mature stage from its initial immature stage. ARFIMA model, FIGARCH model and ARFIMA-FIGARCH model are used to study the long memory effect of unit price. The first moment of the unit price shows no long memory property, but the second moment of the of the unit price shows significant long memory properties. In addition, by using the ARFIMA-FIGARCH model, it is found that there is no long memory property in unit price, but the long memory properties are significant in the volatility of the unit price. In consideration of the Student-t distribution, the results further prove that the unit price sequence has the characteristics of high peak and fat tail.2. The short-term and long-term dynamic relationships between premiums and losses are to studied to find out what causes fluctuations in underwriting cycles of China’s non-life insurance. This is achieved by conducting a cointegration analysis,VAR model, the impulse response function and Granger causality tests on a unique set of industry and line data. The results from the tests suggest that the there is a positive correlationship between premiums and losses. Loss variable, especially future losses forecast errors contribute the significant portions of premiums variable. The empirical results support both capacity constraint theory and rational expectation theory. This indicates that the development of China’s non-life constraints by its own solvency and the loss ratio increase is an important reason leading to higher premium. It is also found that interest rate is another factor which leads to volatilities of underwriting cycle of most lines. However, the impact of interest rate is not significant and the results do not support the theory that interest rate is more important for long-tail lines than short-tail lines. In addition, the uncertainty variable explains significant portions of forecast errors of premiums, especially the act insurance (for example agricultural insurance). For different lines of insurance, the main factors that determine premiums are different. The overall results indicate that no single hypothesis is able to explain the insurance cycle.3. Markov regime-switching model is applied to investigate the dynamic volatility path of China’s insurance growth from January 2000 to August 2009. According to the result of estimation and test, it is found that there are structure breaks in China’s underwriting,i.e. the“rapid growth regime”, the“low growth regime”and the“temperate growth regime”. The continuity and asymmetric characteristics in volatility path of underwriting cycle in different regime are also quite different. The testing results suggest that the underwriting cycles are pro-cyclic. Besides that, the insurance industry possesses, to some extent, the nature of following the same direction as the expansion and contraction path of the whole economic cycle. The volatility of property-liability insurance underwriting cycle is more acutely than that of life insurance. Besides that, the life insurance underwriting cycle possesses, to some extent, the nature of the continuity and following the same direction of path of insurance underwriting cycle. The response of insurance industry to external economical impulse shows lag effect. Moreover, the growth regimes of China’s insurance industry are closely related to the economic policy acts. Property-liability underwriting cycle is mainly affected by fiscal policy while life insurance underwriting cycle is mainly affected by monetary policy. This means that macroeconomic policies can affect the growth rate of insurance industry dynamically. In addition, life insurance has played a more important role to promote the development of insurance industry since 2004. At the same time, the testing results indicate that there exist irrational problems in the structure of life insurance lines.4. ARCH model with asymmetry is used to test the impact of external shocks on China’s Insurance by using the growth rate of China’s premium data. A clear asymmetry or a leverage effect in the external shocks is observed in China’s Insurance development. Positive impacts increase the volatility of China’s insurance significantly while negative impacts affect the volatility of China’s insurance much less. These results shows that“good news”promote the growth of insurance industry more than“bad news”decreases the growth of insurance industry and suggests that China’s insurance industry has a strong ability to withstand negative shocks. Therefore, in the period of rapid economic growth, we should speed up the pace of development of the insurance industry to shorten the expansion stage; in the period of economic contraction, we should effectively change the extensive growth mode and promote China’s insurance industry to more mature.5. By econometrically analyzing the interaction between insurance industry development and economic growth, the correlation between nominal economy and real economy is tested. The rapid growth of insurance industry can prompt economic growth and the social welfare state. At the same time, economic growth also accelerates the growth of the insurance industry. Economic growth and growth of insurance industry in China shows clear dynamic correlation in both short-run and long-run. The results is achieved by conducting a cointegration analysis,the impulse response function and Granger causality tests on a set of quarterly real GDP and total real premiums data from 1999 Q1 to 2009 Q1. The results from the tests suggest that there is a positive correlationship between insurance industry and economic growth level. Therefore, insurance industry possesses, to some extent, the nature of following the same direction as the economic growth, especially when the economy is in the state of a stable development. Insurance industry and economic growth promote each other. Moreover, the results indicate that the insurance industry’s promotion to economic growth is supply-leading mode. Therefore, their positive interaction between insurance industry and economic growth is very important to help fast and steady economic growth.更多还原