Sources,Influencing Mechanisms And System Stability Of Non-life Insurance Underwriting Risk Correlation

The risk correlation between financial institutions is one of the core of modern risk management.Since the global financial crisis in 2008,regulators at home and abroad have paid unprecedented attention to the risk correlation of financial institutions.Internationally,the Financial Stability Board has incorporated "relevance" into its framework for identifying global systemically important banks.In addition,the Basel Committee on Banking Supervision and the International Association of Insurance Supervisors have developed methods to evaluate Global Systemically Important Banks and Global Systemically Important Insurance institutions respectively,in which "relevance" is one of the identification criteria.The same is true of China’s financial regulatory authorities.Since the subprime crisis in 2008,the word "relevance" has frequently appeared in various policy documents of China Banking and Insurance Regulatory Commission,emphasizing that great attention should be paid to the relevance of various risks and the relevance of various business operations.For example,when the management measures for reputation risk,liquidity risk and Enterprise Risk Management are put forward,it is clearly emphasized that the correlation with common risks such as credit risk,insurance risk and other risks should be fully considered.As for the implementation of the recovery and disposal plan of banking and insurance institutions,it is called for banks and insurance institutions to comprehensively sort out the complexity,relevance and substitutability of their business,so as to effectively improve transparency,reduce complexity,enhance selfrescue ability and prevent systemic risks.Therefore,"relevance" plays an important role in insurance industry regulation.For a long time,the existence of systemic risks in the insurance industry has been controversial.This has led to the obvious lack of attention from the regulatory authorities and the academic circles to the main business of insurance companies,such as the lack of attention to the underwriting risks,and the resulting risk correlation and risk spillover effects.However,it is undeniable that,on the one hand,as the contiguous characteristics of natural disaster events become more prominent,and as the population flow,goods flow,and vehicle traffic between provinces increase year by year,the risk of underwriting risks itself Relevance is increasing and becoming more complicated.On the other hand,the geographical distribution of underwriting business among insurance companies are become more and more similar,the structure of insurance types of underwriting business among different insurance companies also shows an increasing trends of homogeneity,and the similarity of the operation and management patterns of different insurance companies is also significantly increasing.This has led to closer links between insurance companies due to business risks underwriting.Therefore,ignoring the underwriting risk and the underwriting risk relevance and then talking about the systemic risk of insurance companies or the insurance industry will inevitably underestimate the risk spillover effects of insurance institutions,thereby weakening the actual impact of the insurance sector.Therefore,this article based on the practice of Chinese non-life insurance industry status quo of management,in a life insurance company accept insurance risks relevance between the research object,uses the qualitative analysis,theoretical analysis,empirical analysis and simulation method with the combination of a variety of ways to study the measurement of underwriting risk relevance,impact mechanism and its effect on the stability of the system.The main contents of this paper include:First,the correlation of underwriting risk among non-life insurance companies is measured based on Pearson correlation coefficient,which confirms that there is a significant correlation of underwriting risk among various insurance companies in China’s non-life insurance industry,and some companies have a large correlation of underwriting risk.Then,this paper explores the sources of underwriting risk correlation between non-life insurance companies from the connection of equity or management rights,to the possible business contacts,and then to the various links of underwriting business.The results of qualitative analysis show that common risk and reinsurance association are important sources of underwriting risk correlation,and they play roles through spatial diversification,diversification,co-insurance and direct or indirect reinsurance business respectively.Second,based on the perspective of common risk and reinsurance association,this paper firstly constructs a theoretical model of the influence of geographical distribution similarity and spatial correlation on the underwriting risks correlation in the spatial dimension.Then,extending the above model to the dimension of product structure,this paper intuitively analyzes the theoretical influence of the structural similarity of insurance product types and the product correlation on the underwriting risks correlation in the form of case analysis.The theoretical results show that the common risk in both the spatial and product dimension have significant positive impact on the underwriting risk correlation between non-life insurance companies.Then,based on the perspective of reinsurance association,this paper constructs a theoretical model of the impact of direct and indirect reinsurance association on the underwriting risks correlation.The theoretical model results show that,in the ideal state of frictionless market,the above two reinsurance association have significant positive impact on the underwriting risk correlation between non-life insurance companies,and they have the same effect.Finally,based on theoretical models and existing literatures,this paper constructs metrics for common risk and reinsurance association among non-life insurance companies,and measures them based on actual operating data.The measurement results show that the common risk among non-life insurance companies has a high concentration in the frequency distribution,but also shows a left-biased and fat-tailed phenomenon as a whole.In addition,reinsurance association between nonlife insurers appear to be driven by catastrophe risk.Third,based on the underwriting risk correlation perspective,this paper designs an empirical model and contribution degree model of the impact of common risk and reinsurance association on the underwriting risk correlation between non-life insurance companies,and conducts an empirical analysis based on the practical operation data of China’s non-life insurance industry.The results show that both common risk and reinsurance association have significant positive impact on the underwriting risk correlation between non-life insurance companies,and the former contributes more than the latter.Furthermore,the robustness test results show that the numerical differences of both common risk and reinsurance association in different measurement methods and measurement models do not significantly interfere with the magnitude and direction of their influence on underwriting risk correlation.Finally,the regression results of secondary indicators and the decomposition results of Shapley value show that the similarity of geographical distribution and the similarity of insurance structure are the key factors affecting the underwriting risk correlation,and the two factors have the greatest impact on the underwriting risk correlation.In addition,direct reinsurance connection and indirect reinsurance connection are the key factors affecting the underwriting risk correlation between non-life insurance companies,and the former has a significantly greater impact than the latter.Finally,based on the perspective of system stability of non-life insurance industry,a network model of underwriting business risk transmission in China’s non-life insurance industry is constructed based on available data.Based on the network model,the dynamic process of underwriting risk transmission and contagion in the insurance system is theoretically analyzed.The system stability measurement index at the industry level,the portfolio stability state measurement index at the company level and the stability characteristic index in the process of risk dissemination are constructed.Then,the simulation results based on a particular shock process show that large companies lose the most in the contagion process,followed by reinsurance companies,medium and small companies lose the least,and the risk loss of the previous round is several or tens of times larger than the risk loss of the next round.Furthermore,with the increase of external loss shock,it can be found that the characteristics of system stability,such as contagion breadth,contagion depth,contagion intensity and contagion density,generally show a trend of increasing first and then decreasing.Furthermore,the simulation results based on the underwriting business risk network model in each year show that the underwriting system stability of the non-life insurance industry generally shows a decreasing trend over time,and there is a significant negative correlation between underwriting risk correlation and system stability.Finally,based on the linear regression model,we examine whether underwriting risk correlation,common risk and reinsurance connection have significant influence on the stability of corporate portfolio.The regression results show that only the reinsurance connection has a significant positive impact on the stability of corporate portfolio before the risk contagion process begins after the occurrence of external loss shock.In the process of risk contagion,both underwriting risk correlation and reinsurance association have a significant negative impact on corporate portfolio stability,while common risk has a significant inverted U-shaped nonlinear impact.After the risk propagation process is over,the underwriting risk correlation has a significant negative impact on the final stability of the company portfolio,and the impact of common risks on the final stability of the company portfolio is a significant inverted U-shaped nonlinear impact.However,the effect of the reinsurance association on the final stability state is not significant due to the offsetting.