An Improvement Of Auto Insurance Pricing Methods In Small And Medium-Sized Property Insurance Company

The market pricing of auto insurance rates has always been a hot topic in academic circles.However,the existing research tends to face all the property insurance companies and ignores the size differences of the property insurance companies,resulting in the strategy is not targeted.At the same time,the micro-data of empirical research is often limited to third-party auto insurance loss data in a Swedish insurance company,and less analysis is conducted from actual operation.The traditional pricing model of automobile insurance products with annual premium does not reflect the risk loss cost per mileage.Therefore,this paper uses the data of Company X in Guangdong Province,makes the driving mileage into the pricing model,and does research based on the auto insurance pricing for small and medium property insurance companies according to the GLM model.Firstly,through the analysis of the current situation,it is pointed out that although the auto insurance of X Property Insurance Company has not lost money under the background of the market-oriented auto insurance premium rate,it is facing the competitive pressure of large-scale property insurance companies with the increasing concentration of the industry.Customers are sensitive to the price;on the basis of the liberalization of independent pricing,only fine quotation can give customers the lowest price.The future of small and medium-sized property insurance companies lies in constantly improving the actuarial level,but the current pricing models of small and medium-sized property insurance companies have the following problems: it mainly carries on around NCD and excessively depends on channel sales;pricing model does not reflect the risk loss costs of mileage per unit.Then the paper applies the generalized linear model to the auto insurance pricing of X Property Insurance Company,and brings the mileage into the pricing model.The study shows that:(1)According to the existing insurance practice data,the corresponding risk variables are screened and classified.Four explanatory variables,including driving mileage,vehicle age,vehicle type and NCD coefficient(non-compensatory preference coefficient),have all passed the test.And it illustrates that these four risk factors need to be considered comprehensively in the pricing process.(2)The net premium calculated by the generalized linear model covers the claim amount,but the net premium only accounts for 31.5% of the premium income,indicating that the net premium used to compensate for the insured's compensation or payment is relatively low,while the additional premium used to cover the expenses of the insurance business process is too large;(3)It is better to introduce the mileage calculation pure premium than to introduce the traditional risk factor model on the discriminative power,which shows that introducing the mileage calculation pure premium into auto insurance pricing has higher efficiency.Finally,the paper puts forward countermeasures and suggestions:(1)Product differentiation.It includes expanding coverage endorsement,improving actuarial level so as to achieve product differentiation,and increasing value-added services to achieve service differentiation;(2)Reducing operating costs.It includes continuously reducing the cost of background management and selling expenses.(3)Actively developing mileage auto insurance products.As a more efficient and fair insurance product,the development of mileage-based auto insurance products can also be regarded as an effective way for small and medium-sized property insurance companies to change their development mode and promote the upgrading of product structure.