Empirical Analysis On The Efficiency Rate Of China Insurance Industry Based On Three-stage DEA Model

Since1980Insurance industry had recovered in China, the domestic Insurance market with unprecedented development speed has made considerable achievements. As one of the three pillars of the financial industry, the insurance industry is one of the important channels of capital flows, and the composition of social security. With the rapid development of our country’s economy, the enterprise demand for social capital circulation and management have become higher and higher, and people’s demand for risk protection of life is more and more urgent, the insurance industry to social function has been gradually revealed. However, as an emerging industry, the domestic insurance industry is still at the initial stage of development, the current situation of China’s insurance industry is not very desirable, and the overall operation efficiency is low, many problems exist there. Therefore, measure our country insurance industry efficiency level, evaluating the operational efficiency of market main body, studying the factors of restricting the development of the insurance industry in China and unfavorable conditions, seeking ways to improve the efficiency of insurance industry have the very vital significance.Data envelopment analysis (DEA) is a nonparametric method, and it is one of the most widely used methods to measuring efficiency. Three-stage DEA model, compared with the traditional DEA model, can eliminating the influence of environmental factors on the efficiency value, measures sample input and output efficiency. Specifically, on the first stage, put the decision making units’multiple input and output into the DEA model, then we can get the original input-output efficiency under the condition of value; on the second stage, we use the stochastic frontier approach (SFA) to get the regression relationship between each input slack variable and environment variables and to adjust the original input using the regression results for eliminating the influence of environmental factors; On the third stage, put the adjusted input variables in the second stage and the original output variables into the DEA model, we can get the adjusted input-output efficiency value without the impact of environment.In this paper, the empirical part of the research of insurance market in China in2011, using three-stage DEA model respectively to measure and evaluate the input-output efficiency of property insurance and life insurance companies, and analyzed the reasons of low efficiency in the insurance market.of our country and puts forward some improving measures.This paper respectively selects the samples from property insurance and life insurance market. According to the sample data availability, consistency and integrity, and ultimately selected the51home insurance companies, including31Chinese insurance companies and20foreign or Sino-foreign joint venture insurance companies;50life insurance companies, including27Chinese life insurance companies and23foreign or Sino-foreign joint venture insurance life insurance companies as the research object, using DEA model to evaluate2011annual insurance market efficiency. This paper selects capital, management cost and number of staff as input variables, investment income and payment and reserve has occurred as output variables, company size, and education rate, the company shares, equity, and age of the company as the environmental factor, studying property insurance and life insurance Company’s technical efficiency, pure technical efficiency and scale efficiency.The empirical results show that regardless of the insurance or life insurance, after adjusting for the environment, there is a slight rise of pure technical efficiency and scale efficiency has declined dramatically which causes the overall technical efficiency level declined obviously. Conclusions obtained from the empirical results are:one, in2011China’s insurance market efficiency value is low, and the whole industry is in a low stage of development; two, the main causes of inefficiency of technical efficiency in the insurance industry is the scale inefficiency; three, the Chinese insurance companies technical efficiency is higher than foreign insurance companies’. This paper puts forward several measures for solving all sorts of problems appeared in insurance industry in China, they are:firstly, we should build a monopolistic competition environment actively to promote a more efficient industry; secondly, we should expand the scale of the industry effectively and promote the scale effect; thirdly, we should encourage the development of the foreign capital insurance company and get a win-win results.The innovation point of this article is, first of all, this paper uses three-stage DEA model to evaluate the insurance market’s efficiency, this model can evaluate samples’real input-output efficiency,overcoming the shortcomings of traditional DEA model which cannot eliminate the environmental factors affect; Then, on the sample selection, this paper not only does the research on property insurance market, at the same time also pays attention to the life insurance market samples, and run the comparison of the two market which is very rare in other papers concerning insurance efficiency field; Finally, on the selection of the number of samples, this paper is different with other research only focusing on the typical samples, it includes almost all the property insurance and life insurance companies as possible. However, concerning the applicability of the data, this paper retain51property insurance companies and50life insurance companies which accounts for90%of their respective markets, these samples can reflect the efficiency of insurance companies of various of sizes and qualities. But, since my capacity and energy is limited, this paper also has many shortcomings. First, this paper only investigates the condition of2011China’s insurance market efficiency, lacking the efficiency value of the longitudinal comparison; Second, I did not do too much study about DEA model, the selection of variables is only on the basis of previous research experiences, theoretical problems such as the selection principle of the variables is without deep step discussion. I hope more and more scholars can focus on this specific field in the future.