Efficiency Measurement of the Life Insurance Markets with Three-Stage DEA Model

Abstract

Efficiency measurement has become popular in the life insurance industry in recent years. Regulators can better adjust the regulatory indicators of the life insurance market by looking at the relevant life insurance efficiency score; investors can determine whether they should invest based on the efficiency of the life insurance market, and company managers can adjust the relevant input value based on the efficiency analysis results. According to Wise (2017), over 190 studies on the efficiency measurement of the life insurance industry were published between 1983 and 2015. To measure efficiency, there are two main methods: stochastic frontier analysis (SFA) and data envelopment analysis (DEA). Some researchers, however, pointed out that the traditional DEA model failed to account for the impact of environmental effects and statistical noise on decision-making units (DMUs). In this thesis, we use the Three-stage data envelopment analysis model to evaluate the efficiency score of 18 selected OECD life insurance markets from 2014-2019 and analyze them. In the first stage, we use the basic DEA model to get the original efficiency score; and in the second stage, we use stochastic frontier analysis regression to remove the impact of environmental effects and statistical noise on the original efficiency score; after the adjustment according to the second stage, we recalculate the efficiency score of each market. Based on the results, we find out the specific ways to improve the efficiency of the inefficient life insurance market and find out how the relevant environmental factors affect the efficiency of the life insurance market. And we examine the relationship between technical efficiency and company size, profitability, and solvency by using a multiple linear regression model. Our main conclusions are that the German life insurance market is the most efficient, followed by the Swedish life insurance market, Irish life insurance market and Italian life insurance market; and the Lithuanian life insurance market is the least efficient, followed by Slovenia life insurance market, Greece life insurance market and Hungary life insurance market; most life insurance markets with increased total factor productivity are due to increased technical efficiency; most life insurance markets with decreased total factor productivity are also due to decreased technical efficiency; and life insurance markets with large size, high profitability and high solvency are usually more efficient. The contribution of this thesis is to produce more precise results by using more comprehensive data and analyzing life insurance markets’ efficiency at more angles and levels.

Efficiency measurement has become popular in the life insurance industry in recent years. Regulators can better adjust the regulatory indicators of the life insurance market by looking at the relevant life insurance efficiency score; investors can determine whether they should invest based on the efficiency of the life insurance market, and company managers can adjust the relevant input value based on the efficiency analysis results. According to Wise (2017), over 190 studies on the efficiency measurement of the life insurance industry were published between 1983 and 2015. To measure efficiency, there are two main methods: stochastic frontier analysis (SFA) and data envelopment analysis (DEA). Some researchers, however, pointed out that the traditional DEA model failed to account for the impact of environmental effects and statistical noise on decision-making units (DMUs). In this thesis, we use the Three-stage data envelopment analysis model to evaluate the efficiency score of 18 selected OECD life insurance markets from 2014-2019 and analyze them. In the first stage, we use the basic DEA model to get the original efficiency score; and in the second stage, we use stochastic frontier analysis regression to remove the impact of environmental effects and statistical noise on the original efficiency score; after the adjustment according to the second stage, we recalculate the efficiency score of each market. Based on the results, we find out the specific ways to improve the efficiency of the inefficient life insurance market and find out how the relevant environmental factors affect the efficiency of the life insurance market. And we examine the relationship between technical efficiency and company size, profitability, and solvency by using a multiple linear regression model. Our main conclusions are that the German life insurance market is the most efficient, followed by the Swedish life insurance market, Irish life insurance market and Italian life insurance market; and the Lithuanian life insurance market is the least efficient, followed by Slovenia life insurance market, Greece life insurance market and Hungary life insurance market; most life insurance markets with increased total factor productivity are due to increased technical efficiency; most life insurance markets with decreased total factor productivity are also due to decreased technical efficiency; and life insurance markets with large size, high profitability and high solvency are usually more efficient. The contribution of this thesis is to produce more precise results by using more comprehensive data and analyzing life insurance markets’ efficiency at more angles and levels.