背景風險對家計單位風險趨避程度及保險需求之影響

Degree of Risk Aversion and Demand for Insurance of Households in the Presence of Background Risk

林豐騰(Feng-Teng Lin)

指導教授 ： 曾郁仁

國立臺灣大學/管理學院/財務金融學研究所/博士(2009年)

https://doi.org/10.6342/NTU.2009.03084

摘要

本論文分為兩部分，第一篇研究利用Halek and Eisenhauer (2001)之研究中所提出利用壽險資料推導出Arrow-Pratt風險趨避之縮減式，並以台灣家庭收支調查資料為樣本，實證估計家計單位的風險趨避係數（包括絕對風險趨避係數及相對風險趨避係數），估計結果與Halek and Eisenhauer (2001)之估計類似，相對風險趨避係數分配為右偏，多數集中在0至4之間。研究進一步探討背景風險是否為影響風險趨避程度高低之因素，背景風險定義為獨立於其他風險之不可投保風險，所得風險是最常用來代理背景風險的變數，本研究以所得變異係數來代表家計單位無法透過保險或其他避險方法來控制的背景風險，迴歸模型並加入了其他解釋變數包括家庭所得及資產等財富變數、戶長及家庭特性等社會變數及其他地理變數。主要實證結果顯示所得風險愈高其風險趨避程度亦愈高，表示樣本家計單位偏好符合Pratt and Zeckhauser (1987)之適當風險趨避(proper risk aversion)， Kimball (1993)之標準風險趨避(standard risk aversion)，和Gollier and Pratt (1996)之風險脆弱性(risk vulnerability)等理論之充分條件及必要條件。
第二篇研究主要探討家計單位保險（包括人壽醫療險及產物保險）之購買決策、消費比例及支出金額是否受到背景風險（以所得變異係數及標準差及減薪廠商比例等所得風險變數作為代理變數）之影響，並實證估計各種保險之所得彈性。利用不同之迴歸模型（包括Logistic、 Tobit及OLS等方法）並控制相關解釋變數包括財富變數及其他社會及地理變數後，實證結果顯示所得風險會正向影響家計單位的保險需求，面對愈高所得風險的家庭會有較高機率購買保險以及傾向購買較多保險，此結果與Guiso and Jappelli (1998)和Koeniger (2004)分別針對責任險及汽車險所做之實證分析結果相同，亦與Eeckhoudt and Kimball (1992)和 Schlesinger (1999)推導出之理論模型一致，家計單位偏好符合Pratt and Zeckhauser (1987)、Kimball (1993)和Gollier and Pratt (1996)之條件。研究並估計保險之所得彈性，實證結果顯示所得彈性為正，表示保險購買支出會隨著所得提高而增加，代表保險屬於正常財，此結論與其他相關保險實證研究結果一致。

關鍵字

風險趨避 ； 背景風險 ； 所得風險 ； 保險需求 ； 所得彈性

並列摘要

Essay 1 of this study uses life insurance expenditure data of Survey of Family Income and Expenditure (SFIE) in Taiwan to estimate the Arrow-Pratt risk aversion coefficient of households empirically by using the reduced form equation derived by Halek and Eisenhauer (2001). This study provides empirical evidence on the nature of the relationship between the risk aversion and background risk which is not under the control of the agent, and that is independent of endogenous risks. Using the coefficient variation of household income as the proxy for background risk, after controlling other factors including household income and wealth, the characteristics of the head of household and other demographic variables, the results suggest that households which are more likely to face higher income risk exhibit a greater coefficient of risk aversion. This finding is consistent with consumer preferences being characterized by proper risk aversion (Pratt and Zeckhauser, 1987), standard risk aversion (Kimball, 1993) and risk vulnerability (Gollier and Pratt, 1996) which are the necessary and sufficient conditions of the optimal risk-taking behavior in the presence of background risk.
Essay 2 of this study investigates how background risk affects households’ insurance purchasing decision, expenditure share and amounts of insurance by using data of Survey of Family Income and Expenditure (SFIE) in Taiwan. Using the income risk as the proxy for background risk and controlling other wealth and demographic factors, the findings suggest that insurance expenditure is positively affected by uninsurable background risk. This results suggest that consumer with more income risk is more risk averse and leads a higher demand of insurance. This finding is similar to the empirical results of Guiso and Jappelli (1998) and Koeniger (2004) and is consistent with the theory models derived by Eeckhoudt and Kimball (1992) and Schlesinger (1999). This finding is also consistent with consumer preferences being characterized by proper risk aversion, standard risk aversion and risk vulnerability. This study also finds that the coefficient income elasticity of insurance is positive that means people tend to increase insurance expenditure with respect to an increase in income. This result is consistent with most empirical studies of insurance demand that suggest that a consumer’s income change has positive effect on the consumer’s demand for insurance and suggest that insurance is a normal good.

並列關鍵字

Risk aversion ； Background risk ； Income risk ； Demand for insurance ； Income elasticity

參考文獻

Tzeng, L. Y., J. L. Wang, and A. Ho (1999). “Demand for Market Insurance and Self-Insurance Under an Increase in Risk Aversion,” Journal of Risk Management 1: 53-64. (in Chinese).

Ho, S. C., and L. Y. Tzeng (2002). “ Revisiting the Demand for Insurance,” Journal of Risk Management 4: 31-44. (in Chinese).

Arrow, K. J. (1965). “Aspects of the Theory of Risk Bearing,” Yrjö Jahnsson Lectures Yrjö Jahnsson Foundation, Helsinki.

Arrow, K. J. (1971). Essays in the Theory of Risk Bearing, Chicago: Markham Publishing Company.

Atkinson, A. B., J. Gomulka, and N. H. Stern (1990). “Spending on Alcohol: Evidence from the Family Expenditure Survey 1970–1983,” Economic Journal 100: 808–827.