The Relationship between Risk and Return: Evidence for Life Insurance Stocks

This paper examines the relationship between realized mean returns and alternative measures of risk for samples of life insurance stocks during the 1961-76 period within the framework of the Capital Asset Pricing Model (CAPM). The results provide some evidence of a significant relationship between mean returns and systematic risk, but they also provide evidence of a significant relationship between mean returns and measures of nonsystematic risk, in contradiction to the principal implication of the CAPM. According to normative financial theory, the rate of return required by shareholders is of critical importance to investment and financing decisions of publicly-held firms. Knowledge of factors that determine required rates of return is needed for rational decision-making. Theorists agree that most investors are risk averse so that required rates of return will be positively related to risk, but there is less than complete agreement on the relevant measure(s) of risk. The Capital Asset Pricing Model (CAPM) of Sharpe [31] and Lintner [ 18] and the zero-beta version of the CAPM developed by Black [ 1] imply that the proper measure of risk for an asset is its market (systematic) risk as measured by beta and that diversifiable (nonsystematic) risk will not affect required rates of return. Thus, the CAPM predicts that realized returns on financial assets should be independent of nonsystematic risk measures, such as return variance, once the influence of beta has been removed. The CAPM has been subject to extensive empirical analysis. Studies by Black, Jensen, and Scholes [2], Fama and MacBeth [9], and Foster [13] suggest that mean realized returns on common stocks are significantly related