
DiversificationThe example on the Portfolio Risk and Return page illustrated that a portfolio formed from risky securities can have a lower standard deviation than either of the individual securities. On this page we shall explore this concept further to demonstrate that the benefits of diversification, i.e., the reduction in risk, depends upon the correlation coefficient (or covariance) between the returns on the securities comprising the portfolio. Consider stocks C and D. Stock C has an expected return of 8% and a standard deviation of 10%. Stock D has an expected return of 16% and a standard deviation of 20%. The concept of diversification will be illustrated by forming portfolios of stocks C and D under three different assumptions regarding the correlation coefficient between the returns on stocks C and D. Case 1: Correlation Coefficient = 1The table below provides the expected return and standard deviation for portfolios formed from stocks C and D under the assumption that the correlation coefficient between their returns equals 1.
When the correlation coefficient between the returns on two securities is equal to +1 the returns are said to be perfectly positively correlated. As can be seen from the table and the plot of the opportunity set, when the returns on two securities are perfectly positively correlated, none of the risk of the individual stocks can be eliminated by diversification. In this case, forming a portfolio of stocks C and D simply provides additional risk/return choices for investors. Case 2: Correlation Coefficient = 1The table below provides the expected return and standard deviation for portfolios formed from stocks C and D under the assumption that the correlation coefficient between their returns equals 1.
When the correlation coefficient between the returns on two securities is equal to 1 the returns are said to be perfectly negatively correlated or perfectly inversely correlated. When this is the case, all risk can be eliminated by investing a positive amount in the two stocks. This is shown in the table above when the weight of Stock C is 66.67%. Case 3: Correlation Coefficient = 0The table below provides the expected return and standard deviation for portfolios formed from stocks C and D under the assumption that the correlation coefficient between their returns equals 0.
When the correlation coefficient between the returns on two securities is equal to 0 the returns are said to be uncorrelated. In this case, some risk can be eliminated via diversification. Notice that when the weight of Stock C is between 100% and 60% the portfolios have a higher expected return than Stock C and a lower standard deviation than either Stocks C or D. This is depicted in the graph by the inward curve in the opportunity set. The Real WorldIn practice, the correlation coefficient between most stocks ranges between 0.5 to 0.7. When this is the case, the opportunity set will have a similar shape to that shown in the case in which the returns were uncorrelated. Thus, risk can be reduced via diversification. You can utilize the Two Asset Portfolio Calculator to explore this relationship. Moreover, the benefits of diversification increase as more stocks are added to the portfolio.
© 2002  2010 by Mark A. Lane, Ph.D.
