Borrowing and Lending Rates
Equally unrealistic is the assumption of identical borrowing and lending rates for the investor. The risks involved in lending money to the federal government are less than the risks of lending money to ordinary investors, and investors therefore pay higher rates of inter est on borrowed funds than they receive through investment in riskless.
The amount of reduction in the slope of the line beyond the point of tangency obviously depends upon the magnitude of the difference between the borrowing rate for the investor and the lending rate, and thisdifference depends in part upon the credit rating of the investor. It is also realistic to acknowledge that the rate paid by the investor depends in part on the amount borrowed. This results in an extrapolation beyond the point of tangency which is curvilinear rather than linear.
The most visible professionally managed portfolios are mutual funds, and it is not surprising, therefore, that most research in the field of investments relating to portfolios is based upon mutual funds. Earlier, in studies of mutual funds were discussed to see whether their performance was consistent with the efficient market hypothesis. Here, the performance of mutual funds is discussed to test the explanatory power of Sharpe’s capital asset pricing model.
There are two excellent studies of mutual fund performance which explicitly discuss the nature of the relationship between the rate of return on portfolios and their riskiness through time. Both are in sub stantial conformity with the implications of Sharpe’s model. The first study was by Sharpe himself. He computed average annual rates of return and standard deviations of those returns for 34 mutual funds for the years 1954-63. The model implies that higher risk portfolios, on the average, will have higher returns. Sharpe’s inquiry indicates that this was true for the 34 funds during the period studied. The correlation between the average returns and their standard deviations was +0.836 indicating that about two thirds of the differences in returns were “explained” by differences in risk.
Further, the relationship between returns and risk was approximately linear, as implied by the model, except for the region of high risk. A possible explanation is that the high-risk portfolios were less efficiently diversified than the others.