not worth shareholders money. The diagram shows how the CAPM works. Safe in- vestments, such as Treasury bills, have a beta of zero. Riskier investments should earn a premium over the risk-free rate which increases with beta. Those whose risks roughly match the markets have a beta of one, by definition, and should earn the market return. So suppose that a firm is considering two projects, A and B. Project A has a beta of 1/2: when the market rises or falls by 10%, its returns tend to rise or fall by 5%. So its risk premium is only half that of the market. Project     identifies the standard deviation of the zero-beta portfolio. Notice in Figure 9.7 that different efficient portfolios such as P and Q have different zero-beta companions. These tangency lines are helpful constructs only. They do not signify that one can invest in portfolios with expected return-standard deviation pairs along the line. That would be possible only by mixing a risk-free asset with the tangency portfolio. In this case, however, we assume that risk-free assets are not available to investors. 3. The expected return of any asset can be expressed as an exact, linear function of the expected return on any two frontier portfolios. Consider, for example, the minimum-variance frontier portfolios P and Q. Black showed that the expected return on any asset i can be expressed as   E(ri) E(rQ) [E(rP) E(rQ)]   Cov(ri, rP) Cov(rP, rQ) P Cov(rP, rQ)   (9.8) III. Equilibrium In Capital Markets 9. The Capital Asset Pricing Model The McGraw−Hill Companies, 2001                   Bs risk premium is twice that of the market, so it must earn a higher return to justify the expenditure.   Never Knowingly Underpriced But there is one small problem with the CAPM: Financial economists have found that beta is not much use for explaining rates of return on firms