The CAPM can be used for: classifying shares, portfolio selection, measuring portfolio performance, finding miss-priced shares and to calculate the required rate of return of an organisations investment. Shares can be classified into three groups by their beta values- these groups are: aggressive, defensive and neutral. Neutral shares (b=1. 0) follow the market exactly, aggressive shares (b>1. 0) exaggerate market movements and defensive shares (b<1. 0) constrict the movement compared to the market.
The ability to classify shares enables investment managers to create portfolios specific to investors risk tolerances through the use of beta. At the end of each financial year the investment manager can calculate how well their portfolios are performing. This is done through comparing the CAPM expected-return with the actual-return. If the return is lower than expected, adjustments are needed for the portfolio. The CAPM is used to identify the shares with abnormal risk-return characteristics in-order to decide which shares to trade.
An organisation can be seen as a group of projects, with each project affecting the value of the organisation. This means investors will require different rates of return depending on the risk of new projects. A company that undertakes a riskier project than normal will have an increased beta value. This also works to reduce the beta of a company, if they embark on a less risky project, the company as a whole will be less risky. Strengths and weaknesses Weaknesses Obtaining the historical beta of a company is a fairly straightforward exercise, although deciding how long the observation period should be isn’t so easy.
The person calculating the beta can choose any period between 30 days to 10 years; this can cause misleading data if the same time frame isn’t always applied. If this is controlled there is still a problem as the data is historical and this is unreliable for making future predictions. Black, Jenson and Scholes (1972) found a positive relationship between beta and returns from their study of 35 years worth of data, although it wasn’t as steep as the CAPM had predicted. They believe this could be because of the different methods of finding the risk-free rate of return and the errors not having a zero mean.
The errors occur because betas are calculated using historical data (ex ante) and are later compared to the actual returns (ex post). Fama and MacBeth (1973) conducted a similar study and found that the relationship between beta and returns was unstable over time although there is a very small relationship between risk and return. Fama (1992) expands on these findings and offers an alternative to the CAPM in an effort to include the other import factors; this model is known as the Fama-French three factor model.
Lakonishok and Shapiro (1986) conducted a study for the period 1962-81 and didn’t manage to find even a simple relationship between i?? and return. As investments usually involve commitment over a long time period the CAPM will fail, as it can’t compensate for the changes in security rate over the life of the investment. Roll (1977) criticises the CAPM as being un-testable as the world market portfolio is not available, so a single stock market must be used in the calculation such as the FTSE 100 index.
If the true world market is not being used then the relationship that is revealed will not be based on the true CAPM so errors arise. The CAPM was formed on a number of assumptions concerning investor’s behaviour and the operation of capital markets which don’t reflect reality. The model assumes that investors are rational and risk adverse, able to assess returns and standard deviations through freely available information and can borrow and lend at the riskfree interest rate.
The model also assumes all assets are traded and fractions of assets can be purchased without transaction costs or taxation. Strengths In economic models it is necessary to simplify in order to explain real-world behaviour; the same is true for the CAPM, as the assumptions do not reflect reality. “As Sharpe (1964) observed: ‘the proper test of a theory is not the realism of its assumptions but the acceptability of its implications'”(Head, 2007, p223). This means the assumptions are not a problem if the CAPM has some predictive power as the technical problems can be over looked.
All of the authors mentioned in the weaknesses section acknowledge that there is some predictive power with the CAPM. Roll (1998) believes from his study of US companies for 1995-97 on credit spread that the CAPM can explain up-to 40 percent of the changes in security returns. Clare, Priestley and Thomas (1998) used data from the London stock exchange for 1980-1993 to determine whether the Fama-French model is better at explaining returns than the CAPM. They found that the CAPM is a powerful tool in explaining expected-returns and that there is no role for the Fama-French model in the UK.
“Graham and Harvey (2001) show that corporations use the CAPM more than any other to estimate the cost of equity capital. Most brokerage and investment advisory firms still offer estimates of beta as part of their service package”(Gitman, 2007, p233). Graham and Harvey (2001) conducted a general survey on corporate finance and achieved a 9% response rate. The results show 73. 9% of the respondents always or almost always use the CAPM to calculate the cost of capital, with larger firms more likely to use the CAPM.