Chapter 7 • Portfolio theory and its relevance to real investment decisions
The invalidity of the last three of these means that, strictly, the model has not been
properly tested. This is to say that the model might work perfectly well, but we sim-
ply do not know whether it does or not.
It seems that the reason that CAPM was broadly supported by early tests of it was
partly ascribable to the nature of stock market returns during this early period, which
seemed to fit with the model. In the light of more recent research, this appears to be
little or no more than a misleading coincidence. Research based on stock market data
from more recent times has shown the model to have been hardly, if at all, able to
explain security returns.7.10 How CAPM is used to derive discount rates for
real investments
There are three factors that need to be estimated for the future so that the basic CAPM
can be used to derive the discount rate in practice. We shall now take a look at each
in turn.The risk measure (beta)
Beta has the role of measuring the level of risk for a particular investment. The appro-
priate beta to use here is the beta of a business that operates in the area of the proposed
investment for which the discount rate is currently required. In practice probably the
best approach would be to take the average beta over several businesses operating in
the area concerned. In practical terms, a reasonable approach seems to be to use some
commercial service’s estimate of the betas for several businesses whose principal
activity is similar to that in which the particular real investment under consideration
lies. Digital Look Ltd (www.digitallook.com) publishes betas.The risk-free rate (rf)
While there is strictly no such thing as a risk-free asset, short-dated UK government
bills probably are as safe an asset as we can find in the real world. The historical inter-
est rates on these are readily accessible to us. Our problem remains one of estimating
likely future rates. Fortunately, government bill rates are fairly stable from year to
year, and seem to be predicted with a fair degree of accuracy by the leading economic
forecasters. Al-Ali and Arkwright (2000) found that nearly all businesses that use
CAPM base the risk-free rate on historic returns on government securities. Dimson,
Marsh and Staunton (2002) found that the average return from government securities
over the period 1900 to 2001 was about 2 per cent p.a., in real (ignoring inflation)
terms. The current real return on UK government bills is about 2 per cent p.a.The expected return on the market portfolio (E(rm))
The expected return on the market portfolio is something of a problem area. This
factor tends to be pretty volatile from year to year and difficult to forecast accurately.
We could use an average of returns for the immediate past periods as a surrogate for
future expectations but, as was shown by Ibbotson and Sinquefield (1979) on the basis
of US data, vastly different results could be obtained depending on which starting