Smithers & Co report on cost of capital (2003)
Smithers & Co Report (154 pages, PDF) by Stephen Wright, Robin Mason and David Miles for a group of UK economic regulators on the measurement of the cost of capital, particularly for regulated infrastructure utilities. This focuses on the cost of equity rather than debt or gearing issues.
The report's main findings are as follows.
- “While the CAPM itself has come in for considerable criticism ... virtually any asset pricing model that is used to derive an appropriate estimate of the cost of capital for regulated industries must inevitably build on estimates of the common components that feed into the CAPM itself.”
- “Regulated industries are unlikely to be "precisely" average, with a beta of unity; but nonetheless the dominant element in their cost of capital will always be the expected market return, with a distinctly smaller role for the risk-free rate.”
- Beta should be estimated with the most frequent share price data that can be trusted to reflect a liquid market, e.g. daily data for well-traded shares. With high-frequency data there is little statistical inaccuracy in the measurement of beta and no significant Bayesian adjustment (towards 1) should be made.
- “The historic size of the equity premium is still the subject of considerable puzzlement and controversy amonst academics; but this is largely due to the historic behaviour of the risk-free rate (proxied by the short-term interest rate). In contrast, we summarise a range of evidence that the equity return has, over reasonably long samples, been fairly stable both over time, and across different markets.”
- Care “should be applied as to whether returns are being measured using arithmetic or "geometric" averaging. The former is conceptually superior, though possibly less stable. The most crucial thing is to be aware that the difference between the two measures can be significant—as much as two percentage points or more.”
- “Our central estimate of the cost of equity capital, derived from a wide range of markets, is around 5.5% (geometric average), and thus 6.5% to 7.5% (arithmetic average).” (These numbers are net of an undisclosed measure of inflation.) “We cannot, however, be at all confident that these estimates are precisely correct: 95% confidence intervals are, at a conservative estimate, of up to two percentage points either side of the point estimates.”
For further information or advice please contact Franck Latrémolière.
Filed under CAA, Ofcom, Ofgem, Ofreg, OFT, Ofwat, ORR, Price controls.
