RDP 9107: The Cost of Equity Capital in Australia: What can we Learn from International Equity Returns? 8. Conclusion

This paper has reviewed the evidence on the cost of equity in Australia. Two standard measures of the cost of equity are the realised rate of return on equity, and the earnings yield. The former measure suggests that the cost of equity in Australia may be similar to or lower than overseas, while the latter suggests that the cost of equity may be higher than overseas. There are a number of reasons, however, why historical rates of return may not provide good measures of the cost of equity. Earnings/price ratios also have drawbacks, but if certain adjustments to earnings are made, and data are averaged over the recent past, they may provide better measures of the required return on equity. In particular, earnings/price ratios provide some evidence of how the stockmarket values earnings flows and it would seem dangerous to ignore this information. However, this paper has also highlighted the need for more detailed analysis of E/P ratios, along the lines of recent work which compares US and Japanese data.

The evidence on E/P ratios suggest that the cost of equity may be higher in Australia, but offers no insights as to why this might be. Accordingly this paper has also estimated a version of the International Asset Pricing Model. I find that the Australian stockmarket appears to respond more to movements in the world market than is justified by its relatively low debt/equity ratio. This implies that the level of non-diversifiable risk may be higher in Australia than overseas. If so, the equity risk premium may be higher in Australia. In addition, I find some tentative evidence that the Australian exchange rate may have some correlation with the world equity market return, which may imply a further risk premium on both debt and equity. The analysis also suggests that further work could profitably investigate the premium (if any) on energy price risk.

To further investigate the risk that is suggested by data from financial markets, the paper has looked briefly for the source of these fluctuations. It seems that a very broad measure of earnings in the Australian economy shows excess sensitivity to world earnings. Hence the riskiness that is suggested by financial markets data may have a fundamental cause in the real economy. This evidence on real earnings risk and financial market risk may help to explain the large flow of investment abroad by Australians, especially by life offices and superannuation funds, when official controls on investment abroad were eased in the early 1980s.