RDP 9205: Measuring the Cost of Capital in Australia Appendix 1: Australian Studies of the Cost of Capital
June 1992
- Download the Paper 453KB
Australian Manufacturing Council (1990)
The AMC sought to examine the competitiveness of Australian companies, particularly in the traded sector. On the basis of interviews with Australian companies, they estimate the real cost of debt as around 6 per cent and the real cost of equity as between 11 and 14 per cent. With a debt:equity ratio of 41:59, this leads them to conclude the real weighted average cost of funds is around 10 per cent.
Brunker (1984)
Brunker sought to specify and measure the user cost of capital for the Australian manufacturing sector in order to include it as an explanatory variable in econometric analysis. His series differs from most others in taking explicit account of each of the various tax components and physical depreciation required to calculate the cost of capital. He first derives a cost of funds and then adjusts this to take specific account of the corporate tax rate, investment allowances, depreciation allowances and valuation adjustments. His final estimate is an aggregate real user cost of capital (including plant and equipment, buildings and structures and inventories and other working capital) for the manufacturing sector. The average cost of capital for the period was 18 per cent.
Carmichael and Stebbing (1981) and Dews (1988,1989)
Although the methodology used is the same in these studies, the objectives were somewhat different. The first paper examined the macroeconomic effects of the interaction between inflation and taxation while Dews' papers emphasised the importance of determining an appropriate measure of the cost of capital and its effect on business investment.
Carmichael and Stebbing (1981) estimate an after-tax, real, weighted-average cost of funds which Dews (1988, 1989) updates. Their approach assumes the firm maximises profits by investing until the after-tax marginal product of capital equals the real after-tax cost of funds. The value of equity and the net cash flow of the firm are defined according to the theory of the valuation of the firm. The expected real return to equity is estimated based on these definitions and the average earnings yield on equity. The debt to equity ratio is estimated from a sample of companies, and the cost of debt proxied by a real after-tax industrial debenture rate. Grossing up Dews' estimates by the complement of the corporate tax rate, then the average cost of capital is around 4 per cent per annum for the period 1962 to 1986 and 8 per cent for the period 1975 to 1986.
Johnston, Parkinson and McCray (1984)
Johnston et al aimed to introduce a rate of return objective for public business authorities as part of a strategy to place their operations on a more commercial footing. They employ two different approaches to estimate a target rate of return for public enterprises based on rates of return achieved in the corporate sector. They use an opportunity cost approach, based on National Accounts data, and a cost of funds approach, based on a grossed up, weighted average of debt and equity costs. Both measures they develop are ex post (realised) returns and they suggest that when applying their measures as a hurdle rate “some allowance can be made for expected future developments”.[27]
To construct the cost of funds measure, Johnston et. al. compare the value of a geometric mean of share market accumulations and the geometric average yield on 10-year government bonds (to proxy for the risk-free rate). They find that a typical private-sector enterprise has a long-run ex post cost of capital of 11 per cent. The cost of capital estimate assumes a debt to equity ratio of 40/60, uses a stock-exchange estimate of the nominal “effective” tax-rate of 37 per cent, and assumes an equity risk premium of just under 3 per cent. Based on overseas and domestic experience they derive a real long-run cost of private-sector debt of 4 per cent (based on a rate of 3 per cent for public debt plus a 1 per cent margin for risk).
Department of Finance (1987) and Swan (1988)
DoF aims to estimate a discount rate that is relevant to a framework for project appraisal in government departments and agencies as well as business authorities. They use a methodology similar to the opportunity cost approach of Johnston et.al., but provide estimates that reflect the realised real rate of return to total assets, rather than the return to fixed assets only.
Measures based on national accounts estimates are sensitive to assumptions made about the value of assets such as land – which are not included in the national accounts, but must be included in the denominator of a cost of capital estimate. Johnston, et.al. show that an estimate of the cost of capital may vary by nearly 7 percentage points in some years depending on which estimate of the capital stock is used. Nevertheless, DoF's estimates of the actual rate of return are reasonably close to some independent estimates, based on stock-market data made by Swan (cited in DoF (1987)). Swan estimates that the average ex post cost of capital over the period 1967/68 to 1982/83 was about 6 per cent.
Unlike Johnston, et.al., DoF allow for expected future developments. They extend their national accounts based analysis of actual returns to debt and equity holders for the period 1976–77 to 1985–86, to derive a measure of the ex ante cost of capital. This measure averages 12 per cent over the period. This estimate includes a real required rate of return on new equities of 18 per cent (pre-company tax). With the real rate of interest on two-year bonds averaging 3 per cent over this period the implied risk premium on equity is around 15 per cent, above most other estimates. Swan re-estimates the DoF estimates, using “better data and methodology”[28] and finds that the ex ante cost of capital is significantly lower at 7.5 per cent.