Estimating Expected Cost of Equity Capital: A Theory-Based Approach

Abstract

In this study, we estimate the expected cost of equity capital using the unrestricted form of the classic dividend discount formula and examine the extent to which these estimates (rDIV ) reliably proxy for expected cost of equity capital. We find that the rDIV estimates are associated with six risk proxies suggested by theory and prior research in a manner consistent with expectations; the explanatory power of the model is approximately 26%. Based on these results we conclude that the rDIV estimates are a valid proxy for expected cost of equity capital. Estimating rDIV requires a terminal value forecast. Since such forecasts are not always available we also assess the reliability of estimates produced by imposing three alternative terminal value assumptions on the dividend discount model. Specifically, rGORDON (Gordon and Gordon (1997)) imposes a firm-specific assumption; rGLS (Gebhardt, Lee and Swaminathan (2001)) imposes an industry-specific assumption and rOJN (Ohlson and Juettner-Nauroth (2000) and Gode and Mohanram (2001)) imposes an economy-wide assumption. We find that the rGORDON estimates have the highest correlation with rDIV and behave in a manner consistent with expectations with respect to their relationships with the risk proxies. We conclude that, when sufficient data to estimate rDIV is unavailable, rGORDON represents a reasonable substitute. Finally, our data indicate that although rDIV and rGORDON reflect the distribution of expected cost of equity capital, neither measure (nor any of the alternatives) should be relied upon to estimate the magnitude of expected cost of equity capital and/or implied risk premiums.