Evaluating Cross-Sectional Forecasting Models for Implied Cost of Capital

The computation of implied cost of capital (ICC) is constrained by the fact that around half of all firms do not have analysts’ earnings forecasts. Hou, van Dijk and Zhang (2012, HVZ) present a cross-sectional model to generate forecasts and compute ICC from these forecasts. However, the forecasts from the HVZ model perform worse than those from a naïve random walk model and show anomalous correlations with risk factors. We present two parsimonious alternatives to the HVZ model: the EP model based on the persistence in earnings and the RI model based on the residual income model from Feltham and Ohlson (1996). We show that both models outperform the HVZ model in terms of forecast bias, accuracy and earnings response coefficients. Further, the ICC metrics generated from the EP and RI models outperform those from the HVZ model in terms of correlations with future returns and risk factors. We recommend that future research use the RI model or the EP model to generate earnings forecasts. *Corresponding author, Rotman School of Management, University of Toronto, 105 St. George Street, Toronto, Ontario M5S 3E6, Canada. We would like to thank Jim Ohlson (the editor), two anonymous referees, Jacob Thomas, Franco Wong and seminar participants at Boston University, Erasmus University, University of British Columbia and University of Toronto for helpful comments. All errors are our own.