Open Access

Paul Wissmann

• This study examines the accuracy of various models for estimating the cost of equity capital, focusing on both market-based and accounting-based approaches. The research evaluates whether implied cost of capital (ICC) models, which employ forward-looking estimates, offer better approximations for explaining stock prices compared to traditional factor models, including the CAPM, Fama-French three-factor, Carhart four-factor, and Fama-French five factor models. An empirical analysis was conducted using data from four major U.S. companies: Lockheed Martin, Johnson & Johnson, XOM and IBM. The expected returns for these companies were calculated using various methodologies and subsequently employed as the cost of equity in the Weighted Average Cost of Capital (WACC) within Discounted Cash Flow models to determine the implied stock price. The primary objective was to identify which method delivers the most accurate forecast results by comparing the percentage deviations of the implied share price from a benchmark. Furthermore, the study assessed the sensitivity of these models to different factors. The findings indicate that, for three out of the four companies analyzed, models based on historical data provided the most accurate forecasts compared to the benchmark. In one instance, results derived from ex-ante expected data using the residual income model were more accurate. While the CAPM and other factor models generally performed well, certain variants of the ICC models, particularly within the residual income model framework, produced results that were closer to the benchmark in specific cases.