On the Practical Estimation of Ex-ante Beta for Unlisted Companies: An Empirical Test of an Approach Based on Corporate Fundamental Values

Abstract

This thesis examines the use of comparable company analysis in relation to the practical estimation of the beta component for unlisted companies. The thesis proposes a new method in selecting comparable companies based on similar fundamental values, and tests the empirical predictive value of this method in a horse-race with the conventional Pure-Play method based on industry affiliation. For each company in the dataset, the two methods estimate a beta value that can be compared to the observed beta. The theoretical framework suggests a valid basis for using fundamental values in explaining the beta value that is later empirically confirmed. The horse-race generally suggests no significant difference in the predictive value of beta between the two examined methods. For both methods, the results suggest not to adjust beta for financial leverage due to the lack of a linear relationship that the Pure-Play based adjustment assumes. Further, since the proposed method can be based on forecasted values from a discounted cash flow valuation, it manages to incorporate the fundamentals under varying conditions and amend beta to what can be theoretically expected. This ability leads to a superior ex-ante beta estimation compared to the related Pure-Play method that implicitly assumes ex-post beta to be a good approximation of ex-ante beta. This thesis therefore suggests that practitioners at least consider the proposed method as a legitimate alternative to the PurePlay method due to its complementary abilities.