A Note on Corporate Valuation Using Imprecise Cost of Capital

We compare the estimator for discount rates according to Elsner and Krumholz (EK) (2013) to the approach propagated in Breuer et al. (2014) (BFM). While the EK estimator is derived analytically and implies a correcting factor by which the original arithmetic mean estimator is adjusted, the BFM approach is based on a simple ad hoc modification and recommends to truncate the time horizon for cash flow projection up to about N = 30 years. The BFM approach is reasonable, as the most relevant bias problems are implied by terminal value computations. Rather surprisingly, for our main simulation analysis based on real-world capital market data for 19 countries over the period 2008-2012, the EK estimator turns out inferior to the BFM approach. However, results depend on the kind of bias measure and on the issue of whether using total historical returns or only excess returns for evaluation purposes. In any case, our findings imply that a rather straightforward rule of thumb for valuation may be of value to practitioners: ‘stick to the simple arithmetic mean estimator without terminal value computation, but consider future cash flows up to a time horizon of about 30 years'