Intangible asset valuation using the Feltham-Ohlson framework and real option analysis: Theory and empirical evidence

Residual income asset pricing models identify potential theoretical differences between the market value of equity and the financial accounting construct book value of equity. Feltham and Ohlson (1995) establish a link between accounting measures of net operating activities, net financial activities and the market price of equity. This dissertation is the first research to operationalize the Feltham-Ohlson (1995) model and subject it to rigorous empirical testing. Furthermore, this study decomposes the unrecorded goodwill (the difference between the market and book value of equity) identified in Feltham-Ohlson (1995).; The analytical framework used in this thesis decomposes accounting bias into components associated with assets-in-place and future growth opportunities. Real option theory is used to value the growth opportunity components having embedded strategic flexibility. The ability of the residual income constructs and growth opportunity measures to control for accounting bias are subjected to empirical validation.; A set of nested hypothesis are developed to test for the presence and source of accounting bias. Empirical tests are constructed to measure the presence and magnitude of accounting bias. Further tests are implemented to determine the ability of alternative residual income measures and growth opportunity proxies to control for observed bias. Feasible generalized least squares is used to control for heteroskedasticity and the 2SLS Instrumental Variables technique is used to control for measurement error associated with abnormal net operating income constructs.; The results generally support the external validity of the Feltham-Ohlson model as a valuation technique. In addition, the inclusion of residual income constructs helps to control for the observed accounting as a bias. Residual income based on I/B/E/S analyst forecast information provides the most consistent control for bias. Bias control for discretionary growth opportunity activities such as R&D and advertising best controls for bias when implemented under the valuation suggested by the real option model. This result also suggests the current disclosure of R&D and advertising expenses is valuation relevant since current expenditures reflect the occurrence and magnitude of the call on the real option. These results provide evidence the model developed and implemented in this thesis has potential use in evaluating the valuation relevance of alternative financial disclosure methods.