Valuation Accuracy of the RIV Model - Is it possible to improve the valuation accuracy of the RIV model when taking both conservative measurement bias and business goodwill/badwill into account?

Abstract: The residual income valuation model (RIV) presented by Ohlson (1995) relies on unbiased accounting. The presence of conservative accounting, however, has been well recognized within research (e.g. Mora and Walker, 2015; Barker, 2015). This thesis aims to investigate if it is possible to improve the valuation accuracy of the RIV model when taking both firm specific conservative measurement bias and business goodwill/badwill into account. We examine two versions of the RIV model, operationalized as Method I and Method II. In Method I we aim to capture the market's implicit estimate of goodwill at the horizon point in time and incorporate it into the terminal value calculation. In Method II we estimate a conservative measurement bias parameter, which enables us to use unbiased accounting in the RIV model. The relative valuation accuracy of the two models is examined by comparing them with the valuation accuracy of a base model. The thesis is based on a sample of companies listed on the stock exchange in Sweden, Finland, Norway and Denmark. We find that Method I exhibits better valuation accuracy compared to the base model, whereas Method II does not exhibit better valuation accuracy. In addition, we elaborate upon the methodological choices made for Method I and Method II. The results indicate that further improvements in valuation accuracy could be obtained.