| |
Abstract:
Recent research concerned with enhancing conservatism corrections of linear information models (LIMs) reports a decrease in bias as compared to the Ohlson (1995) model. However, inaccuracy is not significantly reduced. These findings raise two questions: First, are LIMs able to capture unconditional conservatism? Second, if conservatism can be captured, then why is accuracy not markedly improved? With regard to the first question, we find that conservatism is captured when the Feltham/Ohlson (1995) model is estimated according to the modification suggested by Choi/O'Hanlon/Pope (2006). On average, one dollar of unrecorded reserves, measured as the estimated reserve by Penman/Zhang (2002), results in a correction of market value forecasts of approximately one dollar. Furthermore, our results suggest no improvement of the conservatism corrections for the following cases: (1) disaggregating book value into operating and financial assets and (2) estimating the Feltham/Ohlson (1995) model via the valuation function. Regarding the second question, we argue that the failure of the models to markedly reduce inaccuracy is the consequence of forcing the models to value different firms on the basis of the same conservatism coefficient. We therefore suggest an estimation procedure, in which LIM parameters are estimated separately for different conservatism levels. Our implementation reduces median inaccuracy from 36.8% to 21.1%, which is comparable to implementations of the residual income model based on analyst forecasts.
Accounting Conservatism, Residual Income Valuation, Feltham-Ohlson Model, Linear Information Model, Equity Valuation
|
| |
Abstract:
This study addresses the problem of differences between firms and the impact on valuations based on multiples. We investigate the extent to which industry-based multiples ignore additional firm-specific information and develop measures for identifying peer groups that are not comparable with the target firm. Additionally, we compare the performance of different methods that control for differences between firms. We find that differences between firms lead to systematic errors in the value estimates of different multiples. These errors are consistent with our hypotheses, statistically significant, economically substantial, consistent between different value drivers and robust over time. We find that these errors can be predicted very accurately by comparing the financial ratios of the target firm with the financial ratios of its peer group. We show that when adequately controlling for differences between firms, valuation accuracy is improved substantially and all considered value drivers perform almost equally well.
equity valuation, multiple valuation method, price-earnings, comparables, ratio analyses, financial ratios
|
