臺大管理論叢 NTU Management Review VOL.30 NO.2

Predicting Future Performance Using Fair Value versus Historical Cost: Evidence from Investment Property 316 to disclose the fair values of these investment assets in the footnotes. In order to reduce potential fraud and increase credibility, CAS 3 mandates that firms use the cost model unless the firms provide evidence that the fair values of investment properties can be obtained from active markets or through values of similar properties in an active market, and the fair value estimates are reliable. Unlike IAS 40, Chinese firms however are not allowed to use model estimates when there are no appropriate market estimates available. We believe that the Chinese real estate market provides an ideal setting for testing the predictive value of fair value accounting for the following reasons. First, since CAS 3 does not require the disclosure of fair value in the footnotes for firms using the historical cost model, Chinese investors in such firms do not know the fair value of the firm’s assets. This should provide a superior test of the predictive value of the fair value model. Second, CAS 3 requires that fair value is used only by firms that can provide verifiable market values, which should assist to mitigate the lack of transparency usually exhibited by emerging markets. Indeed, China is an emerging capital market, and there is a significant lack of infrastructure and mature markets. Therefore, in order to reduce potential fraud and increase credibility, CAS 3 mandates that firms use the cost model unless the firms provide evidence that the fair values of the investment properties are from active markets or through values of similar properties in an active market, and that the fair value estimates can both be obtained and are reliable. Unlike IAS 40, Chinese firms however are not allowed to use model estimates when no appropriate market estimates are available. 3. Related Literature and Hypothesis Development 3.1 Related Literature of Fair Value Accounting in Non-financial Assets Several studies examine whether using fair value accounting increases the ability of accounting earnings to predict future firm performance (Barth and Clinch, 1998; Aboody, Barth, and Kasznik, 1999; Lopes and Walker, 2012; Bratten, Causholli, and Khan, 2012; Evans et al., 2014; Cantrell, McInnis, and Yust, 2014). However, the empirical evidence is mixed regarding whether fair value-based earnings provide better predictive ability for future earnings than historical cost accounting. Consequently, the predictive ability of fair value accounting remains undetermined. Relatively few investigations focus on whether using fair value accounting on non- financial assets provides superior information for predicting future firm performance. Using a sample of U.K. firms, Aboody et al. (1999) study whether revaluations of a firm’s