Are Impairments and Unrealized Losses Equally (Un)timely?

Abstract

We compare the timeliness of U.S. GAAP impairments to the timeliness of IFRS unrealized losses for investment property. Using the reporting of economic losses by U.S. and U.K. real estate firms around the time of the financial crisis, our results show that unrealized losses are substantially more timely than impairments. This result is not explained by the presence of U.S. GAAP valuation buffers, the use of undiscounted cash flows in impairment testing, the use of Big 4 auditors or the use of external appraisers, and it also holds when extending the sample to include IFRS firms from other EU countries. We argue that while both U.S. GAAP and IFRS use fair value measurements to evaluate economic losses, the recurring use of fair value measurements under the IFRS revaluation model leads to a routine implementation process, established valuation parameters, and increased investments in the valuation infrastructure. As a result, unrealized losses are likely to possess more timeliness than impairments. By showing how similar underlying transactions lead to different reporting outcomes, we contribute to the comparability literature and challenge the view that the fair value and the historical cost models provide equally timely information when the prices of non-financial assets decline.