Fair value as a relevant metric: A theoretical investigation.

摘要

Fair value has recently gained support from government officials, business professionals, and standard setters, who claim that the metric better reflects the underlying economic condition of a firm and is a relevant form of measurement. In this dissertation, I first provide a brief historical review of the debate about historical cost and current cost, and establish the primary reason standard setters favor the use of market price as the basis for fair value measurement. I also survey a growing body of empirical archival research investigating the relevance of fair value. In view of the mixed results from these studies, I present an alternative method to investigate these claims.; Specifically, I develop a theoretical framework to identify the necessary and sufficient conditions for market price (as a fair value measurement) to be relevant. I then use an economic model to evaluate the information content of the market price of an asset, and investigate whether this signal reflects the underlying economic condition of a firm and whether it can help users to predict firm value. This approach allows me to evaluate the claims made by supporters of fair value directly, instead of relying on inference.; I show that, when some markets are in disequilibrium, the market price of an asset alone does not always allow a user to infer a firm's value. If the user has additional information, such as a firm's economic profits, a fair value metric that is based on market price could be relevant in some cases. In addition, I use a simulation to show that reported accounting profit (using either historical cost or fair value) only provides a reliable signal for economic profit under certain conditions (in the case of a demand shock). This further erodes the user's ability to predict firm value using only accounting information.; These findings suggest that regulators should remain cautious when promoting fair value.