Fair Value Accounting and Informational Efficiency: A Look at the Confirmatory Role of Financial Reports

Abstract

Prior analytical research suggests that independently verified financial reports can enhance informational efficiency by serving a confirmatory role, where they discipline managers’ unverified, but more timely, voluntary disclosures. I study the confirmatory role of financial reports by examining how fair value accounting affects two aspects of informational efficiency: the credibility of voluntary disclosures and the timeliness of price discovery. I argue that greater measurement uncertainty associated with fair values can make the accounting numbers less verifiable and potentially less reliable. This, in turn, can hinder the extent to which financial reports can serve a confirmatory role. Thus, I hypothesize that fair value accounting can reduce the credibility of voluntary disclosures and the timeliness of price discovery.

To examine these hypotheses, I exploit SFAS 133 (FASB 1998), which increases fair value accounting exposure for derivative users by mandating all derivatives to be reported at fair value. I compare the credibility of voluntary disclosures and the timeliness of price discovery of derivative users to those of derivative non-users, pre- versus post-SFAS 133, using a difference-in-differences research design. I identify derivative users using a combination of an engine-based keyword search and manual tracing to the 10-k filings on the SEC’s EDGAR database.

Using management forecasts as a key voluntary disclosure, I find results suggesting that an increase in exposure to fair value accounting impairs the credibility of good news management forecasts, but not of bad news forecasts. A potential explanation for this asymmetric result is that bad news from management is inherently more credible and, thus, less susceptible to credibility concerns. In contrast, I find results suggesting that fair value accounting does not impede timely price discovery, but rather, can enhance timely price discovery in negative return periods. I also identify potential alternative explanations for these results.

In examining the impact of fair value accounting on the timeliness of price discovery, I find that the firm-level intraperiod timeliness metric, used in prior literature, has some limitations. Specifically, large return reversals during the period can lead to values that cannot be clearly interpreted as the timeliness of price discovery. I create a proxy to capture the extent of return reversals and find that the portfolio-level intraperiod timeliness metric mitigates such issues through averaging firm-level returns. In particular, using simulation analysis, I explore portfolio sizes that will mitigate these issues and use this to inform my analysis.

These results suggest that fair value accounting can have unintended adverse consequences for informational efficiency, by weakening the credibility of managers’ voluntary disclosures. My findings are relevant to standard setters and regulators, given a continual transition towards greater fair value accounting. This thesis highlights the importance of considering the system of public financial reporting and disclosure, where the financial report is only one of many sources of information, when assessing the impact of accounting.