Page 143 - 35-2
P. 143
NTU Management Review Vol. 35 No. 2 Oct. 2025
relevance. Expanding such works, Aboody, Barth, and Kasznik (1999) find that upward
asset revaluations are positively linked to future performance (measured as changes in
operating income and cash from operations).
As accounting standards for fair values have evolved over the past decade, a growing
body of research looks at how new standards affect the value relevance of fair value
estimates. Statement of Financial Accounting Standards (SFAS) 157 requires that fair
value assets and liabilities be classified into different levels based on the transparency
of the inputs as a means of valuing the assets/liabilities. Song, Thomas, and Yi (2010)
document that the value relevance of level 1 and level 2 fair value is greater than that
of level 3 and that corporate governance is positively associated with the relevance of
these reported fair values. Barron, Chung, and Yong (2016) find that SFAS 157 adoption
introduces more comprehensive fair value disclosures of level 3 estimates, giving way
to a decline in analysts’ overall uncertainty, as reflected by lower forecast error. Magnan,
Menini, and Parbonetti (2015) also reveal a differential effect on different categories of
financial assets, pointing to an increase in the quality of public and private information
for level 2 but a decrease in that for level 3. Similarly, Riedl and Serafeim (2011) show
that firms with greater exposure to level 3 financial assets have higher betas than level
1 and level 2 financial assets, revealing that this relationship is mitigated by a stronger
information environment.
While fair value accounting improves the value relevance of financial statements, it
also leads to income becoming increasingly volatile and, in turn, less predictable (Barth,
1994). Barth (2006) points out that, as more estimates of the future are incorporated
into current measures of assets and liabilities, the reliability of future estimates plays an
important role in determining the predictability of income. Similarly, Landsman (2007)
points out that the level of informativeness in fair value estimates is determined by the
estimates’ reliability, which is dependent on both the source of the estimates and their
measurement error. He also points out that relying on managers’ estimates of the fair value
of assets and liabilities introduces information asymmetry.
As reliability is a prominent concern for fair value accounting, prior research has
examined how reliability and managerial opportunism influence the consequences of
fair value accounting. Beaver and Venkatachalam (2003) examine bank loan fair values,
showing that investors react negatively to the discretionary part of fair values if the
135

