| In this dissertation, I examine whether market prices are affected by the reporting regime of information items, i.e., being recognized in financial statements or only disclosed in footnotes. This investigation is motivated by the ongoing debate over recognition versus disclosure, and the widely held beliefs regarding implications of the Market Efficiency Hypothesis (semi-strong form).; To avoid limitations in prior studies, this study exploits the unique setting stemming from by firms' recognition of Employee Stock Option (ESO) expense with a time-series research design. I find the information content of recognized and disclosed ESO expense is significantly different. Specifically, it appears that investors value ESO expense reported under the recognition regime more as an expense compared with under the disclosure regime.; To ensure that the main result is not driven by contracting theory-related factors, the second set of tests examines market reactions around adoption announcements. The evidence does not support the conjecture that pricing effects of recognition versus disclosure are driven by contracting factors. Overall, the results suggest that the information aggregation process is not completely independent of whether information is recognized or disclosed, which is consistent with the following explanations: (1) informational efficiency in the presence of information costs (Grossman and Stiglitz, 1980; Barth et al., 2003); and (2) limited arbitrage in the presence of noise trading and arbitrage costs (De Long et al., 1990; Shleifer and Vishny, 1990, 1997). |
