Measuring disclosure using 8-K filings
- PDF / 1,351,775 Bytes
- 60 Pages / 439.37 x 666.142 pts Page_size
- 10 Downloads / 206 Views
Measuring disclosure using 8-K filings Jing He 1 & Marlene A. Plumlee 2 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract We construct four voluntary and two mandatory disclosure measures for 260,880 firmquarters from 2005 through 2016 using 8-K data. The first voluntary disclosure measure is the count of 8-Ks, a proxy used in prior studies. The second and third are the count and word count of the 8-K items classified as voluntary (Items 2.02, 7.01, and 8.01) and the associated exhibits. The final voluntary disclosure measure is the count of the voluntary 8-K items and exhibits that include management guidance, conference calls, non-GAAP measures, or investor day disclosures. We document basic properties of these measures, including their cross-sectional and time-series correlations and persistence, and associations with firm-level characteristics. We show that, although the four measures are highly correlated, each captures unique aspects of firms’ disclosures. Based on how the measures are constructed and supported by our findings, we contend that the word-count-based 8-K measure provides a superior proxy for firmlevel voluntary disclosure. Keywords Voluntary disclosure . Mandatory disclosure . 8-K filings JEL codes M41 . D83
We appreciate extremely valuable feedback from Russell Lundholm (the editor), Holly Yang (discussant), the 2019 RAST conference participants, and two anonymous reviewers. We are also grateful to Philip Berger, David Burgstahler, Mike Cooper, Tricia O’Malley, seminar participants at the 2017 AAC Convention, and workshop participants at CUHK and UCLA for their helpful comments. We benefited from numerous invaluable discussions with Atif Ellahie, Rachel Hayes, Mac Gaulin, Daniele Macciocchi, Xiaoxia Peng, David Plumlee, and Jordan Schoenfeld. Data gathered by Dr. Joel Brown, Aesir Consulting LLC. All remaining errors are the responsibility of the authors.
* Jing He [email protected] Marlene A. Plumlee [email protected]
1
Department of Accounting & MIS, University of Delaware, Newark, DE, USA
2
School of Accounting, University of Utah, Salt Lake City, UT, USA
J. He, M. A. Plumlee
1 Introduction An extensive literature focuses on understanding the sources and implications of voluntary disclosures provided by firms.1 As noted by Beyer et al. (2010), Berger (2011), and others, empirical research in this area relies on a number of firm-level measures of voluntary disclosure, including (1) survey rankings/AIMR scores, (2) researcher-constructed measures, (3) the presence or number of specific firm-level disclosures or events (management earnings guidance, non-GAAP earnings, conference calls, or investor days), and (4) various measures based on 8-K filings.2 While all of these measures capture cross-sectional variation in disclosure, they are subject to limitations that might reduce their usefulness in disclosure research. Some of the measures have limited availability, either in terms of the coverage universe of firms or periods (e.g., AIMR scores and
Data Loading...
