ABSTRACT
We examine how managerial sentiment affects non-generally accepted accounting principles (non-GAAP) earnings disclosure. Following recent studies, we use terrorist attacks in the US to measure the exogenous shocks to managerial sentiment. We find that (1) firms located in the attacked metropolitan areas are less likely to report non-GAAP earnings; (2) there is a significant decline in non-GAAP exclusions for affected firms, indicating more conservative non-GAAP earnings disclosure; (3) the effect of pessimistic sentiment on non-GAAP earnings disclosure is more pronounced when terrorist-attack events are perceived as more salient, and when managers are less experienced. Our main findings are robust to a battery of robustness tests. Overall, results of this study suggest that shocks induced by exogenous negative events affect managerial sentiment and corporate disclosure practice.
Disclosure statement
No potential conflict of interest was reported by the author.
Notes
1. Brown et al. (Citation2012) examine the influence of investor sentiment on managers’ discretionary disclosure of non-GAAP earnings metrics. Our study differs from theirs in that their research finds evidence that managers strategically respond to investor sentiment in their earnings-disclosure process, while our study examines whether managers’ individual behavioral biases play a role in non-GAAP earnings disclosure.
2. The GTD data are available at https://www.start.umd.edu/gtd.
3. The mass-shooting data are available at https://library.stanford.edu/projects/mass-shootings-america.
4. Our sample of terrorist-attack events is consistent with prior literature (Antoniou, Kumar, and Maligkris Citation2016, Citation2017; Chen, Wu, and Zhang Citation2018).
5. This information is available from the FBI’s UCR Program. For more information, please refer to https://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s.
6. Available at http://www3.nd.edu/~mcdonald/10-K_Headers/10-K_Headers.html.
7. Two methods for determining whether firms disclose non-GAAP earnings are as follows: (1) reading corporate press releases (Brown et al. Citation2012); (2) I/B/E/S actual value (Heflin and Hsu Citation2008). This study uses the second method, which has been proven valid (Doyle et al. 2013Citation2013).
8. Available at https://www.huduser.gov/portal/datasets/usps_crosswalk.html.
9. The approach is consistent with the study of Brown et al. (Citation2012).
10. One to multiple matching method is applied in our study to enhance the power of the test. See Smith and Todd (Citation2005) for a guide on propensity-score matching.
11. Prior study has proved that the majority of the non-GAAP numbers in the I/B/E/S database agree with hand-collected non-GAAP disclosure and the data in I/B/E/S is generally informative (Bentley et al. Citation2018). This data set is now publically available at https://sites.google.com/view/kurthgee/data.