Disclosure
Copycat Skills and Disclosure Costs: Evidence from Peer Companies’ Digital Footprints
2020 JAR, Sean Cao (Georgia State), Kai Du (Pennsylvania State), Baozhong Yang (Georgia State), Alan L. Zhang (Georgia State)
The disclosure literature has studied proprietary costs by focusing on the disclosing firms and their disclosure decisions, while offering limited insights into peer companies’ copycatting behavior. In this paper, we examine whether copycats profit from imitating peer companies, the sources of their copycatting skills, and more importantly, under what conditions copycats cause competitive harm and impose proprietary costs on disclosing companies.
We identify copycatting companies by tracking the digital footprints of investment companies that view disclosures on the SEC EDGAR website.
We find that, from the voluminous peer disclosures, copycat companies are able to identify profitable trades that outperform other disclosed trades by 5.5% annually. Such stock-screening skills are related to their sophistication and the intensity of their research.
Finally, we find that proprietary costs are not homogeneous but rather depend on the characteristics of both the copycats and the disclosed information.
Copycats inflict greater damage on the performance of disclosing companies when they possess greater skills, when disclosed trading strategies take longer to complete, and when disclosed stock holdings are characterized by high information asymmetry.
Despite the capital market benefits of greater transparency, public disclosures may reveal proprietary information that ultimately works to the advantage of competitors. Competing firms can view and imitate a disclosing firm’s strategies, which is known as “copycatting.”
We construct a treatment group of companies that view peers’ 13F disclosures on EDGAR and a control group of non-viewing companies that could potentially execute coincidental trades without seeing peers’ 13F disclosures.
Information on organizational IP addresses comes from the Whois database of the American Registry for Internet Numbers (ARIN). To study trading activities, we observe that a change in an investment company’s holding of a stock may take one of four forms: (i) initiating a new position (i.e., from 0 to a positive holding of the stock, known as a “first buy”); (ii) closing out the position (i.e., from a positive holding to 0, or a “last sell”); (iii) increasing a current position; and (iv) reducing a current position without liquidating the position.
How to Talk When a Machine is Listening: Corporate Disclosure in the Age of AI
RFS 2021, Sean Cao (Georgia State), Wei Jiang (Columbia), Baozhong Yang (Georgia State), Alan L. Zhang (Georgia State)
Growing AI readership, proxied by expected machine downloads, motivates firms to prepare filings that are friendlier to machine parsing and processing. Firms avoid words that are perceived as negative by computational algorithms, as compared to those deemed negative only by dictionaries meant for human readers.
The publication of Loughran and McDonald (2011) serves as an instrumental event attributing the difference-in-differences in the measured sentiment to machine readership. High machine-readership firms also exhibit speech emotion assessed as embodying more positivity and excitement by audio processors. This is the first study exploring the feedback effect on corporate disclosure in response to technology.
Our analysis starts with a diagnostic test that connects the expected extent of AI readership for a company’s SEC filings on EDGAR (measured by Machine Downloads), and how machine-friendly the company composes its disclosure (measured by Machine Readability).