The Chilling Effect of Regulation FD: Evidence from Twitter

Regulation Fair Disclosure (“Reg-FD”) was intended to stop the practice of selective disclosure, in which companies provided material information to select analysts and institutional investors prior to public disclosure. It achieved this goal by requiring that material disclosures be broadly disseminated to the public through non-exclusionary channels. While the underlying concept of broad non-exclusionary disclosures is simple, the legislative implementation of this regulation generated significant controversy. In particular, a number of stakeholders believed that the difficulty associated with identifying material disclosures and broad non-exclusionary methods of dissemination would discourage firms from providing informal communications that could potentially violate Reg-FD, thus leading to a deterioration in the overall level of disclosure. While a number of prior studies have documented that Reg-FD has eliminated certain selective disclosures, it remains unclear how Reg-FD affects the firm’s overall disclosure policy and information environment.

In our paper, we contribute to our understanding of how Reg-FD may have influenced firms’ overall disclosure policy by examining one specific aspect—the adoption of new disclosure technologies. More specifically, we provide insights as to whether firms are reluctant to adopt new disclosure technologies without clear guidance from the SEC endorsing their use for the purposes of complying with Reg-FD. We focus on Twitter because prior studies have established that there are positive capital market benefits to Twitter usage, suggesting that Twitter would be broadly adopted if there were limited costs to doing so. In addition, firms do not have to use Twitter to disseminate information provided through traditional channels, which allows us to isolate the voluntary adoption of Twitter as a new disclosure medium.

We conduct our analyses using a comprehensive hand-collected dataset consisting of over 1 million tweets from July 2007 to February 2014 for 1,570 unique firms, representing the tweeting history of all firms listed on the NYSE, AMEX, and NASDAQ. We identify whether tweets relate to financial matters using a textual classification based on a dictionary of financial keywords. We exploit the April 2, 2013 report released by the SEC in response to the use of social media by Netflix’s CEO Reed Hastings to generate identification. This report, which we refer to as Reg-SocMedia, endorsed the corporate use of social media for the purposes of satisfying the requirements of Reg-FD. More important, the report approved the use of social media in strong terms that were unexpected prior to its release, thus generating a substantial change in how firms viewed the suitability of social media disclosures.

Our analyses indicate that there were substantial changes in the use of Twitter and in the market response to firms’ financial tweets following Reg-SocMedia. We find a non-transitory stock price response of 25 basis points to financial tweets following Reg-SocMedia, compared with no detectable response before Reg-SocMedia. This finding is robust to a variety of empirical specifications and cross-sectional tests. We suggest that there are two related mechanisms driving our results.

First, when we examine our underlying data, we find that firms change their tweeting behavior by both increasing their use of Twitter, providing more specificity in the tweets, and by no longer exclusively relying on traditional press releases for information that is disseminated through Twitter. Overall, we interpret the changes in tweeting behavior as evidence that firms believe value relevant information can be safely disseminated through social media without violating Reg-FD. Second, we find that there is a differential increase in twitter followers for firms that use Twitter for financial communications compared to those firms who do not. This finding suggests that investors believe that Twitter can be used to disseminate value relevant information.

Overall, we conclude that firms were reluctant to use Twitter for communicating value relevant financial information and that investors did not expect to find value relevant financial information on Twitter prior to its endorsement by the SEC. This finding is consistent with the notion that Reg-FD has potentially slowed the adoption of valuable new disclosure technologies.

The complete paper is available here.