Is Corporate Tweeting Informative or is it Just Hype? Evidence from the SEC Social Media Regulation

Transcription

1 Paris, France, th Financial Risks International Forum Is Corporate Tweeting Informative or is it Just Hype? Evidence from the SEC Social Media Regulation Mohamed Al Guindy Smith School of Business, Queen s University Kingston, Canada

2 2

3 3

4 4

5 Does Social Media matter? Congratulations to Ted Sarandos and his amazing licensing team. Netflix monthly viewing exceeded 1 billion hours for the first time ever in June Ted, we need even more! 5

6 Trading volume (M shares) Cumulative return (%) Return around Netflix post Day (relative to post) Trading volume around Netflix post Day (relative to the post) 6

7 SEC regulation (Reg-SocMedia) In April 2013, the SEC stated that companies can use Social Media as outlets to disclose information to investors. [the April 2013] decision could prompt a sea change in how companies communicate with investors and comes as regulators more broadly grapple with adapting decades-old regulations to new and evolving technologies. (Scannell, 2013). Financial Times 7

8 Research Question(s) Is corporate tweeting informative or just hype? Before and after the SEC regulation What is informative? What is hype? Interaction between social media and more traditional methods of communicating information Do firms strategically time their news on Twitter? 8

9 Why Twitter? Twitter has become the social medium of choice for Investment Relations Professionals. (Q4 Web Systems, 2012) Twitter is by far the most popular social media service for investor relations. You can use it to post short announcements about earnings releases, webcasts, acquisitions and other corporate news. Dave Hogan, Director of Investor Relations and Corporate Communications, First Financial Bankshares Financial Analysts and portfolio managers are turning to Twitter for financial information (Kiladze, 2013) Discussions between investment firms and ETF issuers to create a Twitter ETF that tracks the mentions of a large number of firms on Twitter 9

10 Return (%) Return (%) Before SEC social media regulation Day (relative to tweet) After SEC social media regulation Day (relative to tweet) 10

11 Literature Review Disclosure literature Diamond and Verrechia (1991) Merton (1987) Media in financial markets Tetlock (2007) Tetlock, Saar-Tsechansky and Macskassy (2008) Engelberg and Parsons (2011) Textual analysis Tetlock (2007) Loughran and McDonald (2011) Technology & Social Media Da, Engelberg and Gao (2011): Google SVI Blankespoor, Miller and White (2014) Jung, Naughton, Tahoun and Wang (2016) 11

12 Classification of Tweets 1. General interest: represents the majority of tweets 2. Financial tweets: targeting investors or potential investors, usually a (much) smaller number of tweets 12

13 What predicts tweeting Logit model to predict tweeting: Which firms are likely to use Twitter? Higher CAPM Beta Lower B/M (growth firms) Larger firms Less leverage Less institutional ownership Tweeting on a given day Tweeting autocorrelation Industry clustering Earnings period Use of Twitter CAPM Beta 0.156*** (0.03) B/M ** (0.04) Size 0.262*** (0.02) Leverage *** (0.11) ROE (0.006) Payout (0.10) Institution *** (0.001) Analysts 0.037*** (0.005) Dispersion (0.02) Year fixed effects Industry fixed effects R 2 N Included Included

14 Regression analysis Abnormal returns Before SEC regulation Financial tweeting day (9.42) Financial tweeting day * after SEC After SEC regulation Effect of SEC regulation (Full sample) 19.47** (8.43) * (12.77) R N Abnormal returns (excluding earnings season) Before SEC regulation Financial tweeting day (7.79) After SEC regulation 15.85* (8.30) Effect of SEC regulation (Full sample) Financial tweeting day * after SEC *** (10.65) R N

15 Vector Autoregression Dep. variable: Returns (basis points) Tweeting Day t *** (5.66) Tweeting Day t (5.65) Tweeting Day t (5.66) Tweeting Day t (5.66) Tweeting Day t (5.66) Tweeting Day t (5.61) AICC N

16 Number of followers Above mean scaled number of followers Below mean scaled number of followers Financial tweeting day *** (21.62) (9.50) R N

17 Newswire press releases Tweet & corporate press release Corporate press release only Return on day *** (15.82) *** (3.39) R N

18 Tweeting negative information Logistic regression predicting negative tweets: Outside market hours 0.259*** (0.08) Earnings day *** (0.11) Week before earnings *** (0.13) Week after earnings *** (0.09) Friday afternoon 0.311* (0.17) Institution ** (0.001) R N

19 Tweeting negative information (II) Little negative news Most tweeting is positive (Verrechia 1983) Short-sale constrains More likely to tweet negative news Investor inattention Outside market hours Friday afternoon Not during earnings season

20 Robustness tests Intraday evidence Do previous returns predict tweeting? Changes to tweeting frequency or tone New tweeting firms after SEC regulation Firms that started tweeting financial information after the SEC regulation 20

21 Conclusions Economic After the SEC regulation informative Before the SEC regulation Hype More followers greater impact Regulation SEC s Reg-SocMedia made social media more pertinent Behavioural Tweeting negative information Investor inattention Friday afternoon effect! Methodology Contributing to the textual analysis literature Dealing with the noise problem in Social Media 21