We observe many CEOs of big companies as easily as movie stars or sports players on Twitter. Why do they appear on social media and what would be the effect of their Twitter network on the corporate performance and shareholder benefit? This study explores big company CEOs on Twitter. We identify Twitterusing CEOs and examine the impact of their social media activity on the corporate performance. While academic literatures have studied the performance of publicly well-known CEOs mainly with media coverage, we focus more on the big company CEOs on the social media and test its effect on the corporate performance. We discovered the determinants of Twitter status of CEOs, in terms of personal attributes, company attributes and industry. We also found the positive impact of Twitter on corporate performance, contrary to previous evidences of negative effect.
The behavior or characteristic of firm executives has been studied in corporate finance and behavioral economics in terms that it can possibly affect the corporate performance, and consequently the benefit of shareholders. Among the attentions to executives, public reputation is one distinct interest to both researchers and shareholders. Especially, the main interest is focused on their individual compensation and the consequences of their distinctive status on the performance. Empirical evidences have shown that public fame has positive effect on the executive compensation but negative ex-post consequences on the corporate performance, leading negative effect for the shareholders [1].
Copyright is held by the International World Wide Web Conference Committee (IW3C2). IW3C2 reserves the right to Cporpoyvriidgehta hcyp2e0r1l6inhketlod tbhye aauutthhoorr’(ss)s/iotewniferth(se);Mcaotpeyriinagl ipseursmeidttiend oPenullebyclitfsrohorenpdircaivmsapteeadriatan.odf atchaed#emMicicrpoupropsotsse2s0.16 Workshop proceedings, aWvaWilaWble20o1n6linCeoams pCaEnUioRn,VAolp-r1i6l9111(–h1t5t,p:2/0/1c6e,urM-wosn.torréga/l,VoCla-n1a69d1a). #Microposts2016, Apr 11th, 2016, Montréal, Canada.
ACM XXX-X-XXXXX-XXXX-XXXXXXXXXXXXXXX. DOI: http://dx.doi.org/XXXXXXX
We test such argument in the context of social media. While the
previous literature used award-winning event as means of
obtaining the superstar CEO status [1] or press coverage as a
proxy of reputation [
We mainly propose two questions: What are the determinants of Twitter-using CEO? And what is the impact of Twitter on the corporate performance. These questions are described by the following hypotheses.
H1. Twitter activities of CEOs will be explained by personal attributes and firm attributes.
H2. Twitter activities of CEOs will have an impact on the corporate performance.
We describe the Twitter-using CEOs by personal attributes and firm attributes. Personal attributes include age, gender and total compensation of CEOs, while firm attributes include company size, return on assets (ROA), return on equity(ROE), and leverage. Industry dummies are also used.
Regarding the corporate performance, we apply stock market variables (price, trading volume, bid and ask, shares outstanding) and compute the market-based performance measures(return, spread, turnout) to build up the following models. 1.1 = ! + !,! ! + !,! ! + 1.2 ! = ! + !,! ! + !,! ! + + !,!!! !!! 2.1 ! = ! + !,! ! + !,! !
Underlying dataset is the S&P 500 constituents in 2014 and their CEOs on Twitter. Companies’ financial data, and executive information are obtained from CRSP, COMPUSTAT, and EXECUCOMP. After refining preferential/ordinary stocks with classes, we maintain 484 firms and corresponding CEOs, whose screen names and twitter status are verified through Twitter API. We discovered the CEOs who have Twitter accounts and finally identified 109 CEOs. Table 1(b). Sample Overview: Industry Average Twitter status is also different through industries. The industries in Consumer discretionary sector (e.g. Media, Retailing, Apparel), and IT sector (e.g. Software) turn out more active on Twitter. CEOs of IT companies tweet more, having more friends, more followers for a longer period. Average Twitter status of CEOs is shown in Table 2. On average, female CEOs have more followers. Considering less number of female CEOs, it seems to show the public interest toward female CEOs of big companies.
According to Table 3, age and compensation turn out as significant determinants of having Twitter accounts and holding period. The younger and the more-paid CEOs are likely to have Twitter accounts with significance. Industry dummy test shows that the CEOs of consumer service-oriented industries and IT industries have more Twitter accounts. However, variables related to twitter status (number of tweets, friends and followers) are explained mainly by other twitter status variables one another, implying the fundamental character of networking. Among company variables, only current total asset shows significance. Meanwhile, there exists heteroskedasticity according to BreuschPagan LM test. However, this seems due to significant correlation between the Twitter status variables as shown in Table 4.
To find the Twitter impact on the corporate performance, we use simple model with Twitter variables and Market capitalization as a company variable. Assuming that potential retail investors are more interested in the stock and performance variables are related to stocks, market capitalization can be tangible proxy for the company size that investors perceive.
Return CEO tweet has positive effect on the market return of company share with weak significance. The number of followers and holding period also shows positive effect on the return, even with less significance. Hence, being popular on social media may not always be bad to company. Even though further study will be carried for it, this positive effect opens a room for new insights. Spread defines the difference between bid and ask price of share on the trading day. Therefore, it is often regarded as a measure of market efficiency, regarding the symmetry of information between traders. In our results, negative effect of CEO tweets on the spread means that CEO tweets may reveal better signal and bring less inefficiency in the share price. However, the number of friends still raises the question with its positive impact.
Share turnout defines the period trading volume compared to the number of shares outstanding on the same period. Therefore, it implies how volatile the market would react for specific factor. The result shows that the number of followers of CEO Twitter has significant impact on the volatile trading mood. Holding period is also showing positive effect on the share turnout. Market cap has opposite effect on the share turnout.
We currently work on the model, data, and measures. In order to develop the notion of the fame of CEO on Twitter, more measures from Twitter will be adopted, on top of the basic figures of Twitter. Furthermore, the influence of CEO tweets on the corporate performance will be studied on firm-specific issues. By measuring the abnormal returns on the event window of related tweets, we expect to observe its instant impact on the market response. The remaining part of this work will be better improved through the opportunity of peer review on the workshop.
[1] Malmendier U, Tate G. 2009. Superstar CEOs. Quarterly
Journal of Economics 124(4): 1593-1638.