Social media as a voluntary disclosure
dissemination channel, welcome addition or
uncontrollable mess?
Bachelor Thesis Accountancy & Control
Milan C.W. van den Abeele 10634029
Supervisor Ms. Ejona Gjata MSc University of Amsterdam
Faculty of Economics and Business
Final version
2 Statement of Originality
This document is written by Student Milan van den Abeele who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than mentioned in the text and its reference have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
This study examines whether voluntary disclosure dissemination through social media reduces information asymmetry and if so, what the benefits and drawbacks are compared to other dissemination channels. Information asymmetry can be reduced by increased disclosure of firms’ private information. However, due to multiple reasons, firms do not disclose all their private information and therefore information asymmetry exists. Historically, firms had to rely on third parties like analysts and the press to disseminate their disclosures. But since the rise of social media, firms now have the possibility to disseminate their own disclosures. By conducting a literature review it can be concluded that voluntary disclosure dissemination through social media reduces information asymmetry. Social media is a helpful channel to widely disseminate disclosure in a short amount of time. Especially for less-followed firms this can be a helpful channel to reduce information asymmetry. The interactivity of social media can also be a disadvantage, when the dialog on the disclosures has a negative tone which the firm cannot control. This makes social media less favourable for well-followed firms. This paper contributes to the voluntary disclosure literature by combining prior limited empirical studies to give further insight in the capabilities of social media as a disclosure dissemination channel.
3
Samenvatting
In deze paper wordt onderzocht of het verspreiden van vrijwillig naar buiten gebrachte informatie op social media informatie asymmetrie verlaagd en als dat zo is, wat de voor en nadelen van social media zijn ten opzichte van andere beschikbare kanalen. Informatie asymmetrie kan door een bedrijf worden gereduceerd door extra informatie naar buiten te brengen die de investeerders niet kennen. Door verschillende redenen brengen bedrijven niet alle informatie waarover ze beschikken naar buiten, waardoor er informatie asymmetrie bestaat. Voor de opkomst van social media moesten bedrijven op derde partijen zoals analisten en de pers vertrouwen voor het verspreiden van de informatie. Met social media kunnen bedrijven makkelijker zelf de informatie verstrekken. Na het uitvoeren van een literatuuronderzoek kan worden geconcludeerd dat het verspreiden van vrijwillig naar buiten gebrachte informatie met behulp van social media informatie asymmetrie verlaagd. Vooral voor kleine bedrijven kan het een bruikbare manier zijn om informatie asymmetrie te verlagen. De interactiviteit van social media kan ook een nadeel zijn, namelijk wanneer de dialoog over de gepubliceerde informatie een negatieve toon aanneemt welke
oncontroleerbaar is door het bedrijf. Dit maakt het verspreiden van informatie op social media minder aantrekkelijk voor grote bedrijven. Deze paper draagt bij aan het
wetenschappelijk onderzoek door het combineren van de literatuur over het effect van extra informatieverschaffing op informatie asymmetrie en de literatuur over het gebruik van social media door bedrijven en zo meer inzicht te geven in de capaciteiten van social media als een medium om bedrijfsinformatie te verspreiden.
4 Table of contents Abstract ... 2 Samenvatting ... 3 1 Introduction ... 5 2 Information asymmetry ... 6
3 Theories on voluntary disclosure ... 7
3.1 The Disclosure Principle ... 8
3.2 The Signalling Theory ... 9
4 The effect of voluntary disclosure on information asymmetry ... 12
5 Voluntary disclosure dissemination ... 15
5.1 Press release ... 15
5.2 Press ... 16
5.3 Conference call ... 17
5.4 Social media (Facebook and Twitter) ... 18
5.4.1 Characteristics ... 18
5.4.2 Effect on firms ... 19
5.4.3 Interactivity ... 21
5.4.4 Conclusion ... 23
6 Discussion and conclusion ... 23
References ... 26
5
1
Introduction
In July 2012, Netflix CEO Reed Hastings posted on his personal Facebook that the monthly online viewing of Netflix had exceeded one billion hours for the first time, which resulted in an increase of the stock price of Netflix from $70.45 at the time of the post to $81.71 at the close of the following day. However, according to the Regulation Fair Disclosure (Regulation FD) it was still prohibited to announce key information via social media. The Division of Enforcement of the U.S. Securities and Exchange Commission (SEC) launched an inquiry on the post, which was the reason for the SEC to issue a report stating that companies can use Facebook and Twitter for corporate disclosure, as long as investors have been alerted about which social media will be used to disseminate such information.1 Since then, social media has become an increasingly important medium for firms to voluntary disclose and disseminate financial and non-financial information (Zhou, Lei, Wang, Fan, & Wang, 2015). Facebook and Twitter are the most used social networks by firms. Of all S&P 1500 firms, 47% adopted Twitter, 44% adopted Facebook and 52% adopted either of the two (Jung, Naughton, Tahoun, & Wang, 2016). When comparing this to the still increasing amount of internet users that are active on Facebook or Twitter, the adoption by firms of Facebook is relatively low while on the other hand, the adoption of Twitter is relatively high. In 2016, 79% of the internet users in the U.S. were active on Facebook and 24% on Twitter.2 This raises the question why firms prefer Twitter over Facebook, while internet-users prefer Facebook over Twitter. One of the largest contributing factors to the growth of social media users is the development of smartphones and the increased mobility that comes with it. This enables users to read, share and comment on information about firms quicker and more often and it enables firms to reach a large group of investors and other stakeholders (Brown, Stice, & White, 2015).
Many studies have addressed the effect of voluntary disclosure on information asymmetry and since the issue of the report of the SEC there have been some studies on the use of social media to disseminate voluntary disclosures. However, these studies are limited in sample size and focus on a single industry. By combining these studies, this paper tries to find a more conclusive answer to the question whether dissemination of voluntary disclosure
1https://www.sec.gov/news/press-release/2013-2013-51htm
6 through social media reduces information asymmetry and if so, what the benefits and drawbacks of social media are compared to other dissemination channels. This is an interesting question, because by reducing information asymmetry a firm can avoid the possible consequences of the agency problem between managers and investors and the adverse selection problem between informed and uninformed investors. This literature study will contribute to the existing literature by combining these studies to find a more conclusive answer to the research question. For multiple reasons, this study will focus on the social media platforms Facebook and Twitter: (1) Facebook and Twitter are the most adopted social media platforms by firms and internet-users, (2) both platforms offer the possibility of direct engagement between the firm, analysts, and investors, and (3) the report of the SEC only mentions Facebook and Twitter.
Based on prior empirical research this literature review concludes that voluntary disclosure dissemination through social media reduces information asymmetry for both highly-followed as less-highly-followed firms. However, the negative relation between increased dissemination of voluntary disclosures and information asymmetry is significantly bigger for less-followed firms. This is explained by the fact that highly-followed firms already reach most interested investors. Because of the greater interactivity of Facebook, the positive impact of dissemination of disclosures on this platform is smaller than on Twitter. This greater interactivity makes it hard to control and direct the dialog on the disclosures. Firms therefore often choose Twitter over Facebook to disseminate voluntary disclosures.
The next chapter provides background information on the disadvantageous effect of information asymmetry on the functioning of the capital market. In chapter three, the theories that underlie the mechanics of voluntary disclosure are discussed. Subsequently, in chapter four, prior empirical research on the effect of voluntary disclosure is compared. A comparison of the different channels to disseminate corporate disclosures is made in chapter five. Finally, in chapter six the findings are discussed and the research question is answered.
2
Information asymmetry
One of the main challenges in the capital market economy is the so called “Lemons” problem (Akerlof, 1970). As an example, Akerlof outlines a situation where there are just four kinds of cars: (1) new cars, (2) used cars, (3) good cars, and (4) bad cars (“lemons”). An individual in the market for a new car cannot distinguish between a good car and a bad car.
7 However, during the ownership of the car, the owner gains more and more knowledge on the quality of the car. When the owner wants to sell his car, he has more knowledge on the quality of the car than the buyer. Information asymmetry exists between the buyer and the seller. Because it is impossible for the buyer to know the quality of the car, good cars and bad cars must sell for the same price. This results in owners of good cars not receiving the true value of the car and lemons driving good cars out of the market.
In the capital market, the same problem exists between investors and managers. First, managers, as insiders, typically have more information about the value of the firm and its performance than investors, so they can overstate the value of the firm (information asymmetry). Second, managers have incentives to use this private information for their own interests instead of the investors’ interest (agency problem). Because of the lack of information and uncertainty, investors will value “good” and “bad” investments equally, which causes “good” investments to be undervalued and “bad” investments (lemons) to be overvalued. According to Akerlof (1970), this may eventually lead to a disruption of the capital market.
One of the solutions of the lemons problem is full corporate disclosure (Healy & Palepu, 2001). In the capital market, firms are required by law to disclose certain information, which is called mandatory disclosure. The private information firms do not have the obligation to disclose is called voluntary disclosure. There are extensive regulations on mandatory disclosures all around the world. For example, in the United States all domestic, listed firms have to follow the United States Generally Accepted Accounting Principles (US GAAP) set by the SEC. However, these Standards do not fully solve the lemons problem, because they do not require full disclosure. In the ideal situation, the lemons problem would be solved by full disclosure from firms. However, the lemons problem still exists in the capital market. What is the reason for firms to not disclose all their private information?
3
Theories on voluntary disclosure
There are two theories/principles underlying the mechanics of voluntary disclosure. The Disclosure Principle predicts that firms will voluntarily disclose all private information, whereas the Signalling Theory predicts that firms will voluntarily disclose more private information as long as the extra benefits exceed the extra costs.
8 3.1 The Disclosure Principle
According to the persuasion game (Milgrom, 1981), a salesman selectively provides information which is relevant to the decision to influence the decision maker so there is an adverse selection problem. In equilibrium, the salesman discloses all information. With the Disclosure Principle, Dye (1985) generalizes the persuasion game to the agency problem between managers and investors/shareholders.
The fact that in equilibrium firms should disclose all their private information is based on game theory: investors have certain expectations about future profits of the firm, which determine the market price of the firm. When no firms disclose private information, the expectations of investors would be identical for each firm and therefore the market price of each firm would be identical. When a single firm discloses positive, private information, the price of that firm will rise above the equilibrium and the price of all other firms will decrease. This will force non-disclosing firms to disclose their private information to raise their market price above the non-disclosure price which, again, will cause the price of the non-disclosing firms to decrease. This will lead to an equilibrium where every firm discloses all private information. However, there are four conditions that must be met for the Disclosure Principle to hold: (1) disclosure is costless, (2) it is known to the buyer/investor what private information the firm has, (3) it is impossible for the firm to lie, and (4) the disclosed information is interpreted homogeneously by all buyers/investors (Grossman & Hart, 1980; Grossman, 1981; Milgrom, 1981). Prior research on the Disclosure Principle finds that the assumptions of the Disclosure Principle do not hold.
Disclosure costs
The first condition for the Disclosure Principle to hold is that disclosure is costless. However, in the capital market disclosure is never costless (Verrecchia, 1983). Firms have to incur costs to prepare and disseminate the disclosures. Therefore, the condition that disclosure is costless does not hold in the capital market.
Investor does not know what private information the firm has
Dye (1985) finds that investors do not always know what private information the firm has. Therefore the condition that investors always know what private information a firm has
9 does not always hold. In the case that investors do not know what private information the firm has, the firm can successfully suppress bad information because the investor cannot distinguish firms with bad information from firms with no information.
Untruthful disclosure
Not all disclosures have to be audited, which creates the possibility for the firm to disseminate untruthful disclosures. Bertomeu and Marinovic (2016) call these soft disclosures, such as press releases, forecasts, and unaudited statements. The condition that it is impossible for the firm to lie does therefore not hold in the capital market.
Information not interpreted homogeneously
Dutta and Trueman (2002) conclude that managers face uncertainty about the interpretation of disclosed information. Therefore the condition that the disclosed information is interpreted homogeneously does not hold. When managers face such uncertainty, they change their disclosure strategy in two significant ways. First, they base their expectations on how disclosures will be interpreted on prior interpretations of investors. Second, firms’ private information cannot be characterized in a single good news/bad news partition anymore. For example, the firm can think of a high inventory level as good news, but analysts can think of it as bad news. As a result, managers will disclose information that is sufficiently low or sufficiently high when analysts previously released a positive assessment of the firm. In contrast, information that is sufficiently low or sufficiently high will be withheld when analysts previously released a negative assessment of the firm.
In a theoretical setting, the conditions of the Disclosure Principle hold and managers disclose all private information. However, as multiple studies show, in the real capital market none of the principles hold. This results in firms not disclosing all private information which causes the existence of information asymmetry between managers and investors.
3.2 The Signalling Theory
Spence (1973) introduced the signalling theory by using the labour market as an example. When hiring a new employee, the employer does not know the productive capabilities
10 of the employee, it takes time to learn those capabilities. In other words, the hiring decision is made under uncertainty. The employee then has to determine the wage, which is equal to the employee’s marginal contribution to the firm. The employer cannot directly observe the marginal contribution of the employee to the firm, but he can observe the personal characteristics and attributes of the individual. This includes, among other things, sex, education and previous jobs. Some of these attributes, such as education, can be modified, whereas sex cannot be modified. Spence refers to unmodifiable attributes as indices and to modifiable attributes as signals.
When determining the wage to be offered to the applicant, the employer will base the expected marginal product of the applicant on the signals and indices of that applicant. Whereas indices are unmodifiable, signals can be manipulated by the applicant. This means that job applicants try to reduce information asymmetry between them and the employers. There may be costs involved in making these adjustments. Spence refers to these costs as signalling costs. Job applicants will therefore select signals that maximize the difference between the offered wage and the signalling costs. It is likely that more suitable job applicants will send these signals, as it is less costly for them to send these signals than for less suitable applicants and they want to distinguish themselves from less suitable applicants.
When applying this theory to corporate communication, high performing firms similarly disclose extra information to reduce information asymmetry between the firm and its investors. However, previous research shows conflicting outcomes. Some studies find a positive relationship between a firm’s profitability and its level of voluntary disclosure (Haniffa & Cooke, 2002; Lim, Matolcsy, & Chow, 2007; Wang, Sewon, & Claiborne, 2008), while other studies find no relationship (McNally, Eng, & Hasseldine, 1982; Meek, Roberts, & Gray, 1995; Raffournier, 1995; Ho & Wong, 2001; Chau & Gray, 2002; Eng & Mak, 2003; Cahan, Rahman, & Perera, 2005; Patelli & Prencipe, 2007; Chau & Gray, 2010). According to Dainelli, Bini, and Giunta (2013) “there are two main causes for these inconclusive results:
First, because these studies aim to investigate how corporate characteristics influence the amount of disclosure included in annual reports, they develop disclosure indexes that encompass different topics (see Meek et al., 1995). Assuming that different disclosure topics are managed using different communication strategies, these indexes might have obstructed the clear identification of signalling strategies.
11 Moreover, the presence of disclosure items that are not considered to provide incremental value information (i.e. public information) probably contributes to obfuscate any signalling mechanism as well (Chau & Gray, 2002; Lim et al., 2007; Meek et al., 1995)” (p. 267).
To overcome these limitations, Dainelli et al. (2013) only focus on incremental value information that should be used to communicate the signals to investors. The signalled information is attractive to investors if it has the following characteristics: (1) it has to be relevant for the decision process of the investors, and (2) it has to be private. If the information possesses these characteristics, it represents incremental value information (Beyer, Cohen, Lys, & Walther, 2010). They conclude that the most profitable firms on average disclose more voluntary information than less profitable firms and therefore the signalling theory is valid in the corporate communication environment.
Guay, Samuels, and Taylor (2016) focus on a different angle. Based on a growing amount of literature on the negative effect of complex financial statements on the information environment, they examine if firms indeed use voluntary disclosures to mitigate this negative effect. They study the period until twelve months after the 10-K filing and find that a growing complexity is associated with increased voluntary disclosures. Consistent with the signalling theory, they find that firms with relatively poor performance and high earnings management undertake less action to mitigate the negative effect of complex financial statements and vice versa.
To summarize: according to the Disclosure Principle, firms will voluntarily disclose all private information. However, according to different studies on the subject, four conditions must be met for the Disclosure Principle to hold. In the capital market, these conditions do not hold which causes information asymmetry to exist. The Signalling Theory predicts that firms will disclose more information as long as the extra benefits exceed the extra costs. As a result, more profitable firms will disclose more information than less profitable firms and information asymmetry exists when the costs of disclosure are greater than the extra benefits. This theory is substantiated by empirical research. In the next chapter, the literature studying the effect of voluntary disclosure on information asymmetry is discussed.
12
4
The effect of voluntary disclosure on information asymmetry
According to the Lemons Problem discussed above there is an information problem in the capital market which can be solved by full disclosure (Healy & Palepu, 2001). In this chapter I will discuss prior, empirical research measuring the effect of disclosure on information asymmetry.
Multiple proxies are used to measure information asymmetry. Lev (1988) states that the bid-ask spread is a good proxy to explicitly measure information asymmetry. The reason for this is that the adverse selection problem, arising from the trading of investors with asymmetric information, expresses itself directly in the bid-ask spread. According to Leuz and Verrecchia (2000), the turnover ratio is another good proxy to measure information asymmetry. The willingness of investors to sell and buy is found to be inversely related to information asymmetry (Easley, Kiefer, O’Hara, & Paperman, 1996; Grammig, Schiereck, & Theissen, 2000). Lang and Lundholm (1993) introduced the share price volatility as a proxy for information asymmetry. The thought behind this is that big changes in share price suggest information asymmetry. Some other used proxies are cost of equity capital (Botosan, 1997), trading volume (Welker, 1995; Cheng, Courtanay, & Krishnamurti, 2006), and market depth (Shroff, Sun, White, & Zhang, 2013).
The most common used proxy for disclosure level is the AIMR rating. The AIMR rating is provided in the annual Report of the Association of Investment Management and Research Corporate Information Committee. It is based on the evaluations by analysts on the disclosure practices of firms. The downside of the AIMR rating is that it only takes in consideration the most heavily followed firms in an industry. This makes it hard to compare firms from the AIMR report in a cross-sectional analysis, since studies like Lang and Lundholm (1993) find a positive correlation between firm size and disclosure levels. It also provides little variation in analyst following and firm size. To overcome the limitations of the AIMR rating, some studies (Botosan, 1997; Leuz & Verrecchia, 2000; Petersen & Plenborg, 2006; Cheng et al., 2006; Shroff et al., 2013) construct their own measure for the disclosure level.
Welker (1995) studies the cross-sectional relation between firms’ bid-ask spreads and their AIMR ratings. He tries to extend prior research by investigating the relation between the firms’ spread and its general disclosure practices during periods in which no disclosures are made. Welker (1995) finds that there is a negative relation between bid-ask spreads and the level of disclosure. This negative relation still holds after controlling for return volatility,
13 trading volume, and share price. The cross-sectional tests show that firms with a high level of trade of informed traders have a more negative coefficient relating disclosure level to bid-ask spreads.
Instead of using the AIMR rating to measure disclosure level, Botosan (1997) uses a self-constructed measure and selects a sample from a single industry, with greater variation in analyst following and firm size. This will potentially create a greater cross-sectional variation between the firms in the sample, but the sample size is limited to, in this case, 122 firms. Her proxy for disclosure level, which she calls DSCORE, consists of five categories of voluntary information in the annual report that are useful for the decision making of investors and analysts: (1) background information , (2) summary of historical results, (3) key non-financial statistics, (4) projected information, and (5) management discussion and analysis. Information asymmetry is measured through cost of equity capital. To measure the cost of equity capital, Botosan (1997) uses the earning-to-price ratio adjusted for growth and dividend pay-out. She concludes that the cost of equity capital, and thus information asymmetry, does not decrease statistically significant in relation to voluntary disclosure for firms with high analyst following. However, for firms with low analyst following she does find a negative relation between disclosure level and information asymmetry.
Healy, Hutton, and Palepu (1999) try to contribute to the disclosure literature by studying the effect of “sustained improvements” of disclosures on stock liquidity. From the AIMR reports between 1978 to 1991 they selected 97 firms which had the largest increase in average relative ranking. For this sample, they collected quartiles of the annual relative ratings from the two years before the disclosure change, the year of the change, and the two years after the change. They find that a higher disclosure level is associated with reduced relative bid-ask spreads and thus information asymmetry.
Leuz and Verrecchia (2000) state that results from prior research on disclosures and information are mixed, depending on the used disclosure metric and research design. As an explanation for these mixed results, they state that the disclosure environment in the U.S. is already very rich which makes it hard for researchers to measure detectable effects of more disclosures. To avoid this, they study the German market where disclosure levels under German GAAP are relatively low compared to the American market. They measure the effect on information asymmetry for German firms switching from German GAAP to either IAS or U.S. GAAP. The sample consists of 102 firms listed on the DAX 100. All these firms made a switch
14 from German GAAP to either IAS or U.S. GAAP. They conclude that an international reporting strategy, i.e. a higher level of disclosure, is associated with higher share turnover and lower bid-ask spreads and thus reduced information asymmetry.
The study of Petersen and Plenborg (2006) is motivated by the fact that most of the research in this area is based on American data and by the study of Verrecchia (2001), in which he argues that the existing theory on this topic is convincing, but that prior empirical research has proved the link between disclosure and information asymmetry to be elusive. They construct their own disclosure index because there is no AIMR index in Denmark. Their disclosure index is based on five categories: (1) strategy, (2) competition and outlook, (3) production, (4) marketing strategy, and (5) human capital. They identified 62 indicators within these categories. None of these indicators are required to be disclosed in Denmark, so they are all voluntary. For each indicator a point is assigned to the firm. The proxies used to measure information asymmetry are the bid-ask spread and the turnover ratio. Their sample consists of 36 industrial firms listed on the Copenhagen Stock Exchange. For each firm, they collected the annual report and determined the voluntary disclosure index based on the 62 indicators. They find a negative relation between disclosure level and both bid-ask spread and a positive relation between disclosure level and turnover ratio.
Cheng et al. (2006) also study the relation between the level of voluntary disclosure and the information asymmetry. Their sample consists of 104 firms listed on the Singaporean stock exchange (SGX) and thus is similar in size to the samples of Botosan (1997) and Leuz and Verrecchia (2000). They construct a voluntary disclosure index based on a checklist developed by Luo, Courtanay, and Hossain (2006), which consists of three categories: (1) business data, (2) management’s discussion and analysis, and (3) forward-looking information. The proxies used to measure information asymmetry are bid-ask spread, trading volume, and price volatility. The study concludes that higher levels of voluntary disclosure reduce bid-ask spreads, trading volume and price volatility.
In a more recent study, Shroff et al. (2013) examine the effect of the Securities Offering Reform on the voluntary disclosure behaviour of firms and the economic consequences of the change in behaviour. As a measure for disclosure level they use press release frequency. To measure changes in information asymmetry, they use the bid-ask spread, market depth, and analyst forecast accuracy as proxies. Their results show that the Reform caused firms to increase their voluntary disclosure level which, in turn, decreased information asymmetry.
15 According to economic theory, the information problem in the capital market can be solved by full disclosure. Multiple empirical studies on the subject find this negative correlation between increased disclosure level and information asymmetry. While the proxies used to measure information asymmetry are suitable and well substantiated, the proxies used to measure disclosure level are reliable, but they are all subjectively determined which may create biased outcomes. Another limitation is that the AIMR reports only include the biggest, most followed firms and that the studies with self-constructed indexes contain a relatively small sample. Thus, it is necessary for the voluntary disclosure literature to find a more suitable, scalable proxy for disclosure level.
5
Voluntary disclosure dissemination
The main conclusion of the literature discussed above is that increased disclosure levels decrease information asymmetry. However, if these extra disclosures do not reach all stakeholders, the effect of the increased disclosures on the information problem may be relatively small. It is therefore important that the disclosures are widely disseminated. This distinction between disclosure and dissemination is also made in recent literature. In their study on the effect of the Corporate Filing Alert service on stock prices, Li, Ramesh, and Shen (2011) find that the alerts have a positive effect on the stock price and thus contain value-relevant information. Bushee, Core, Guay, and Hamm (2010) conclude that information asymmetry is reduced by greater press coverage and that the broadness of the dissemination has a bigger impact on information asymmetry than the quality or quantity of information in the press release. Twedt (2016) studies the effect of newswire dissemination on the price discovery process and finds a significant positive relation. These studies highlight the importance of the dissemination of disclosures. In this chapter, I will discuss and compare the different channels firms use to disseminate their voluntary disclosures.
5.1 Press release
One of the major channels used by firms to disseminate their corporate disclosures is through press releases. A press release is news that is written and distributed to the press by the firm. Firms therefore have full control over the content in the press release which is advantageous for the firm, but can be disadvantageous for other market participants if the
16 information in the release is biased. Over the years, the number of words used and the information content in press releases has increased significantly. Davis, Piger, and Sedor (2012) find that the amount of words in press releases by the end of 2013 has increased 90 percent relative to 1998. In response to raising concerns on the usefulness of press releases, Landsman and Maydew (2002) measure the information content of quarterly earnings press releases during the period 1972-1998 and conclude that the increase in used words is accompanied with an increase in information content. In their study on the use of press releases to signal the firms’ expected future performance, Davis et al. (2012) conclude that managers use press releases to communicate expected future performance and that the market responds significantly to opportunistic language. This indicates that market participants consider an opportunistic press release as a credible signal, despite the fact that managers can use opportunistic language on purpose to hide bad performance. Press releases can therefore be a useful channel for underperforming firms to disseminate their disclosures. However, this will not be beneficial for the information problem.
5.2 Press
The press plays an important role in the dissemination of corporate disclosures and is assumed to be the largest source of corporate information with a larger audience and more influence than analysts (Bushee et al., 2010). First, firms have to rely on the press to disseminate the information in their press releases. Second, the press writes about upcoming scheduled firm disclosures, which alerts interested investors. Finally, the press creates new information by writing articles in which they combine the information in a press release, quotes from analysts and management and their own interpretation of this information.
There are two main streams of literature on media coverage: (1) the literature on the monitoring role of the media in the capital market and (2) the literature on the effect of media coverage on the dissemination of corporate disclosures. The overall conclusion in the first stream of literature is that the press serves an important monitoring role, often identifying management malfeasance well before analysts and other intermediaries (Miller, 2006; Core, Guay, & Larcker, 2008; Dyck, Volchkova, & Zingales, 2008). The second stream of literature investigates the role of the press in disseminating disclosures. By minimizing or controlling for new information, the studies on the effect of media coverage on the dissemination of corporate disclosures try to isolate this effect (Sant & Zaman, 1996; Bushee et al., 2010; Engelberg &
17 Parsons, 2011; Drake, Guest, & Twedt, 2014). These studies collectively conclude that the press plays an important role in the dissemination of corporate disclosures, but that there is a bias towards highly visible firms because these firms generate more revenue than small, less-followed firms.
5.3 Conference call
Another way to disseminate disclosures is through conference calls. In a conference call, managers first present the firm’s financial performance and, if they wish, additional voluntary disclosures. After the presentation, analysts can ask questions and discuss the presented disclosures. Despite the Regulation FD, which prohibits firms to disclose information to only a select group of market participants, conference calls have become an increasingly common form of voluntary disclosure in the last twenty years (Bushee, Matsumoto & Miller, 2004). Similar to the studies on disclosure and the information problem presented earlier in this study, prior studies on the effect of conference calls on information asymmetry find that conference calls have a negative effect on information asymmetry (Frankel, Johnson & Skinner, 1999; Brown, Hillegeist, & Lo, 2004). Another conclusion from these studies is that conference calls have a greater negative effect on information asymmetry than press releases.
There are two reasons why conference calls can be more informative than press releases. First, because of vocal cues, a conference call can be more informative for market participants than a press release (Mayew & Venkatachalam, 2012). Second, conference calls are less formal than press releases and are therefore bound to less legal liability (Frankel, Johnson, & Skinner, 1999). Based on transcripts of more than 10.000 conference calls, Matsumoto, Pronk, and Roelofsen (2011) study whether conference calls have benefits relative to press releases. According to their findings, the main benefit of a conference call compared to a press release is due to the discussion part of the call. When a firm performs poorly, the discussion part is relatively more informative for analysts because of the unwillingness of managers to voluntarily disclose this poor performance. In the discussion it is possible for the analysts to gather more information by requesting more detailed information and asking critical follow-up questions. In line with this conclusion is the finding that conference calls of poor performing firms contain more non-financial, forward-looking disclosures relative to conference calls of better performing firms. Again, for small firms with little analyst following, conference calls may not be suitable to widely disseminate disclosures.
18 5.4 Social media (Facebook and Twitter)
5.4.1 Characteristics
“Social media refers to Web-based technologies that enable people to create, share, and exchange information in virtual communities and networks” (Lee, Hutton, & Shu, 2015, p. 372). This includes popular platforms such as Facebook and Twitter, video platforms such as YouTube and Twitch but also blogs and forums. As mentioned earlier, this study will only focus on Facebook and Twitter as channels to disseminate voluntary disclosures. The number of users on those platforms is still increasing. In 2016, 79% of the internet users in the U.S. was active on Facebook and 24% on Twitter against 67% and 16% in 2012.3 In July 2012, Netflix CEO Reed Hastings posted on his personal Facebook that the monthly online viewing of Netflix had exceeded one billion hours for the first time, which resulted in an increase of the stock price of Netflix from $70.45 at the time of the post to $81.71 at the close of the following day. The Division of Enforcement launched an inquiry on the post, which was the reason for the SEC to issue a report that states that companies can use Facebook and Twitter for corporate disclosure, as long as investors have been alerted about which social media will be used to disseminate such information.4 Since then, social media has become an increasingly important medium for firms to voluntary disclose and disseminate financial and non-financial information (Zhou et al., 2015). Of all S&P 1500 firms, 47% adopted Twitter, 44% adopted Facebook and 52% adopted either of the two (Jung et al., 2016). Noticeable is the fact that firms have a low adoption of Facebook and a high adoption of Twitter relative to internet-users.
Both Facebook and Twitter allow users to create a profile for their firm or products, through which they can post information and communicate and discuss with all interested users, including investors and analysts. However, the way firms can interact with users and manage their followers differs between the platforms. See appendix A for a comparison. The way in which firms can post messages also differs significantly. On both platforms it is possible to post text, pictures and hyperlinks. However, a Facebook message is limited to 63,206 characters, whereas a Tweet is limited to 140 characters. The study of Zhou et al. (2015)
3http://www.pewinternet.org/2016/11/11/social-media-update-2016/ 4https://www.sec.gov/news/press-release/2013-2013-51htm
19 highlights the effect this difference causes. On average, the messages on Facebook contain more characters, whereas the messages on Twitter contain more hyperlinks. Among the messages posted by firms, 3.45 percent of Twitter messages and 7.06 percent of Facebook messages contain corporate disclosures. On Facebook these corporate disclosures mainly consist of non-financial disclosures, while on Twitter these corporate disclosures mainly consist of financial disclosures. Some Twitter accounts are used to exclusively release financial disclosures. These findings highlight that firms prefer Twitter over Facebook to release financial information (Zhou et al., 2015). When focusing on user responsiveness and user engagement, Zhou et al. (2015) find that followers respond more quickly to disclosures released on Twitter than those on Facebook (13 minutes versus 25 minutes), while the engagement time of disclosures released on Facebook is much higher than those on Twitter (427 minutes versus 10 minutes).
5.4.2 Effect on firms
As mentioned above, before the rise of social media firms had to rely on the press to widely disseminate their disclosures. This is disadvantageous for small firms, because the press usually focuses on well-known firms which attract a large readership (Miller, 2006). Additionally, investors have limited time and resources and therefore rely on a few sources to obtain firm information (Hong and Stein, 1999; Hirshleifer and Teoh, 2003). Thus, if a firm that is not highly visible disseminates firm information through a small number of channels, it is likely that a portion of investors will not obtain the information on time. With the rise of social media, firms may now have a channel to more easily disseminate their disclosures to investors on a frequent and real-time basis and bypass information intermediaries such as the press.
Blankespoor, Miller, and White (2014) examine whether the effect of dissemination through social media on information asymmetry differs between highly-followed and less-followed firms. Because IT firms tend to adopt technology early, they limit their sample to 102 IT firms. The outcomes might therefore not generalize to other industries. Three measures are used to capture dissemination of corporate disclosures: (1) Tweets that provide a hyperlink to a press release, (2) an “abnormal” dissemination measure which captures Tweets with a link to a news article, a third-party blog or a corporate blog containing a link to the press release, and (3) a measure that captures the acquisition of the disseminated corporate disclosures. The
20 proxies used to measure firm visibility are: (1) firm size, (2) number of investors in the firms’ stock, and (3) number of institutions investing in the firms’ stock. To measure information asymmetry, the proxies abnormal bid-ask spread and market depth are used. According to Blankespoor et al. (2014), this removes correlation between the used proxies and any firm-fixed effects. Using abnormal bid-ask spread as proxy for information asymmetry, the results show that for highly-followed firms none of the dissemination proxies have a significant effect on information asymmetry while for less-followed firms all coefficients are negative and statistically significant (p ≤ 0.06). Using the abnormal market-depth as proxy for information asymmetry, the results show a statistical significant effect on information asymmetry for less-followed firms. There is also some evidence for a positive effect on information asymmetry for highly-followed firms. However, most coefficients for less-followed firms are significantly more positive than for highly-followed firms. These results suggest that less-followed firms benefit more from dissemination of disclosures through social media than highly-followed firms in the form of reduced information asymmetry.
Prokofieva (2015) examines the effect of social media as a secondary dissemination channel, i.e. the dissemination of corporate disclosures when they are already publicly available and tries to extend the findings of Blankespoor et al. (2014) by studying a more diverse set of firms in different industries. The sample consists of 109 listed Australian companies from the ASX 200 Index, which released 3,516 announcements via Twitter. To study the difference in effect of additional dissemination on highly-followed and less-followed firms, the sample is divided in two subsamples: (1) firms from the sample which are also incorporated in the ASX 100 and (2) firms which are incorporated in the ASX 200 but not in the ASX 100. In Australia it is prohibited for listed firms to disclose new information via any other channel than the Australian Stock Exchange (ASX). Once the information is submitted to the ASX Announcements Platform, firms can use other channels to disseminate the information. This distinguishes the Australian market from the market in the U.S., where firms can use their corporate website or social media to publish new information.
Within a three-day window around the event-day, information asymmetry is measured by the abnormal bid-ask spread (SpreadAbn). To measure the dissemination of ASX announcements via Twitter, Prokofieva (2015) introduces three variables which are similar to the variables used by Blankespoor et al. (2014):
21 1. TweetASX. Measures the use of Twitter to disseminate ASX announcements during the three-day period. The variable equals 1 if a Tweet is posted that relates to the ASX announcements and equals 0 otherwise.
2. TweetAbn. Measures whether the Twitter activity during the three-day period changes. It therefore also includes Tweets with links to news articles or blogs.
3. Retweet. Measures the acquisition of the disseminated information. The variable equals 1 if the ASX announcement-containing Tweet is retweeted and/or favorited at least once and equals 0 otherwise.
The results for the complete sample show that TweetASX (p = 0.01), TweetAbn (p = 0.1) and Retweet (p = 0.05) all have a significant negative relation with SpreadAbn. This suggests that dissemination of corporate disclosures through social media negatively affects the bid-ask spread and therefore information asymmetry. Consistent with Miller (2006) and Blankespoor et al. (2014), the results for the two subsamples suggest that less-followed firms benefit more from the dissemination through Twitter than highly-followed firms. For non-ASX 100 firms, both TweetASX and TweetAbn are more negatively correlated with SpreadAbn (p = 0.01 and p = 0.05) than for ASX 100 firms (p = 0.1 and p = 0.1). Retweet is negative and significant for non-ASX 100 firms (p = 0.05) and insignificant for ASX 100 firms.
5.4.3 Interactivity
According to prior empirical literature, dissemination of corporate disclosures through social media has a negative effect on information asymmetry which is stronger for less-followed firms than for highly-less-followed firms. Social media may therefore be a useful channel for less-followed firms to more widely disseminate corporate disclosures to reduce information asymmetry. However, because social media is more interactive than the traditional dissemination channels, firms relinquish full control over the content. This makes it possible for malicious people to induce a negative tone into the dialog. What is the effect of this interactivity on the reduction of information asymmetry?
Lee et al. (2015) investigate the capital market consequences of the interactivity of social media in the context of consumer product recalls. Dysfunctional products can cause both legal and reputational damage to the firm. It is therefore needful for firms to initiate a product recall, which has to reach all consumers as quickly as possible. This has two potential effects.
22 First, firms can inform consumers and the public and respond to false or misleading posts to reduce the scope of the crisis. Second, however, this may also cause the news about the recall to spread to a wider audience. Because of the interactivity of social media, consumers can disseminate and comment on news about the recall, potentially spreading a negative tone and/or misinformation. The sample consists of 405 consumer product recalls between 2000 and 2012. In this period social media became increasingly more interactive which creates a relatively suitable empirical setting for this study. The results suggest that by using social media to disseminate the product recall, the negative price reaction to the recall is attenuated. But as social media becomes more interactive, this attenuating effect lessens (but is not fully lost) because of the negative tone of dissatisfied customers in the dialog. To reduce this negative tone, firms can actively engage in the online dialog to influence and direct the thoughts of consumers and the public. This study highlights the disadvantage of voluntary disclosure through social media, i.e. the possibility that everyone can give their opinion on the disclosed information and the broad public that this opinion can reach.
Jung et al. (2016) examine the effect of the dissemination of earnings announcements (EA) through social media. Their sample consists of Tweets from all S&P 1500 firms with an active Twitter account, it therefore includes firms from multiple industries. They split their sample into a direct and an indirect audience. The former being the amount of users who receive the Tweet directly (followers of the firm) and the latter being the amount of users who receive the Tweet through a retweet (non-followers of the firm). Users who follow the firm choose to do so and are likely familiar with the firm’s performance, while non-followers are not. These users may increase information asymmetry if they behave as uninformed investors (Diamond & Verrecchia, 1991; Kim & Verrecchia, 1994). The proxy used to measure information asymmetry is abnormal bid-ask spread (ABN_SPREAD). EA_FIRM_FOLLOWERS is the proxy used to measure the amount of followers of the firm at the moment of the EA Tweet and RETWEET_FOLLOWERS is the proxy to measure the amount of followers who receive the Tweet through a retweet. The results show a significant negative relation between EA_FIRM_FOLLOWERS and ABN_SPREAD, which indicates a smaller absolute bid-ask spread for firms with more followers and suggests that improved dissemination reduces information asymmetry. This result is consistent with the results found in Blankespoor et al. (2014) and Prokofieva (2015). However, Jung et al. (2016) also find a significant positive relation between RETWEET_FOLLOWERS and ABN_SPREAD, which indicates that the
23 reduction in absolute bid-ask spread is attenuated by retweets to users who do not follow the firm. This result is consistent with the results found by Lee et al. (2015), suggesting that user-initiated dialogs may have a countervailing effect on the reduction of information asymmetry.
5.4.4 Conclusion
The overall conclusion of prior literature on voluntary disclosure dissemination through social media is that it is a helpful tool to widely disseminate corporate information in a short amount of time and as a result decrease information asymmetry (Blankespoor et al., 2014; Prokofieva, 2015; Jung et al., 2016). However, the interactivity of social media attenuates the positive effect on information asymmetry, but it does not eliminate it completely (Lee et al., 2015; Jung et al., 2016). This may be the reason for the fact that firms have a higher adoption of Twitter and lower adoption of Facebook relative to internet-users. As mentioned above, Twitter has a higher user responsiveness and lower engagement time, which highlights the fact that Twitter is more focused on information dissemination while Facebook is more focused on user engagement and interactivity. This makes it easier on Twitter to control and direct the dialog on the disclosures (Lee et al., 2014). Results from prior empirical studies suggest that the negative relation between dissemination through social media and information asymmetry is bigger for less-followed firms than for highly-followed firms (Blankespoor et al., 2014; Prokofieva, 2015). This is explained by the fact that highly-followed firms already reach a broad set of stakeholders without extra dissemination through social media. Then, only the downside of social media remains, which is the possibility that the firm loses control over the dialog about their disclosures. This has a countervailing effect on the reduction of information asymmetry through dissemination of disclosures (Lee et al., 2015; Jung et al, 2016).
6
Discussion and conclusion
The goal of this paper is to provide insight in the effect of voluntary disclosure dissemination through social media on information asymmetry. By combining the literature on the effect of voluntary disclosure on information asymmetry and the literature on the use by firms of social media and other channels to disseminate voluntary disclosures this paper tries to find an answer to the question whether dissemination of voluntary disclosures through social
24 media reduces information asymmetry and if so, what the benefits and drawbacks are compared to other dissemination channels.
Prior empirical research finds a negative relation between voluntary disclosure and information asymmetry. While the proxies used to measure information asymmetry are all suitable and well substantiated, the proxies used to measure voluntary disclosure level are all subjectively determined which may create biased outcomes. Another limitation is the fact that the AIMR reports only include the biggest, most followed firms, while the self-constructed proxies only contain a relatively small sample. It is therefore necessary for the voluntary disclosure literature to construct a more suitable, scalable proxy to measure voluntary disclosure level. Nevertheless, the economic theory that the information problem can be reduced by increased disclosures is substantiated by empirical research.
Firms have a couple of available channels to disseminate their disclosures. Until recently, firms could only use press releases and conference calls to disseminate their disclosures. Firms then had to rely on analysts and the press to further disseminate their disclosures. This made the positive impact of increased disclosures on the information problem relatively small for less-followed firms (Miller, 2006). However, since 2013 it is allowed by the SEC to use Facebook and Twitter to disseminate corporate disclosures. Blankespoor et al. (2014) find that dissemination of voluntary disclosures through social media reduces information asymmetry and that less-followed firms indeed have a relative advantage compared to highly-followed firms. However, their sample is small and limited to a single industry. Prokofieva (2015) draws the same conclusion based on a sample containing firms from multiple industries, but the sample size is still limited. With a sample size of all S&P 1500 firms with an active social media account, Jung et al. (2016) draw the same conclusion. According to Lee et al. (2015), the positive effect on the information problem is attenuated as the interactivity of social media grows. This explains the preference of firms to use Twitter instead of Facebook to disseminate corporate disclosures.
The drawback of social media compared to other dissemination channels is the fact that the press and analysts also serve a monitoring role. Market participants consider opportunistic language as a credible signal, despite the fact that firms can use opportunistic language to hide bad performance. While press releases and conference calls are monitored by the press and analysts, Facebook posts or Tweets are not.
25 Based on economic theory and prior empirical research, it can be concluded that dissemination of voluntary disclosures through social media reduces information asymmetry. As long as the firm can control the dialog on the disclosures, social media can be a helpful channel for firms to widely disseminate their disclosures and thereby reduce information asymmetry.
This study provides insight for firms in the benefits and drawbacks of using social media to disseminate voluntary disclosures. It adds to the disclosure literature by combining multiple limited empirical studies to find a more conclusive answer on the research question.
26
References
Akerlof, G. A. (1970). The market for "lemons": Quality uncertainty and the market mechanism. The Quarterly Journal of Economics, 488-500.
Bertomeu, J., & Marinovic, I. (2015). A theory of hard and soft information. The Accounting
Review, 91(1), 1-20.
Beyer, A., Cohen, D. A., Lys, T. Z., & Walther, B. R. (2010). The financial reporting environment: Review of the recent literature. Journal of Accounting and
Economics, 50(2), 296-343.
Blankespoor, E., Miller, G. S., & White, H. D. (2014). The role of dissemination in market liquidity: Evidence from firms' use of Twitter™. The Accounting
Review, 89(1), 79-112.
Botosan, C. A. (1997). Disclosure level and the cost of equity capital. Accounting review, 323-349.
Brown, S., Hillegeist, S. A., & Lo, K. (2004). Conference calls and information asymmetry. Journal of Accounting and Economics, 37(3), 343-366.
Brown, N. C., Stice, H., & White, R. M. (2015). Mobile communication and local information flow: Evidence from distracted driving laws. Journal of Accounting Research, 53(2), 275-329.
Bushee, B. J., Core, J. E., Guay, W., & Hamm, S. J. (2010). The role of the business press as an information intermediary. Journal of Accounting Research, 48(1), 1-19.
Bushee, B. J., Matsumoto, D. A., & Miller, G. S. (2004). Managerial and investor responses to disclosure regulation: The case of Reg FD and conference calls. The Accounting
Review, 79(3), 617-643.
Cahan, S. F., Rahman, A., & Perera, H. (2005). Global diversification and corporate disclosure. Journal of International Accounting Research, 4(1), 73-93.
27 Chau, G. K., & Gray, S. J. (2002). Ownership structure and corporate voluntary disclosure in Hong Kong and Singapore. The International Journal of Accounting, 37(2), 247-265.
Chau, G., & Gray, S. J. (2010). Family ownership, board independence and voluntary disclosure: Evidence from Hong Kong. Journal of International Accounting,
Auditing and Taxation, 19(2), 93-109.
Cheng, E. C., Courtenay, S. M., & Krishnamurti, C. (2006). The impact of increased voluntary disclosure on market information asymmetry, informed and uninformed trading. Journal of Contemporary Accounting & Economics, 2(1), 33- 72.
Core, J. E., Guay, W., & Larcker, D. F. (2008). The power of the pen and executive compensation. Journal of Financial Economics, 88(1), 1-25.
Dainelli, F., Bini, L., & Giunta, F. (2013). Signaling strategies in annual reports: evidence from the disclosure of performance indicators. Advances in Accounting, 29(2), 267-277.
Davis, A. K., Piger, J. M., & Sedor, L. M. (2012). Beyond the numbers: Measuring the information content of earnings press release language. Contemporary Accounting
Research, 29(3), 845- 868.
Diamond, D. W., & Verrecchia, R. E. (1991). Disclosure, liquidity, and the cost of capital. The
journal of Finance, 46(4), 1325-1359.
Drake, M. S., Guest, N. M., & Twedt, B. J. (2014). The media and mispricing: The role of the business press in the pricing of accounting information. The Accounting
Review, 89(5), 1673-1701.
Dutta, S., & Trueman, B. (2002). The interpretation of information and corporate disclosure strategies. Review of Accounting Studies, 7(1), 75-96.
Dyck, A., Volchkova, N., & Zingales, L. (2008). The corporate governance role of the media: Evidence from Russia. The Journal of Finance, 63(3), 1093-1135.
28 Dye, R. A. (1985). Disclosure of nonproprietary information. Journal of Accounting
Research, 123-145.
Easley, D., Kiefer, N. M., O'Hara, M., & Paperman, J. B. (1996). Liquidity, information, and infrequently traded stocks. The Journal of Finance, 51(4), 1405-1436.
Eng, L. L., & Mak, Y. T. (2003). Corporate governance and voluntary disclosure. Journal of Accounting and Public Policy, 22(4), 325-345.
Frankel, R., Johnson, M., & Skinner, D. J. (1999). An empirical examination of conference calls as a voluntary disclosure medium. Journal of Accounting Research, 37(1), 133-150.
Grammig, J., Schiereck, D., & Theissen, E. (2000). Informationsbasierter Aktienhandel über IBIS. Schmalenbachs Zeitschrift für betriebswirtschaftliche Forschung, 52(7), 619-642.
Grossman, S. J. (1981). The informational role of warranties and private disclosure about product quality. The Journal of Law and Economics, 24(3), 461-483.
Grossman, S.J., & Hart, O.D. (1980). Takeover bids, the free-rider problem, and the theory of corporation. The Bell Journal of Economics, 11(1), 42-64.
Guay, W., Samuels, D., & Taylor, D. (2016). Guiding through the fog: Financial statement complexity and voluntary disclosure. Journal of Accounting and Economics, 62(2), 234-269.
Haniffa, R. M., & Cooke, T. E. (2002). Culture, corporate governance and disclosure in Malaysian corporations. Abacus, 38(3), 317-349.
Healy, P. M., Hutton, A. P., & Palepu, K. G. (1999). Stock performance and intermediation changes surrounding sustained increases in disclosure. Contemporary
Accounting Research, 16(3), 485-520.
Healy, P. M., & Palepu, K. G. (2001). Information asymmetry, corporate disclosure, and the capital markets: A review of the empirical disclosure literature. Journal of accounting
29 Ho, S. S., & Wong, K. S. (2001). A study of the relationship between corporate governance structures and the extent of voluntary disclosure. Journal of International Accounting,
Auditing and Taxation, 10(2), 139-156.
Jung, M. J., Naughton, J. P., Tahoun, A., & Wang, C. (2016). Do firms strategically disseminate? Evidence from corporate use of social media.
Kim, O., & Verrecchia, R. E. (1994). Market liquidity and volume around earnings announcements. Journal of Accounting and Economics, 17(1-2), 41-67.
Korn, E., & Schiller, U. (2003). Voluntary disclosure of nonproprietary information: A
complete equilibrium characterization. Journal of Business Finance &
Accounting, 30(9‐10), 1327-1339.
Landsman, W. R., & Maydew, E. L. (2002). Has the information content of quarterly earnings announcements declined in the past three decades?. Journal of Accounting
Research, 40(3), 797-808.
Lang, M., & Lundholm, R. (1993). Cross-sectional determinants of analyst ratings of corporate disclosures. Journal of Accounting Research, 246-271.
Lee, L. F., Hutton, A. P., & Shu, S. (2015). The role of social media in the capital market: evidence from consumer product recalls. Journal of Accounting Research, 53(2), 367-404.
Leuz, C., & Verrecchia, R. E. (2000). The economic consequences of increased disclosure (digest summary). Journal of Accounting Research, 38(3), 91-124.
Lev, B. (1988). Toward a theory of equitable and efficient accounting policy. Accounting
Review, 1-22.
Li, E. X., Ramesh, K., & Shen, M. (2011). The role of newswires in screening and disseminating value-relevant information in periodic SEC reports. The Accounting
Review, 86(2), 669-701.
Lim, S., Matolcsy, Z., & Chow, D. (2007). The association between board composition and different types of voluntary disclosure. European Accounting Review, 16(3), 555-583.
30 Luo, S., Courtenay, S. M., & Hossain, M. (2006). The effect of voluntary disclosure,
ownership structure and proprietary cost on the return–future earnings relation. Pacific-Basin Finance Journal, 14(5), 501-521.
Matsumoto, D., Pronk, M., & Roelofsen, E. (2011). What makes conference calls useful? The information content of managers' presentations and analysts' discussion sessions. The Accounting Review, 86(4), 1383-1414.
Mayew, W. J., & Venkatachalam, M. (2012). The power of voice: Managerial affective states and future firm performance. The Journal of Finance, 67(1), 1-43.
McNally, G. M., Eng, L. H., & Hasseldine, C. R. (1982). Corporate financial reporting in New Zealand: An analysis of user preferences, corporate characteristics and disclosure practices for discretionary information. Accounting and Business
Research, 13(49), 11-20.
Meek, G. K., Roberts, C. B., & Gray, S. J. (1995). Factors influencing voluntary annual report disclosures by US, UK and continental European multinational corporations. Journal of International Business Studies, 26(3), 555-572.
Milgrom, P. R. (1981). Good news and bad news: Representation theorems and applications. The Bell Journal of Economics, 380-391.
Miller, G. S. (2006). The press as a watchdog for accounting fraud. Journal of Accounting
Research, 44(5), 1001-1033.
Patelli, L., & Prencipe, A. (2007). The relationship between voluntary disclosure and independent directors in the presence of a dominant shareholder. European
Accounting Review, 16(1), 5-33.
Petersen, C., & Plenborg, T. (2006). Voluntary disclosure and information asymmetry in Denmark. Journal of International Accounting, Auditing and Taxation, 15(2), 127-149.
Prokofieva, M. (2014). Twitter-based dissemination of corporate disclosure and the intervening effects of firms' visibility: Evidence from Australian-listed companies. Journal of
31 Raffournier, B. (1995). The determinants of voluntary financial disclosure by Swiss
listed companies. European Accounting Review, 4(2), 261-280.
Shroff, N., Sun, A. X., White, H. D., & Zhang, W. (2013). Voluntary disclosure and information asymmetry: Evidence from the 2005 securities offering reform. Journal of Accounting Research, 51(5), 1299-1345.
Spence, M. (1973). Job market signaling. The Quarterly Journal of Economics, 87(3), 355-374.
Twedt, B. (2016). Spreading the word: Price discovery and newswire dissemination of management earnings guidance. The Accounting Review, 91(1), 317-346.
Verrecchia, R. E. (1983). Discretionary disclosure. Journal of Accounting and
Economics, 5, 179-194.
Verrecchia, R. E. (2001). Essays on disclosure. Journal of Accounting and
Economics, 32(1), 97-180.
Wagenhofer, A. (1990). Voluntary disclosure with a strategic opponent. Journal of
Accounting and Economics, 12(4), 341-363.
Wang, K., Sewon, O., & Claiborne, M. C. (2008). Determinants and consequences of voluntary disclosure in an emerging market: Evidence from China. Journal of
International Accounting, Auditing and Taxation, 17(1), 14-30.
Welker, M. (1995). Disclosure policy, information asymmetry, and liquidity in equity markets. Contemporary Accounting Research, 11(2), 801-827.
Zhou, M., Lei, L., Wang, J., Fan, W., & Wang, A. G. (2014). Social media adoption and corporate disclosure. Journal of Information Systems, 29(2), 23-50.
32
APPENDIX A
Comparison of interactivity Twitter and Facebook
Follow a page: ability for users to follow
the page. Tweets from the page will show on the feed of the follower.
Favourite a page: number of users that
added the page to a list. Users see all the messages from pages in that list.
Favourite a tweet: save the tweet in a list
of favourite tweets.
Retweet a tweet: retweet the tweet to all
followers.
Reply to a tweet: reply to the tweet, can
also be seen by all followers.
Like a page: ability for users to like the
page. Posts of the page will show on the news feed of the user.
Comments on posts: users can comment on
posts of the page, this will show on the news feed of friends of the commenter.
Private message: users can send a private
message to the page owner.
Share a page: users can share the page
with all their friends.
Like a post: users can like a post, which
will then pop up on the news feed of their friends.
Share a post: users can share a post, which
will then pop up on their own page and in the news feed of their friends.
Comment on a post: users can comment on
a post, which will then show up on the news feed of their friends.
