Essays in Corporate Finance: The Role of Social Media in Financial Markets

In this thesis, I examine the role of investor-oriented social media platforms in the financial markets. This thesis comprises three standalone yet interconnected chapters in the field of corporate finance, which investigates and discusses three vital constructs of attention allocation in the financial markets.

Chapter two focuses on the attention allocation by investors on social media and its consequences for investors in the financial markets. Using more than 32 million tweets and a market microstructure dataset, this chapter investigates the impact of social media attention (SMA) on investors' trading behaviors. Through a battery of experiments, I find that SMA is a unique proxy of retail investors' attention and is different from other direct proxies. Aggregate data suggests that SMA can predict the financial markets, i.e., higher SMA results in short-term price pressures generated from more buys than sells. However, short-term price pressures are reversed the next day. Using social media heterogeneity, I test the information diffusion hypothesis. The empirical findings suggest that the impact of social media is amplified within the large social network. Moreover, tweets by verified users, replies, and retweets further increase the credibility of the information. This chapter contributes to the emerging body of knowledge that investigates the impact of social media attention on financial markets and provides valuable insights for a diverse set of market participants, mainly –retail investors.

In chapter three, I investigate whether disagreement on StockTwits provides firm-specific information. Using supervised machine learning approaches and a novel dataset, I predict investors' recommendations and measure disagreement among investors on StockTwits. The findings suggest that an increase in investors' disagreement results in a drop in return synchronicity. The negative impact of investors' disagreement on return synchronicity suggests higher inflows of firm-specific information. In line with this view, I find that disagreement improves price informativeness by increasing the price leads of earnings. Further empirical evidence suggests that the negative impact of disagreement on return synchronicity is more pronounced for firms with less transparent information environments and higher salience on StockTwits.

In chapter four, I examine the role of investor-oriented social media platforms to predict crash risk. Using the investor-level novel dataset from StockTwits and developing a unique proxy of investors' update in sentiment i.e., sentiment oscillations, I find that sentiment oscillations on StockTwits are significantly and positively related to the firm-level future crash risk. These results remain consistent after using a battery of tests to account for unexpected market events, heterogeneity of investors, and endogeneity concerns using the instrumental variable approach. Further tests suggest that the impact of sentiment oscillations is more pronounced in firms with a lower level of accounting conservatism, less analyst coverage, less product market competition, and positive market sentiment. Overall, this study highlights the significance of social media platforms for investors and sheds light on the behavioral explanation of firm-level crash risk.