Three Essays in Empirical Asset Pricing

This dissertation consists of three essays in empirical asset pricing concerning how investor attention and social interactions impact information diffusion in financial markets. In the first essay, I investigate how different investor attention facilitates information diffusion through the customer-supplier network. I find that retail attention improves information incorporation into asset prices and plays a stabilizing role in financial markets. In contrast, institutional attention exhibits a diminished role if I control for retail attention. In addition, I show that the attention of a potential group of informed retail investors, local retail investors, plays a price-stabilizing role beyond that of uninformed (non-local) retail investors. My results provide a refinement on the view of the role played by retail investors. While the literature argues that retail investors destabilize prices, my findings suggest that at least a group of informed retail investors can stabilize financial markets. In the second essay, I examine how social connectedness affects fund manager stock holdings during the COVID19 pandemic. I exploit the recent outbreak of COVID-19 as an exogenous shock to people's beliefs on the future economic condition to examine how fund managers from different regions make different decisions on stock holdings. By applying a unique dataset from Facebook that measures social interaction among different regions, I am able to identify managers from COVID-19 hotspot counties and those highly socially connected to the hotspot counties. I am also able to identify fund managers that are skilled using standard methodologies exploiting fund alpha and other performance metrics. The results show that managers located in or socially connected to hotspot counties sold more stock holdings during the outbreak of COVID-19 in the first quarter of 2020. However, such reductions appear panic-driven given subsequent behavior and outcomes and in particular given the contrasting behavior and outcomes for skilled versus unskilled fund managers. The evidence suggests that social interaction can intensify salience bias even for institutional investors if they are unskilled, but skilled managers appear relatively impervious to the deleterious effect of social networking. Finally, in the third essay, I explore the role of institutional and retail attention in the context of the media news releases and find nuanced evidence of the costs and benefits to market price adjustments flowing from investor attention to news. I show that retail attention does indeed destabilize financial markets by inducing price overreactions to positive news, but only if it is from uninformed retail investors. I find that when retail attention destabilizes the market, it is when retail investors appear to struggle digesting complex business information and then only if the news is of a positive sentiment; negative sentiment news and retail investor attention are not associated with market instability, possibly a result of the well-known reluctance of retail investors to short sell. I also find that institutional attention plays a stabilizing role in any context I explore, complex or simple news, positive or negative news sentiment, with or without retail investor attention.