Projects

This paper examines how and to what extent financial media coverage is affected by events in cryptocurrency markets. I find that the financial media responds to the rising demand for news about the cryptocurrency market by either delaying earnings news coverage or lowering the quality of earnings news articles. On crypto event days, earnings news articles are shorter and rely more on quantitative data suggesting that fewer resources were allocated by media firms to earnings announcement news articles. Capital markets react less to earnings announcements as a result of the reduced coverage by financial media on crypto event days. Interestingly, media increases coverage of firm-specific events in the 30-day period after the crypto event to cater to the informational demand from equity investors.

Using realized earnings over long periods of time, we investigate errors in earnings expectations implied by stock prices of firms. We compute realized lifetime earnings for each firm starting at the IPO date and the beginning of each subsequent year and compare it to stock price on that date. Of the 16,386 firms examined, only 17% survived till 2019, 42% merged with other firms and the rest were delisted for other reasons. While the average lifetime earnings at the aggregate level slightly exceeds first day price, the results are driven by roughly 33% of the firms in the sample. Mergers account for most of success in recovering the first day stock price and appear to be the best way to generate enough earnings to justify valuations. Even among firms that survived, over 46% have yet to generate enough earnings to justify their first day valuations even though they have been in business between 15 and 45 years. Aggregate free cash flows over the lifetime of all firms are lower than the lifetime earnings and justify 80% of the first day trading price. We relate our paper to Bessembinder (2018) by examining the link between lifetime earnings (as a measure of fundamental wealth creation) and stock returns-based wealth creation. We find that lifetime earnings are positively associated with future returns-based wealth creation while current returns-based wealth created itself is negatively associated with future returns-based wealth creation. These results point to a disconnect between returns-based wealth creation and fundamental wealth creation in the short to medium term that eventually corrects.

We provide evidence that a firm’s investor base is a key factor in determining its media coverage. Using a large sample of U.S. public companies spanning the period 2000-2019, we find that firm-specific media coverage is negatively associated with leverage. Our results suggest the media caters relatively more to equity investors, who largely rely on public information, than to debt investors, who are often privy to privileged information. We arrive at similar conclusions when we exploit plausibly exogenous changes in investor base and investor access to privileged information. Among debt investors, the media appears to cater to public bond holders more than private banks. Among equity investors, the media appears to cater to quasi-index institutions. Additionally, cross-sectional variation in media coverage based on investment and article types reveals that the media’s role is far more nuanced than previously documented. Overall, our findings suggest that financial media coverage is significantly influenced by less sophisticated professional investors’ demand for information to help monitor firms.

In this paper, we study the relationship between the cryptocurrency market and investor reactions to earnings news. Building on the limited investor attention literature, we hypothesize that the cryptocurrency market distracts investors from earnings news. We measure the cryptocurrency distraction as the magnitude of two-week cumulative cryptocurrency returns before earnings announcements, and find that the cryptocurrency distraction is associated with a weaker price reaction to earnings surprises. Our result is robust when using the Chinese bans on cryptocurrency offerings and exchanges as sources of exogenous variation in cryptocurrency distraction. Consistent with the limited attention hypothesis, we document that busy earnings announcement days experience greater cryptocurrency distraction. To corroborate the main finding, we further show that the Google search volume and abnormal trading volume around earnings announcements are lower when there is a greater amount of cryptocurrency distraction. Cryptocurrency distraction effect is concentrated in retail trading rather than institutional trading, and consistently, is associated with more severe information asymmetry. Overall, our study provides evidence that cryptocurrency, a class of asset with no intrinsic value, disrupts information processing and thus price formation in the equity market.