Recently, I came across an interesting white paper by Jeremiah Green et. al. looking into the sentiment biases in the media. Similar to my previous research, they find that the number of good-news events reported in the business press exceeds the number of bad-news events. Despite this positive news sentiment bias, they find that bad-news events tend to receive broader dissemination after they first occur, especially for small firms and less newsworthy events. They find evidence that this dissemination bias in favor of bad news arises because bad news grabs journalists’ limited attention, not because journalists respond to incentives to increase readership by "leading with what’s bleeding". Consistent with managers exploiting journalists’ limited attention, they find that firms in general, and large firms in particular, issue more good-news press releases than bad-news press releases.
Understanding these different types of biases in news is important, especially when trying to capture the prevailing sentiment trend on a given market, sector, or stock, as it highlights the necessity to condition on what entity initiated the news (a company vs. business press) when coming up with such measures. Furthermore, it may be necessary to adjust for the dissemination bias when ranking stocks according to sentiment, particularly in a universe of both large and small cap stocks.
You may download this study here.
Subscribe to:
Post Comments (Atom)
Posts
0 comments:
Post a Comment