Abstract
By quantifying the tone of firm-specific articles in leading national newspapers between 1989 and 2010, we propose a bottom-up measure of aggregate journalist disagreement. In line with theoretical considerations, our novel high-frequency proxy for differences of opinion negatively forecasts the market return, in particular during recessions. Moreover, it has predictive power for the cross-section of stock returns. Collectively, our insights support asset pricing theories incorporating belief dispersion and highlight the role of the media in this context.
Original language | English |
---|---|
Pages (from-to) | 57-76 |
Number of pages | 20 |
Journal | Journal of Financial Markets |
Volume | 41 |
DOIs | |
State | Published - Nov 2018 |
Externally published | Yes |
Keywords
- Differences of opinion
- Journalists
- Media
- Return predictability
- Textual analysis