CNBC mention — good or bad — helps a company stock price

CNBC is good for a company’s stock price regardless of whether the news is excellent or terrible, according to a new paper from a graduate student at the University of California at Berkeley.

Economics PhD student Reza Shabani writes, “There is a signifi…cant increase in stock price at the precise time that a company is mentioned on CNBC following a positive news event. Surprisingly, there is also a signifi…cant increase in stock price at the precise time that a company is mentioned on CNBC following a negative news event.”

Beginning in June 2009 through March 2011, Shabani recorded the CNBC broadcast on each trading day from 9 a.m. to 7 p.m. eastern time. He used a customized program to extract the closed captions from the video recordings along with an intrasecond time-stamp indicating the precise time at which a caption appears on the screen.

To identify companies, he developed a complex tagging algorithm that accounts for complications such as punctuation, wild cards, and colloquialisms. He found more than 22,000 examples of company news events mentioned, and used transactions data from the NYSE’s Trade-and-Quote database to calculate returns and volume in the 20-minute window surrounding announcements of news events and their subsequent mentions.

Shabani also found a signifi…cant increase in stock price at the precise time a company is mentioned on CNBC following a news event with an expected outcome. This price response is smaller and less statistically signifi…cant than the price response following subsequent mentions of positive and negative news events.

“The evidence implies that at the time of a CNBC mention, some investors are swayed by the opinions of pundits and analysts while others are introduced to an investment opportunity that they had not previously considered,” writes Shabani. “Media influence is strongest when the implications of a news event are more ambiguous and susceptible to a wide range of interpretations, as is likely with expected news events.”

If you’d like to read the full research paper, go here.