Conventional wisdom has been that newspapers don’t add much to stock markets. However, don’t write off the power of the press too fast.
Here’s a message for business people wondering whether it’s worth their time talking to journalists: “If you want to get your company noticed favourably by investors, speak to the press.”
Sounds obvious? Not if you believe the efficient-market hypothesis, which holds that news is incorporated into stock prices as soon as it is made public. Since the 1970s, investment theory has been strongly influenced by the notion that markets react immediately to new information. Under such a scenario, there is little scope for the media to influence prices.
Now, however, INSEAD Associate Professor of Finance Joel Peress has found evidence suggesting that markets may need the media more than previously assumed – even by researchers such as himself. Drawing on data for selected national stock markets, he has shown that the availability (or otherwise) of newspapers, independently of what they are actually reporting, is a factor affecting trading activity.
In a study covering two decades, he found that both trading volume and price volatility fell on days when newspapers were unavailable because of strike action. What’s more, he found the impact was strongest for small-capitalisation stocks, commonly traded by retail investors who rely more than big institutional investors on the media for information.
“This suggests that the media does play a role in supporting market activity,” Peress told INSEAD Knowledge. In essence, he explains, the media underpins markets by providing investors with an information flow that enables them to feel comfortable about trading. “It’s about access to news. There’s an influence of the press on investors, for good or for bad. The press influences the way people behave by providing the news that they need.”
For his research, Peress sought to get as close as possible to a controlled environment for observing stock market trading with and without media presence. Going back to the early nineties, when trading volume data for national stock markets first became widely available, he trawled through databases in search of trading days when newspapers were unavailable due to nationwide strikes in the media sector. He excluded general strikes affecting multiple sectors, in order to avoid the influence of other extraneous factors on stock prices, and days on which markets were not open.
After extensive investigation, he ended up with several dozen eligible national newspaper strikes, concentrated in just four countries: France, Greece, Italy and Norway. (Other countries with bigger stock markets failed to make the grade because they didn’t provide appropriate examples of nationwide media black-outs.)
Most of the strikes that fulfilled Peress’s criteria were due to disputes over pay and other aspects of working conditions in the media sector, though some were called in protest against threats to the freedom of the press. In France, most strikes were by print and distribution workers, rather than by journalists. But in the other three countries, strikes were mostly staged by journalists, going on strike in the summer of 2004 in Greece for higher pay, for example, or in Norway over vacation benefits. In Italy, journalists staged numerous strikes over the 20-year period, including one in June 2003 in protest against the concentration of media owned by then prime minister Silvo Berlusconi, and another in July 2010 to challenge a proposed law deemed to violate media freedom.
In all cases, Peress made sure that the strike actions were exogenous to (i.e. not driven by) stock market movements on the day of the strike or on preceding days. He also excluded strikes by production and distribution workers after 1996, when Internet-based editions of major newspapers began to be widely available.
No news, less trading
The results of his research were surprising. “I found that on the day that a newspaper strike occurred, share turnover on the country’s stock market averaged 14 percent lower, while remaining unchanged on the days before and after,” he reports.
The strike effect was strongest for the shares of small and medium-sized companies, while being close to zero for the biggest companies. For shares of the bottom quintile of companies by market capitalisation, trading volume dropped by an average 24 percent.
Price volatility also fell on strike days, particularly for smaller companies. In addition, small-cap companies showed a smaller-than-normal price reaction to trading outcomes for large-cap firms on the previous day. As Peress notes, published studies have shown that price swings in the shares of smaller companies typically tend to lag a day behind those of bigger stocks. The new wrinkle produced by his study is that on the days when newspapers were unavailable, small-cap shares missed a beat, reacting to broader market movements only on the trading day following the strike.
Do his conclusions open up new avenues to wily traders hoping to make a quick profit? To the extent that they confirm that markets are not totally efficient all of the time, possibly yes. But with Internet-based news media ever more widely available, information gaps in stock markets are becoming more limited and the impact of strikes by “traditional” media is shrinking.
Perhaps the greatest comfort to be drawn from Peress’s study is for journalists wondering whether anyone pays attention to their work.
There, Peress has no doubt. “Firms can communicate effectively with prospective investors by communicating with newspapers,” he says. “The media play a critical role in allowing people access to news.”
Joel Peress is Associate Professor of Finance at INSEAD.