Media and Finance – An Interesting Chimera

By Prof. Dr. Thorsten Hens, University of Zurich
in collaboration with Prof. Dr. Michal Dzielinski from Stockholm University

In a financial market people trade opinions on the future development of assets. Thus nothing is more important than information. If you know better or quicker that a company has developed a great product, in the long run you can make a fortune by buying shares of it. And if you know how others think about the future returns on a company’s shares then you can benefit from this also in the short run. Being on top of the news is a comparative advantage for any investor. However, news comes from media and media have a life on their own. Understanding the interaction of media and finance is a key for success in the financial markets. In this article we give a rough overview about the leading edge research on media and finance.

New research on media finance has for example shown that the tone with which information is given in the media influences the way investors react to it. Based on huge data sets and clever algorithms to detect news tone researchers found that media makes emotions which then drives prices away from fundamentals. Media reports can therefore be valuable because they let us know what other investors think at the moment, or even what their mood is. In fact, Aeron Davis from the University of London proved the so-called "third-person effect" in a study in 2005; namely that investors react to media reports, because they believe that other market participants will be influenced by them. Moreover, Gur Huberman and Tomer Regev from Columbia University analyzing the particular case EntreMed have shown that media reports can trigger large price changes even if no new information is provided.

On Sunday, May 3, 1998, the New York Times reported on its front page about a promising cancer therapy which has been developed by the company EntreMed. The closing share price of EntreMed on the previous Friday was around 12 $. On Monday after the publication the share opened at $ 85 - this despite of the fact that the findings of the article had been known for months, since the same findings were already reported in the journal Nature (as well as even in the New York Times itself). Three aspects of this story are from the perspective of market efficiency particularly puzzling: (1) the response to this "no news" event was much greater than at the time when the information on the research EntreMed restarted been published, (2) the price rash was partly permanent - months later, the stock price was still twice as high, (3) some other biotechnology companies have been swept away by the eruption. In finance therefore more and more scientists are working with the interplay of media and financial markets. This article reports on their fascinating results.


Do media dampen or amplify volatility of financial markets?

A critical journalist should know that financial markets are developing out of themselves cycles that lead to major changes to the fundamentals. It is precisely this so-called "excess volatility" -- proven already in 1980 in the American Economic Review by Robert Shiller – winner of the Nobel Prize in economics in 2013. So one should expect that the media represents a critical look from the outside and helps dampen these fluctuations. This idealistic view fails, however, ignores that media are part of the system itself. At its input side, they depend on information about companies and macro- economics and on its output side they want to make sure that they are read. If you swim then too long against the flow, your sources dry up because the informant will be surprised each time when the media comments on their views to the contrary. Moreover, you cannot be heard, because according to the famous "confirmation bias" people do not consume media to get informed, but to find confirmation of their views. In fact, Anya Schiffrin (Columbia University) shows nicely in her book "Bad News. How America's Business Press Missed the Story of the Century" that the American media have not pointed to the financial crisis before it happened. On the other hand media in boom times too often put fuel to the fire. Then they justify severe departure from fundamentals with terms like "New Era Economics" to suggest that one is now living in special times in which the old economic trade-offs between risk and return no longer apply the old economic relationships.

A picture is worth a thousand words

The case EntreMed raises the wider question, whether investors generally react to irrelevant aspects of media coverage. In 2008, US scientists Matteo Arena and John S. Howe have investigated this question by the "Who's News" column examining company reports in the Wall Street Journal. They found that reports provided with an image of CEOs provoked a stronger reaction than those without image, even if both are next to each other directly published in the same issue. Even if the decision to illustrate an article is not random it is nevertheless the case that the media coverage causes an additive effect on the pure information addition.

All that glitters

A possible interpretation of the results of Arena and Howe is that media get more attention for company reports with pictures. Private investors who do not constantly deal with the financial market are prone to the "attention-grabbing" phenomenon. This has been shown by Brad Barber and Terry Odean from UC-Berkeley in California. In 2008 they have studied millions of trade orders that households and professional investment managers have delivered in the period from 1991 to 1996. Their results clearly show that media coverage generated attention effects from the private investors who tend very much to buy stocks that are currently mentioned in the media. Professional managers on the other hand do not have similar tendencies. Unfortunately, attention-driven assets do not generate excess returns and the associated frequent shifts in the depot may even be harmful for the investors.

From media exposure to media mood

The previously presented research has dealt only with the issue of media coverage, comparable to the "agenda setting" theory of media studies. The other very important and interesting aspect of the influence of media on the financial market is whether they also have a "Persuasion" effect, that is, whether the mood in the media can affect the mood in the financial market. There are now very efficient algorithms with which one can "quantify" the spirit of the language of thousands of media releases, in which one example the percentage of negative words (according to a predefined dictionary, the Harvard IV-4 Sociolinguistic Dictionary, for example) determined in the text. In two studies, which were based on this simple measure, Paul Tetlock from Columbia University and his co-authors were able to demonstrate that one can predict not only short-term returns but also corporate earnings better with media mood. Our own research shows that media influence the financial market much stronger in phases in which the uncertainty is high. For example, in periods of high uncertainty the markets are more responsive to positive and negative earnings surprises which has interesting implications for investment strategies on the "post-earnings announcement drift". For example, if the actual profit is above expectations of analysts, not only at the day of publication, but also thereafter the stock price has positive returns. For negative surprises the drift shows the mirror image of the positive development, which speaks for delayed information processing. The success of a strategy that buys shares with positive surprises and sells short those with negative surprises has been demonstrated in several studies. It can even be improved taking into account the mood of quarterly reports drift.

Where do we go from here?

The above examples show that, contrary to the efficient market hypothesis of traditional finance, media have an active role in the financial market. For these observations behavioral finance provides a framework. In fact, many aspects of the media impact on investor behavior and market returns can be traced back to so called behavioral biases – systematic deviations from rational behavior. The research on media and finance is very promising. We can only hope that the media will report objectively about.