Candid Camera

New research suggests media coverage makes a difference in corporate governance. Coverage of corporate governance violations in the Anglo- American press increases the probability that these violations will be reversed. The media seems to exert pressure on corporate managers and regulators, forcing companies to behave more in the interest of shareholders.

In September 2003, Richard Grasso, then chairman of the New York Stock Exchange (NYSE), lost his job after his lavish pay was exposed and vilified in the press. Although all directors of the NYSE had voted in favor of his compensation, once the information became public,and even the most probusiness newspapers characterized Grasso's compensation in a very negative light,many directors changed their position.

Reputation considerations play an important role in corporate governance decisions, according to University of Chicago Graduate School of Business professor Luigi Zingales, Alexander Dyck of the University of Toronto, and Natalya Volchkova of Russia's Center for Economic and Financial Research (CEFIR) and New Economic School.

The existing literature, starting with a seminal paper by University of Chicago Graduate School of Business professor Eugene Fama, claims that corporate managers are constrained in their behavior by the impact that their actions have on their reputation vis-a-vis their future employers or the capital market in general. However, the authors argue in their paper "The Corporate Governance Role of the Media: Evidence from Russia," that reputation is an effective constraint only if the future employers or partners learn about their actions. Unfortunately, collecting information is costly, and thus the relevant audiencemight not become informed. The media, however, can change this state of affairs. By collecting and diffusing information about managers' actions, the media pressures managers to behave in the interest of their shareholders.

In the case of Grasso and the NYSE, the SEC began to ask the NYSE board about its compensation practices after news of Grasso's compensation was published in The Wall Street Journal. The publication of that news and ensuing public outcry forced the same directors who voted for his compensation package to fire him.

To document the empirical relevance of these effects, the authors looked for a natural experiment. "As social scientists, we're not often allowed to do real experiments, so we do the next best thing, which is using real world situations that approximate the situation of a controlled experiment," says Zingales. "In a real experiment, we would take a sample of corporate governance violations, randomly trigger news coverage about some of them, then watch the results. For this study, we focused on Russia, because there was a large sample of very visible corporate governance violations and an institutional investor, William Browder of the Hermitage Fund, who fought hard to expose violations affecting the companies in which he invested. Since Browder's initial portfolio was not assembled with this scope in mind, it is pretty close to a random sample of companies and hence to a random experiment."

Russia: The Right Setting

Several conditions make Russia an ideal laboratory setting. First, during the time frame covered in the study (1998-2002), especially during the late 1990s, corporate governance abuses in Russia were extreme, common, and visible. This provides a wide field of inquiry. Second, the standard mechanisms to correct these abuses in Russia were either nonexistent or completely ineffective (for example, courts were easily corruptible). This allowed the authors to identify whether the media had an independent effect on outcomes. Third,and most important,William Browder, manager of the Hermitage Fund, had a conscious media strategy for addressing corporate governance abuses.

The Hermitage Fund was founded in 1996 as a generic hedge fund with a Russian focus and found itself drawn into corporate governance battles. As the largest foreign investor in Russian equities, Browder could not be passive about major corporate governance abuses. However, the legal remedies at his disposal were extremely weak. Thus, to protect his fund's investments, Browder chose to actively shame Russian companies in the international press, hoping to hurt their reputation and the reputation of government officials who regulated these firms.

Since Browder thoroughly researched and packaged stories for the media for companies in which Hermitage owned shares, the authors had an outside component in news coverage with which to test the causality effect of the media.

"From a scientific point of view, the issue is how to distinguish between a simple correlation and causation," notes Zingales. "It's no surprise that big news is reported in the newspaper, and major action at the corporate level may follow. But merely documenting a correlation between news and outcomes proves nothing. One goal of this paper is to show causality through natural experiment."

Whether or not Hermitage owned a stake in a company proved to be an effective instrument for showing causality, since Hermitage had an incentive to act only against companies in which it owned shares. Therefore, companies engaged in significant corporate governance abuses were more likely to be exposed in the international press if they had the Hermitage Fund as a shareholder.

A Case Study

To understand the relation between the Hermitage Fund and press coverage, and between press coverage and outcomes, the authors analyzed two companies: Sidanco, a little- known but important holding company in the oil and gas sector, and MGTS, the telecommunications company for Moscow. In 1998, both companies a ttempted to dilute investors by issuing shares to parties linked to insiders at well below the prevailing market price.

While both companies committed similar corporate governance abuses, they differed in their investors. Hermitage had a significant stake in Sidanco and no stake in MGTS. Consequently, Hermitage became actively involved in trying to generate media coverage about Sidanco.

The authors found that Sidanco's actions were reported in 23 news articles, 14 of which appeared in credible international publications, including the Financial Times, various editions of The Wall Street Journal, and the Economist. In contrast, MGTS had only three articles mentioning its corporate governance violation in the credible international press, all of which were in the Financial Times. In the case of Sidanco, the dilution was reversed, while in MGTS the dilutive share issue was quickly approved.

Aware that this case is only one example and has its limitations, the authors sought support for their argument by collecting data on all major corporate governance abuses that occurred in Russia from 1999 to 2002 and studied the determinants of the probability these abuses were readdressed. They found that the probability for a reversal was significantly affected by the exposure of the news in the Anglo-American press, even after controlling for other potential determinants of the outcome (such as the degree of foreign ownership and the involvement of international organizations such as the European Bank of Reconstruction and Development).

Between the two most credible foreign newspapers, coverage of corporate governance abuses in The Wall Street Journal typically has more impact than coverage in the Financial Times. This could be due to either higher credibility of the former or different audiences. To separate the effect of audience from that of credibility, the authors use the Russianlanguage publication Vedomosti for comparison. Since Vedomosti is a joint venture of The Wall Street Journal and the Financial Times, its credibility is likely to be similar to the two larger papers. However, as a Russian-language paper, it only reaches a Russian audience. The authors found that coverage in Vedomosti had no significant effect, suggesting that all leverage on reputation comes from exposure in the Anglo- American community.

Most importantly, the authors were able to show that it was a component of foreign press coverage driven by the Hermitage Fund action that drove the results, suggesting the link between coverage and outcome is more than a mere correlation.

The Media Makes a Difference

The authors also studied the multiple channels through which press coverage succeeds in changing outcomes. For example, there had been widespread concerns about deals between related parties at Gazprom, Russia's largest gas company. These concerns only became a focus of attention (and were addressed seriously) when Hermitage made Gazprom the focus of a media campaign. One method Hermitage uses to generate news is to conduct research and then present and document the information to a select group of reporters.

When Gazprom's violations were exposed, the first reaction was taken by government officials on Gazprom's board. These officials felt compelled to side with minority shareholders, and passed a motion requiring board approval for any subsequent dilutions. Publication of this news also helped coordinate the actions of small institutional shareholders who demanded an audit of these transactions. These stories inflamed the investment community and helped Hermitage convince other investors to sign their proxies to obtain the necessary 10 percent required to demand an independent audit of the transactions in question, along with other transactions, in December 2000. The revelations also provided additional motivation for the government to change Gazprom's CEO in May 2001.

As this example illustrates, the mechanism by which media coverage affects outcomes is complex, making it difficult to identify a single force. For example, all factors playing a role in the Gazprom case seem to have gained strength as a result of news coverage, even though the final difference was made by a political intervention.

In looking at the effect of the press, the authors find that in roughly half the cases, media pressure leads a regulator or a politician to intervene; in the other half, it is the company itself that relents after realizing the reputational costs of continuing the battle.

Zingales, Dyck, and Volchkova conclude that media coverage has an effect primarily by increasing the reputation cost of misbehavior vis-a-vis a relevant audience,in this case Anglo- American investors. The success of this strategy is highly dependent on the importance those concerned attribute to their reputation within a certain constituency. In this sense, Russia from 1998 to 2002 may be a special situation if Russian companies felt pressure to restore their international credibility after the Russian default on public debt.

Nevertheless, these results could have very important policy consequences. By interacting with developing countries, the United States can exert a positive influence not only on public governance, but also on corporate governance. Since politicians and businessmen are eager to "look good" in Anglo-American public opinion, they can be steered toward improving their governance standards.