媒体新闻报道的声誉溢出效应

本文聚焦于公司违规曝光这一特定事件,使用三重差分的计量分析方法,探讨了媒体的事前新闻报道是否会影响其在投资者群体中的声誉,即是否存在声誉溢出效应。结果表明媒体对违规公司的事前新闻报道越正面(负面),其日后针对非涉案公司发布的新闻报道的市场反应就越弱(强),即存在双向声誉溢出效应。然而该效应存在不对称性,表现为负面报道带来的正向溢出效应明显强于正面报道带来的负向溢出效应。进一步研究发现,声誉溢出效应取决于投资者对于媒体声誉的主观感知以及对于违规事件的信息解读能力。当媒体的既有声誉水平较低、违规案件较为严重,或非违规公司机构投资者比例较高时,媒体的声誉溢出效应更加明显。本文论证了媒体客观公正性对于声誉积累的重要性,为媒体是否应当以及如何维护在投资者群体中的声誉形象提供了经验证据。本文对投资者如何判断和使用媒体发布的信息,进而做出正确投资决策也具有重要意义。

The Reputational Spillover Effect of Media News Reports

The media play an important role as an information intermediary and monitor in the capital market. The media can earn a good reputation if their news reports reflect the truth, and the journalists behind the reports can also obtain better career prospects. However, false reports may damage the media's reputation and even hinder their normal operation. However, due to economic interests or a lack of professionalism, morals, and legal awareness, bias in media reporting is quite common.
In this paper, we examine whether a media outlet's news reports prior to a firm's fraud exposure affect the outlet's reputation among investors (i.e., whether there is a reputational spillover effect). More specifically, we examine whether investors' perceptions of the outlet's reputation change immediately after a fraud is exposed and how they respond to news stories by the same outlet about other no-fraud firms. On one hand, if the media outlet releases positive reports about the fraudulent firm ex ante, rational investors should question its professionalism and/or independence, making its reports on other no-fraud firms less likely to be trusted in the near future (negative reputational spillover effect). On the other hand, the media outlet's negative reports may be regarded as a positive signal by investors, which can produce a positive reputational spillover effect.
We obtain firm fraud information from the China Stock Market and Accounting Research (CSMAR) database and set the lockup period as the window -14,-7] before fraud exposure. First, we identify the treatment media outlet that reports on the fraudulent firm during the lockup period. Next, we identify other no-fraud firms reported on by the same outlet in both windows -7,0] and (0,7) around the exposure date. Then, for each no-fraud firm, we find all other media outlets that release reports in both windows but have no reports on the fraudulent firm and use them as the control group. Finally, all news reports on no-fraud firms in the window -7,7] by the treatment media outlet and the control media outlets are identified and included in our sample.
We then use a difference-in-difference-in-differences (DDD) method for estimation. The results show that a more positive (negative) tone in the media's pre-exposure news report leads investors to react less (more) strongly to its news reports on no-fraud firms soon afterwards, indicating a two-way reputational spillover effect. However, the effect is asymmetrical in that the positive spillover is significantly stronger than the negative spillover. Further results show that the effect depends on investors' subjective perception of the media outlet's prior reputation and its ability to interpret the relevant information. The effect is more pronounced in cases with lower media reputation prior to the fraud, more severe fraud, or higher shareholding percentage of institutional investors in the no-fraud firm.
This paper provides empirical evidence on whether and how the media should maintain its reputation among investors. We demonstrate the importance of media objectivity to reputation accumulation and point out specific influencing factors that determine spillover effects. In addition, we find that the information content of media news reports is highly related to the outlet's most recent reporting quality, indicating that the information content of its news reports can change dynamically. Therefore, we also provide important guidance on how to use media news reports for making better investment decisions.
We acknowledge that the quality of media reports is also closely related to the individual characteristics of journalists, and these characteristics may also matter for the spillover effect discussed in this paper. In addition, the media's reporting behavior may directly affect the careers of journalists and even the reputations of peer outlets. What's more, the exposure of false reporting behaviors may also exert a long-term deterrent effect on media reporting quality. These issues can be explored in the future to deepen our understanding of media reputational concerns.