Courtesy of Sanjiv Sabherwal and Mohammad Uddin
There are many studies on insider trading by corporate CEOs. There are also some studies that examine whether celebrity, or star CEOs, create more value for their firms and investors. However, no studies, until now, have attempted to identify whether there is a difference in the information content of trades of star and other CEOs. In a new paper, we examine whether the star status of a CEO affects the informativeness and predictability of his trades.
Premise
Prominent CEOs in the United States enjoy celebrity status. Star CEOs attract more media and public attention than their less prominent counterparts. We argue that although both star and non-star CEOs possess useful private information about their firms, star CEOs are less likely to exploit that information in their trades due to the greater attention and scrutiny they attract from the media and regulatory bodies. As a result, trades of star CEOs are more likely to be routine trades that are driven by liquidity needs or excess liquidity. That is, their trades are less likely to be informative and to have the power to predict future returns. In contrast, non-star CEOs receive less coverage in the media, attract less attention from investors, and face less scrutiny from regulatory bodies. Therefore, their trades are more likely to contain useful private information and the power to predict returns.
Empirical Approach
We compile a sample of CEOs of S&P 1500 (including S&P 500, S&P MidCap 400, and S&P SmallCap 600) companies from 2004 to 2011. We classify those CEOs into two groups: stars and non-stars. We use three different measures for the classification of CEOs: award-based, web-based, and media-based. In the award-based measure, we identify CEOs as stars based upon winning prestigious contests organized by highly regarded international business publications. We construct a matching group of non-star CEOs, which consists of non-winning CEOs with characteristics closest to those of award winners. The other two measures identify CEOs in the top and bottom quintile based on their Google search volume index (SVI), and the quantity of media coverage they receive in sources included in the Factiva database.
Key Findings
We regress future six-month cumulative abnormal returns (CARs) for a firm’s stock on indicators of star and non-star CEO trades. We find that star buys and star sells are not good predictors of future abnormal returns, whereas non-star buys and non-star sells are strong predictors of future abnormal returns. These findings support the argument that insider trades of a non-star CEO possess useful information about a firm’s future prospects.
We also disentangle trades of both star and non-star CEOs into two distinctive categories based on whether a trade is consistent with a seasonal pattern of trades by that insider. If a trade is consistent with a seasonal pattern, then that trade is considered as routine and it does not possess private information. If a trade is not consistent, then that trade is most likely motivated by private information and is considered an opportunistic trade. We find that star CEOs place more routine trades whereas non-star CEOs place more opportunistic trades. We also find that most of the predictive power of non-star CEO trades arises from opportunistic trades, and not from routine trades.
We also examine whether non-star CEO trades possess future earnings-related information of firms by regressing CARs over 3- and 5-day windows surrounding the first two quarterly earnings announcements that take place after an insider’s trade on indicators of star versus non-star CEO trades. We find that pre-earnings announcement trades of non-star CEOs predict CARs associated with quarterly earnings announcements, whereas trades of star CEOs do not. We also find that opportunistic insider purchases of a non-star CEO contain information about the firm’s future earnings while routine purchases do not.
Scrutiny Argument vs. Inferior Information Argument
An alternative explanation to our argument that increased scrutiny for stars discourages them from placing informed trades is that star CEOs have inferior information about their firms because their stardom distracts them from managing the firm. We investigate whether the increased scrutiny or the inferior information argument explains our findings by classifying firms into lightly and highly regulated. With highly regulated firms under greater scrutiny be regulators, the increased scrutiny argument suggests that not only star, but non-star CEOs of such firms are also less likely to place information-based trades. In contrast, the inferior information argument suggests that even for highly regulated firms, only the trades of star CEOs are likely to be uninformative.
We regress future CARs on indicators of monthly star and non-star CEO trades separately for each group. We find that for lightly regulated firms, the coefficient of non-star buy (sell) is positive (negative) and significant, whereas the coefficients of star buy and star sell are insignificant. For highly regulated firms, the coefficients of buys and sells are insignificant for both non-star and star CEOs. These results are consistent with the increased scrutiny of star CEOs argument and not with the inferior information of star CEOs argument.
Additional Analyses
We perform four additional analyses. First, we confirm that our main results are robust to estimating abnormal performance using calendar-time portfolios instead of CARs. Second, we test the argument that trades of star CEOs are less likely to be informative by focusing only on award-winning CEOs and examining if the insider trades of those CEOs in the same firm become less informative after they achieve star status by winning an award. We find that there is a significant decline in the positive abnormal CARs associated with insider purchases by CEOs after they receive an award. Third, we examine the trading responses of institutional investors to trades by star and non-star CEOs. We find that institutional investors act as liquidity providers for trades of non-star CEOs. Finally, while we have controlled for attributes such as firm size and CEO compensation in all our analyses, we perform our investigation for a subsample of only large (S&P500) firms. We find no qualitative change in our results.
Conclusions
Our study contributes to the existing literature on insider trading by providing a new framework to disentangle insider trades by CEOs. Specifically, we classify CEO trades into two categories: trades by star and non-star CEOs; and empirically show that non-star CEO trades have the power to predict returns whereas trades of star CEOs do not have such power. Ours is the first paper to examine the impact of insiders’ star or celebrity status on equity returns. We use three different approaches for identifying star and non-star CEOs, which shows the robustness of our empirical findings. This study has useful practical implications for naïve investors as it shows that they should carefully interpret trades by top corporate executives before mimicking them. The findings also help regulators, investors, and other active market participants identify which insiders and insider trades provide insights into firm value.
