Print
E-MailGeorgy Chabakauri is Associate Professor of Finance at the London School of Economics; Vyacheslav Fos is Associate Professor of Finance at Boston College Carroll School of Management; and Wei Jiang is Arthur F. Burns Professor of Free and Competitive Enterprise in the Finance Division at Columbia Business School. This post is based on their recent paper. Related research from the Program on Corporate Governance includes Insider Trading Via the Corporation by Jesse Fried (discussed on the Forum here).
All major securities markets have developed laws, rules, and systems that regulate trades by insiders and their affiliates who have privileged access to material nonpublic information, and criminalize insider trades that are based on, or misappropriate, such information. While the theory and practice of insider trading law and regulation have evolved over time, the boundary of insider trading remains blurry and becomes more so with new developments of the market. In this study, we explore the possibility of insider trading on non-insider information in a setting where an insider (i.e., a CEO) makes trading decisions on their firm’s stock based on assessed possibilities of trading by activist shareholders. Though the insider does not have direct information about the arrival of the “barbarians at the gate,” privileged information about their own firm’s fundamentals helps the insider to filtrate public information and eventually trade on public information with a distinct advantage.
In recent decades, real-time trades/orders have essentially become public information. Modern “tape readers” specialize in looking at electronic order and trade books to hypothesize the motives underlying any unusual trading patterns and to analyze where a stock price may be headed. Compared to other forms of informed trading by outsiders (such as those betting on takeover prospects or earnings surprises), activists are better positioned to camouflage their trades due to their ability to spread the trades to time market liquidity. This is because the deadline of the private information, in the form of a Schedule 13D, is largely self-imposed. However, given the concentration of trades in the relative short period of time (usually 2—3 months), and a hard deadline of ten calendar days after the 5% crossing-date (the disclosure triggering event), it becomes increasingly difficult for activists to hide their trades in market liquidity as they approach Schedule 13D filing. Now the question becomes: Are insiders better equipped to detect activist trading than outside investors and the market makers prior to Schedule 13D filing?
We hypothesize and solve in a model to conclude that the answer is a “yes” based on both incentives and capabilities. First, insiders have stronger incentives than general investors to get informed of activist plans. The information about an upcoming Schedule 13D filing is valuable to general investors due to the significantly positive average announcement return. Because job turnover increases and compensation drops for senior executives at the targeted firms, hedge fund activism often meets defense from the management. Therefore, the information is of greater value to insiders because being prepared is a premise to an effective defense. Companies that recognize their vulnerability from activist targeting are incentivized to detect activist movement ahead of the public. Financial advisories and other intermediaries also emerge to help firms deploy defensive strategies ahead of the barbarians’ arrival at the gate.
Second, insiders enjoy an information advantage in an indirect way. Conditional on both insiders and outsiders observing the same order flows and trades, insiders have a more refined information filtration to isolate trades potentially generated by activist interests from those motivated by leakage of or speculations on firm fundamentals, such as earnings or sales growth. Suppose a piece of public information is the union of two disjoint components, where composite set is public information but not its composition. If insiders observe one of the two components (firm fundamental hence trades motivated by it), then combining with the public information of total trading flows allow the insiders to filter out potential activist interest. More importantly, if insiders trade based on information inferred from public information, they do not run afoul of the insider trading rule. In fact, if they buy to counter activist, the insiders do so precisely because of a lack of private positive information about firm fundamentals.
Empirically, we show that corporate insiders engage in share purchase, at frequencies significantly above normal, on the same day when activists trade, and three days afterwards (corresponding to the T + 3 settlement). Such a coincidence is intriguing given that activist trading is not observable in real time. Further tests strengthen our hypothesis. First, we document a significant relationship between activist trading and insider trading during the 60-day window prior to Schedule 13D filings. We find that the likelihood of insider buying (selling) is 13 (78) basis points higher (lower), relative to days outside the time window. The difference, statistically significant, amounts to 15% (37%) of the normal pace of insider buying (selling). The combination of more buying and less selling leave more shares, and hence voting and control power, in the hand of the management at the dawn of an activist campaign.
Second, we rule out the alternative hypothesis that the concurrence of activist and insider trading could be due to activists piggybacking on insider buying as the latter might be motivated by private positive information about the firm. Under the T+3 settlement rule prevailing during most of our sample period (till 2017), a transaction will finish the ownership record change three days after the trade. If companies or investors actively monitor ownership changes—with the help of the intermediaries such as proxy solicitors—then they might get informed three days after the trade. If insiders buy in response to activist trades, we should observe a significant response on T+3. In the reverse direction, activists could potentially be informed of the trades place by insiders after just two days given that insider trading requires disclosure within 48 hours. We find that insider buy is significantly (at the 5% level) higher than usual on T+3 days relative to activist trading; but there is no significant correlation between activist trading and insider trading two days (or any days) prior. Therefore, results attribute the “source” trades to be placed by the activists and insiders trade in response.
Finally, we empirically test the mechanism that insiders are better positioned to isolate unusual trade flows from activist interests from those motivated by speculation on firm fundamentals. We find that the abnormal insider buying on the days with activist trading is solely driven by the subsample without positive earning surprises. This test affirmatively differentiates insider trading in our set-up from the conventional insider trading based on private information about firm fundamentals, and raises new questions for the legal implications of such “insider trading based on non-insider information.”
The complete paper is available for download here.
