In 2016, Lazard counted some 149 campaigns initiated by activist shareholders to obtain board representation and also found that these campaigns netted a record 131 board seats in that year.[1] But, 95 percent of these seats were the result of settlement negotiations, not proxy contests.[2] (Indeed, in the relatively few cases that did go to a shareholder vote, management generally won[3]). This suggests that corporate managements have become highly risk averse and will do much to appease activist funds. They have good reason to do so, because other studies have shown that the majority of CEOs depart within two years of an activist going on the board.[4] A defeat in a hostile proxy fight (even if only for one or two directors) means that the CEO has been humiliated and probably is on his way out. Hence, appeasement is the new norm.

This pattern of the vast majority of activist challenges being settled through private negotiations makes this settlement process academically interesting. What gets negotiated? And with what outcomes? Working with two Columbia colleagues, we have surveyed a data set consisting of 475 such settlement agreements, extending up to 2015.[5] What stands out from this study that would interest practitioners?

The most striking finding is that the appointment of activist directors quickly leads to an increase in informed trading. Once a fund-nominated director or directors go on the board, a pattern of information leakage begins, with the result that the corporation's stock price begins to anticipate subsequent public disclosures. Whether the news is good or bad, the market seems to have largely already learned it. Moreover, this pattern is not only inconsistent with the prior stock movement in the target corporation's stock price, but also with that of control groups of similar firms (which differ primarily in that they do not have an activist nominee on their boards).

This ability of the market to anticipate subsequent disclosures is most plausibly explained as the consequence of informed trading, which seemingly begins at the point at which the activist nominees join the board (or are given “observer” status and permitted to attend board meetings). Reinforcing this hypothesis is another finding that our study demonstrates: the pattern of information leakage is much more pronounced when the hedge fund's nominees include a hedge fund employee. Put simply, hedge fund employees seem to leak much more than independent experts nominated by activists to the board (or at least the presence of an in-house employee increases the amount of information leakage and market anticipation of subsequent disclosures). Intuitively, this is plausible because hedge fund employees live in a world of traders, whereas the independent directors nominated by activist funds (typically, former CEOs or CFOs from the same industry) do not.

Another of the clearest findings in our study involves the bid/ask spread. Bid/ask spreads widen when activist-nominated directors are appointed to the board (in comparison both to the pre-activist spread and to the spreads for the firms in our control groups). But, once again, the pattern is far more pronounced in the case of activist directors who are hedge fund employees. Spreads for the treatment and control groups are similar and parallel prior to the appointment of an activist director, but then diverge sharply after that appointment (with the spread widening only in the case of the board with an activist director). Widening spreads imply that the market expects informed trading; in effect, market traders are protecting themselves by expanding the spread defensively. We have heard the interpretation from proponents of hedge fund activism that this widening of the spread may be because the target firm faces greater volatility because its business model is to be changed. Whether this explanation works as a matter of finance theory is debatable, but it cannot explain a key aspect of our data: the widening of the spread is greater when the fund's nominees include one of its own employees. Seemingly, the market expects more informed trading in such a case and widens the spread.

One must be cautious here and not jump to the conclusion that the hedge fund employee/director is engaging in unlawful insider trading. That inference of illegality cannot be drawn from our data. Still, whether the spreads widen because of unlawful behavior or simply because of negligent handling of confidential information, the increase in spread width is still an agency cost borne by the other shareholders. Moreover, tipping data can be a useful strategy for the lead activist to employ, even if it is careful not to trade, itself.

Not all hedge funds or their nominees behave alike. Clearly, there is much variation. What factors correlate with high information leakage and wider spreads? Policies appear to vary considerably among target corporations with respect to information confidentiality. In our data set of 475 private settlement agreements, only a minority of these agreements address the confidentiality of the information acquired by fund-nominated directors at board meetings, but a high correlation is evident between information leakage and whether the settlement agreement addresses the handling of confidential information by the new directors. In some of these private settlement agreements (but very few), there are strict controls on information sharing by the nominee directors (i.e., no sharing without the target company's consent). Still, in the majority, there is nothing similar to such a rule. Often, all that is stated is that the parties to the agreement recognize that federal law restricts the use of material, nonpublic information for purposes of trading on securities. Another common pattern is to provide that confidential information can only be shared by the nominee or the hedge fund if the recipient is first informed that the information is confidential.[6] We find that both information leakage and the widening of bid/ask spreads are concentrated in those settlements that lack an explicit rule on information-sharing (either in the settlement agreement or a separate confidentiality agreement). A possible inference here is that these cases with no or only weak controls are those in which target management lacks the leverage to insist upon a stricter rule on information sharing by fund-nominated directors.

Beyond this initial point that shareholders pay a hidden price for activist-nominated directors, there is a larger more macro-economic conclusion that logically follows from our data. The ability to engage in informed trading based on access to material, non-public information from fund-nominated directors represents a significant subsidy to activist hedge funds. To be sure, we do not know whether this benefit accrues to the activist hedge fund, itself, to its employees, or to allies in its “wolf-pack” network. But, someone benefits. The key point here is that subsidies, by definition, increase the level and volume of the subsidized activity.

Thus, it seems a fair prediction there will be more hedge fund engagements under rules that permit easy information sharing than under a system that restricted the ability of activists to profit from access to material, non-public information. To see this point, consider this example: an activist fund proposes a change in the business model of a target firm. Its proposals are highly debatable, and many other investors are not convinced that the proposals will enhance shareholder value. But they also know that if they join the “wolf-pack” supporting the lead activist, they will have access to material, non-public information about the target firm during a very volatile period in which its stock price is likely to fluctuate sharply. Access to inside information promises to be very profitable during such period. If so, other hedge funds may want to stay in the “wolf-pack” and remain at least loosely affiliated with the lead activist, even if they doubt the wisdom of its proposed changes in the target's business model. Not only will there be more activist engagements under rules that subsidize activism by tolerating easy information sharing, but marginal campaigns that would not otherwise be supported will gain support for this reason.

This premise leads to a generalization: access to material, non-public information may be the “social glue” that holds together the “wolf-pack” (which otherwise would be a very unstable and short-term entity). Empirically, only the filing of a Schedule 13D that announces the lead activist's presence and plans predictably elicits a significant abnormal gain (usually 6 percent to 7 percent).[7] But, if one knows what will be disclosed publicly in the near future, “wolf-pack” members can make lucrative profits, even if the stock moves erratically and with no ultimate net change.

So, what should be done? In my article posted on SSRN,[8] I make some suggestions, including expanding the definition of “group” under the Williams Act to cause the sharing of information to trigger an earlier Schedule 13D filing. But one further step can be predicted:

By analyzing our data, it should be possible to identify which activist firms are more closely associated with information leakage and widened bid/ask spreads—and, by implication, with informed trading. Such finger-pointing is not our goal, and we are seeking to avoid unnecessary controversy. But, over time, target managements may begin to employ this tactic, and the data is easy to assemble on an activist-by-activist basis. Also, at some future point, Assistant United States Attorneys may want to know which activist funds are most closely associated with information leakage and, based at least in part on that clue, obtain wire taps. If they are looking to determine who is most likely to be engaged in unlawful trading, we have the roadmap.

Of course, criminal enforcement may not be necessary. A more benign scenario begins from the premise that “sunlight is the best disinfectant.” Once it becomes understood that the appointment of hedge fund employees to boards is associated with information leakage and widened spreads, activists may be embarrassed or shamed into using only independent directors (and not employees) as their agents. In-house nominees are already the minority of fund-nominated directors. Also, the SEC could begin to inquire into why so few funds and target corporations place any real limitations on information sharing between fund-nominated directors and others in the “wolf-pack.”

Of course, we may be too optimistic in our expectation that the number of fund employees nominated as directors will decline. One category of fund-nominated employee/director is the fund's own founder, and these persons often have very large egos. Or, founders and senior fund employees may use confidential information as a means of holding together the “wolf-pack” that they assembled.

Predictably, there will be counterattacks on our findings. Diehard academic proponents of activism will argue that widened spreads are a small price to pay for the enormous gains they believe are realized from hedge fund activism. From this perspective, Ivan Boesky and Raj Rajaratnam (who both obtained material information from friendly directors) can be seen as the martyred heroes of the efficient market. Such “the-end-justifies-the-means” arguments should embarrass legitimate activists. Activism, itself, can survive without the subsidy of a steady supply of material, nonpublic information. But perhaps not all contemporary activists can make that leap.

Endnotes:

[1] See Lazard, “Review of Shareholder Activism in 2016” (February 2017) at p.1.

[2] Id. In contrast, in 2014, 34 percent of board seats won by activists were won through actual proxy elections. Thus, this trend toward private settlement is increasing.

[3] In 2016, Lazard found that activists had a “win rate” in actual proxy contests of only 38 percent. Id.

[4] A study by FTI Consulting places the departure rate for CEOs after activist nominees are added to the board at 34.1 percent over one year and 55.1 percent over two years. See Sonali Bazak and Beth Jinks “Activist Directors Double Chance of CEO Exits, Study Shows,” Bloomberg News, Oct. 12, 2016.

[5] I describe this study, which is still ongoing, in an article recently posted on SSRN. See John C. Coffee Jr., “The Agency Costs of Activism: Information Leakage, Thwarted Majorities, and the Public Morality” (Oct. 25, 2017). The results of this study were first reported in Robert Bishop, Robert J. Jackson Jr., and Joshua R. Mitts, “Activist Directors and Information Leakage” (2017). Publication of this study has been slowed by Professor Jackson's nomination by the President to the Securities and Exchange Commission, but we expect to release it early in 2018.

[6] A cynic might say that this gives the hedge fund, as tipper, the best of both worlds, as it can claim that it only tipped those who promised confidentiality (which is permissible under Regulation FD) and was “shocked” to learn that they had traded.

[7] A number of studies have found that there is an abnormal return of 6 percent to 7 percent on the filing of a Schedule 13D by an activist, but no further increase in the stock price unless and until a merger transaction or major asset disposition is announced. For an overview of these studies, see John C. Coffee and Darius Palia, “The Wolf at the Door: The Impact of Hedge Funds on Corporate Governance,” 41 J. of Corporation Law 545 (2016). But informed traders can profit on volatile price swings even if there is no net change over the long-run.

[8] See Coffee, supra note 5.

John C. Coffee Jr. is the Adolf A. Berle Professor of Law and Director of the Center on Corporate Governance at Columbia Law School.