Leadership1 publisher3 min readPublished
Fragmented proxy voting is making shareholder intent harder to predict
Wachtell Lipton counts six US activist campaigns reaching a vote through August 2026, down from nine a year earlier, and says settlements now often land before the activist has agitated in public.
The Board Room · Leadership desk
What happened
- Activism ran at elevated levels globally through the first half of 2026, and about 40% of campaigns carried an M&A-related thesis, according to a Wachtell Lipton memorandum.
- The memo attributes a more fragmented and less predictable proxy vote to the Big Three splitting voting teams, wider pass-through voting, and large investors adding AI to their voting models.
- Activists are opening private engagement earlier, sometimes in the summer months after an annual meeting, nearly a year before they could win seats in a contested election.
Compiled by The Board RoomSomething wrong?How this is made
Why it matters
- constraint A board weighing whether to fight or settle used to lean on a canvass of its largest holders. With voting teams split inside the same manager and ballots passed through to underlying owners, that input is thinner, and the call rests more on the board's own conviction in its strategy.
- decision A summer approach makes a board set access terms a year before any seat is winnable, and what it grants in July becomes the baseline the activist expects through the next annual meeting.
- exposure Boards in the middle of a CEO search are the reachable ones: an activist arriving then can push for a transition announcement before a successor exists, or for an interim in the chair.
- precedent Settlements agreed before any filing, and sometimes with no written agreement at all, leave other shareholders learning of board changes after the terms are set. The proxy record shows nothing of what was traded.
The count that used to come before a proxy fight is what is thinning out. Elina Tetelbaum and Loren Braswell name three causes: the Big Three institutional investors splitting their voting teams, pass-through voting becoming more prevalent, and large investors incorporating AI into their voting models [6]. Each breaks a different assumption a board once worked from. Splitting a stewardship team splits the answer a company gets when it asks how a manager is leaning, and pass-through voting moves the ballot to underlying holders the company has no relationship with. The memo calls the settlements reached before public agitation "a significant number" and pass-through voting "more prevalent," and it gives no number for either [22].
Six contests reached a vote in the United States at companies above $250 million of market value as of August, and activists took seats in one [8]. Six is what is left after the settlements, and Tetelbaum and Braswell put the vast majority of campaigns in the settlement column, often resolved before preliminary proxy materials are filed [7]. Wachtell attaches a condition to the record as well, writing that shareholders back an engaged board especially where the activist cannot recruit quality candidates or hold a clear message [17]. One win in six contests is a 17 percent hit rate on a sample of six [19]. The number of contests reaching a vote fell by three year on year, a third fewer than 2025 [18].
Fall used to open the season, timed to the conferences where funds name their targets [10]. First contact can now come in the months just after an annual meeting, nearly a year before the activist could win a seat in a contested election [9]. That changes the decision on the table when the call comes. The question is how much access to grant, and for how long, at a point when no seat is at stake for months. The authors wrote that early engagement is useful for understanding the activist's perspective and showing responsiveness, and that it has to be managed so it does not create unreasonable expectations about access over the course of a year [12]. They also wrote that the nature of an initial contact offers clues about the seriousness of the thesis, since the more prolific funds screen several targets at once [11].
The biggest cost the memo identifies is in CEO succession. An activist arriving mid-search can create pressure to announce a transition before a successor has been identified, to install an interim CEO, or to accelerate a search the board had been running on another timetable [13]. "There is no one-size-fits-all model for handling such pressure," the authors wrote [14]. What they say matters is that the board can demonstrate to investors that the decision was made thoughtfully and deliberately [15].
The two issues move on different timescales. The timing question is live this quarter: a summer approach arrives while a board is still months from its next meeting [9]. The change in how shareholders vote is the slower one, and it cuts against the ballot-box record, since a company choosing to fight is relying on a register it can read. The ten trends this pair published a year ago, for 2026, already listed less visibility into shareholder views [2].
What to watch
- Whether the count of US contests reaching a vote recovers after August 2026 or settles below the nine seen in 2025.
- Any disclosure from large managers on how AI enters their voting models; the memo mentions the practice but does not size it.
- Whether settlements struck without a formal written agreement start surfacing in disputes over what shareholders were told.