DuPont beat Trian's proxy fight in May 2015 — and replaced its CEO five months later

On May 13, 2015, DuPont's board did what boards under siege dream about: it counted the ballots and found it had won. All twelve of its director nominees were reelected, and the competing slate put up by Nelson Peltz's Trian Fund Management lost.¹ Ellen Kullman, DuPont's chair and chief executive, had faced down one of the most feared activist investors on Wall Street and kept her board intact. It looked like a clean vindication of her strategy. It was nothing of the sort.

The story that hardened afterward is that DuPont beat the activist decisively and management's plan carried the day. Almost every part of that is misleading. The vote was won by a whisker, tipped by a handful of index funds. The people paid to advise shareholders had actually sided with Trian. And within five months the CEO who fronted the victory was gone and DuPont was moving toward exactly the kind of breakup Trian had spent eighteen months demanding.⁵ The 'win' wasn't a verdict. It was a stay of execution.

The narrowest possible landslide: flip one of three index funds and the whole result inverts

Start with the margin, because the margin is the whole point. Trian wasn't storming the gates. It held only about 2.7% of DuPont's shares and sought just four of twelve board seats — a minority-representation bid against a company worth roughly $68 billion, not a takeover.⁴⁵ And yet, holding around 24.6 million shares, it came within an eyelash of getting them. After certified results were released, Trian stated plainly that the vote 'was extremely close,' and that if any one of three large index funds had voted all its shares for Trian's nominees, Trian would have won board representation.³ Even as DuPont was projected to win on the day of the meeting, sources described the tally as still very close.⁷ A win this fragile isn't a mandate. It's a coin that landed on its edge and rolled the right way.

The establishment wasn't on management's side: the referees called it for the challenger; three big passive holders overruled them

Here is the part that scrambles the usual script. You'd expect the institutional establishment to close ranks behind an incumbent board against a raider. It didn't. The two most influential proxy advisory firms, ISS and Glass Lewis, both recommended shareholders back Trian's slate.⁴ On the merits, as read by the people whose job is to read the merits, the activist had the better argument. DuPont survived not because it out-argued Trian but because its three largest holders — Vanguard, BlackRock and State Street — voted with management despite that advice.⁴ Those are index funds, structural owners of nearly everything, and their loyalty on the day was the difference between keeping a board and losing one. DuPont didn't win the debate. It won the rolodex.

Five months to the same conclusion: the CEO exits, and a Tyco breakup veteran walks in the door

A proxy win is supposed to buy a management team time and cover. It bought Kullman about five months. On October 5, 2015 — roughly 4.7 to 5.1 months after the vote — DuPont announced she would retire as chair and CEO, effective October 16.² The same release cut the company's full-year 2015 operating earnings guidance, and in the interval since the vote earnings had continued to deteriorate.²⁶ Stepping in as interim chief was board member Edward Breen — a man whose résumé included overseeing the breakup of Tyco International.⁶ Sit with that casting choice. The board had just spent enormous energy defeating an activist who wanted DuPont carved up. Then it handed the company to an executive famous for carving a conglomerate up. The activist lost the vote and won the argument, on a delay.

Was the exit really the activist's doing?: the honest counter is that no press release ever admits defeat, and the timing is murkier than the myth

The fair objection is that this is too clean a morality tale. DuPont called Kullman's departure a retirement, and contemporaneous coverage noted that observers were divided over whether the step-down was voluntary or forced.⁶ The exit coincided with a same-day earnings guidance cut, which offers a simpler explanation than 'Trian won by other means': the numbers were bad, and bad numbers cost CEOs their jobs whether or not an activist is circling.²⁶ All true — and it doesn't rescue the tidy 'DuPont won' narrative, it just relocates the loss. Whether the board fired her or accepted a graceful resignation, the outcome converged on Trian's thesis: performance was deteriorating, the incumbent leadership left, and a restructuring specialist took the chair. A proxy win that leaves management installed and unchanged is a victory. One that ends with the CEO out and a breakup artist in is a negotiated surrender dressed as a triumph.

DuPont's board held the line and kept its twelve seats. It just couldn't keep the thing the seats were meant to protect. Within five months the CEO who embodied the winning strategy was gone, the earnings that were supposed to prove her right had gotten worse, and the company had reached for a leader whose specialty was the very outcome the whole battle was fought to prevent. Trian lost the vote and, on a five-month delay, got most of what it came for. The lesson isn't that activists always win. It's that a victory measured in retained board seats can hollow out completely while the trophy is still on the shelf — and the surest way to read a proxy 'win' is to ignore the count and watch what the winner does the moment the meeting adjourns.