Not mad, just disappointed.

Photographer: Lisa Kyle/Bloomberg News.

DuPont Shareholders Don't Want Trian's Advice

Matt Levine is a Bloomberg View columnist. He was an editor of Dealbreaker, an investment banker at Goldman Sachs, a mergers and acquisitions lawyer at Wachtell, Lipton, Rosen & Katz and a clerk for the U.S. Court of Appeals for the Third Circuit.
Read More.
a | A

Yesterday DuPont Co. won its proxy fight against Nelson Peltz's Trian Fund Management, after a genuinely suspenseful campaign in which it really looked like DuPont might lose. People are already looking at this fight as a sort of Siege of Vienna for activism: It's a high point of activist aggression, but it might also mark the limits of activist success. 

The story of DuPont and Trian, as told by for instance Joe Nocera or Stephen Gandel, is that a basically well-run company with good performance and shareholder-friendly policies was targeted by a basically thoughtful, credible, long-term-oriented activist investor. There were no real villains; neither side's hate was pure. The two sides had behind-the-scenes discussions that led to mutually agreeable corporate changes, but Trian wanted board seats and DuPont didn't want to give them, so the discussions degenerated into a proxy fight. Proxy fights being proxy fights, this one was nasty and contentious, but in sort of an unfocused way: Trian's proxy statement did not call for DuPont to replace chief executive officer Ellen Kullman; nor was it advocating for a breakup or sale or anything crisp and drastic. Trian just wanted to move its discussions into the boardroom; DuPont wanted to keep Trian at arms' length.

So Trian's presentation on DuPont had the boring title "A Referendum on Performance and Accountability," and it argued that "a vote for Trian's nominees is a vote for four highly qualified individuals who will seek to work collaboratively with the Board to: Assess the corporate structure and determine whether management is capable of achieving best-in-class revenue growth and margins with the existing portfolio or whether there is a need to separate the portfolio" etc. etc.; it goes on from there but you get the idea. It's a vague idea. The argument is not: Here's what DuPont needs to do, but management won't do it, and we will, so vote for us. It's: DuPont is a company, and as a company it has a board of directors, and its current board is okay-ish, but if you add us it will be an even better board, and then it will go around doing board things, just like it does now, only better.

The rest of the presentation goes into specific criticisms of corporate costs, poor M&A decisions, executive compensation programs and so forth. But the Trian pitch was not that its nominees would take specific actions or reverse specific decisions, but rather that they'd bring a better decision-making process to the board. Slide 17 explains that process:

At large capitalization companies, it is not unusual for Board and Committee meetings to entail a thousand pages of material; as a result, it is difficult, if not impossible, for any director without a team of analysts to thoroughly review all of the information

When a Trian partner joins a Board, the resources of Trian are mobilized:

  1. Trian maintains the confidentiality of the Board materials
  2. An operational, financial, strategic and legal diligence request list is sent to the company
  3. The Trian team examines all the information
  4. Trian employees interact with several layers of management

The Trian director is therefore able to:

  1. Reduce management’s “information advantage” over Board members
  2. Provide thoughtful analysis and recommendations based on material, non-public information
  3. Ensure decisions are made by the Board in a dispassionate and clinical fashion

This slide is so important that Trian printed it twice; it's also slide 32. They really like their board process! But not everyone liked it as much. Steven Davidoff Solomon summarizes the lessons of the DuPont fight, including the reaction to that board process:

The biggest issue forestalling settlement in the DuPont matter was the issue of whether to place a Trian representative on the company’s board. DuPont’s management objected, according to people close to the company, because it did not want a “back office” boardroom filled with Trian directors second-guessing its decisions. Trian directors would have their own staff members at Trian who could conduct independent analysis.

In a classic high-stakes proxy fight that's a good thing: The activist has a big plan (sell the company, break it up, fire the CEO), or even a medium-sized plan (do a big buyback, fix broken governance), and doesn't trust management to execute it. So the activist's nominees come in with their own analysis of how to fix things, provided by the activist's own team, because they've been elected specifically to overrule management and implement their plan. In the DuPont fight, though, this process conflicts with the whole "work collaboratively with the Board" thing. Either you're there to fight with management, or you're there to work with management. If you're there to work with management, there's a limit to how much second-guessing you can do without being rude. 

Of course in many cases directors probably should be ruder to managers. But one lesson of DuPont might be that while activists can win a mandate to radically change a company, or to throw out and replace management, it's harder to win a mandate to just help out in managing the business. Managing is for management, not for shareholders. This sort of fits with the director-primacy theory of corporate governance, as described for instance by Stephen Bainbridge:

Shareholder voting is properly understood not as an integral aspect of the corporate decision-making structure, but rather as an accountability device of last resort to be used sparingly, at best. ... Throughout corporation law preservation of managerial discretion is the default presumption. Because the separation of ownership and control mandated by U.S. corporate law effects just such a presumption, by constraining shareholders both from reviewing most board decisions and from substituting their judgment for that of the board, that separation has a strong efficiency justification.

Shareholders delegate their decision-making rights to directors. Those directors normally have only very limited accountability: They can do more or less whatever they think is right, have no specific obligations to shareholders and can't be held accountable except in exceptional circumstances. Of course, activists are an accountability mechanism: When the board has gone way off the rails, an activist can come in and run a proxy fight to try to get the company sold or broken up or otherwise drastically changed.

But activists themselves apparently don't get the same free pass on accountability. An incumbent board can tell shareholders: Elect us and trust us to run the company. That's how boards always work. But activists don't do as well by asking for trust. They get their best results by presenting clear plans that other shareholders can evaluate. If they're going to bring their own outside analysis into the boardroom, other shareholders want to see it first. Shareholders have a lot of experience trusting management; they are not quite as willing to trust one another.

  1. I mean, Trian certainly discussed a breakup with management, and its presentation mentions the possibility that Trian might want to "separate the portfolio," but insists that Trian would be "open-minded as to the best path forward." DuPont's management said that Trian wanted "to break up the Company and increase both risk and leverage. "

  2. I kid, sort of; slide 32 is part of the "Executive Summary," which repeats some slides. Also, can you blame them for liking the process? I wish I had a team of experts to summarize thousand-page documents for me. 

  3. There might be other lessons in DuPont's shareholder base. For instance, DuPont's "army of retail shareholders" were mobilized to vote against Trian's nominees. Trian was a bit snide about that:

    DuPont “clearly did a better job with the retail shareholder who clearly doesn’t understand the issues,” Peltz told reporters in Wilmington immediately after the meeting. “They did a better job scaring people.”

    But that's because the issues were hard to understand: Instead of a single big decision, there was a debate over board processes, and the retail shareholders opted for the default process of trusting management. And it wasn't just retail: Trian also lost in part because it couldn't win over the biggest shareholders, index funds run by Vanguard and BlackRock and State Street. Remember when we talked about the guys who think that index funds should be illegal because they don't put enough pressure on managers to behave competitively? You could read this result as a half-vindication of their argument: Trian won over a lot of institutional investors, as well as proxy-advisory services like Institutional Shareholder Services and Glass Lewis, but was unable to win over index funds, perhaps because their mechanical diversification makes them less interested in putting pressure on managers to perform.

  4. This theory is perhaps most famously championed by Marty Lipton (disclosure: my old boss). One oddity of the DuPont battle is that, after ISS recommended that shareholders vote for the Trian slate, Lipton wrote a client memo saying that Trian and some other activists "have clearly established credibility and acceptability" and "become respected members of the financial community," and suggesting that sometimes the best course for boards is to settle. But he also viewed the fight through the lens of board accountability:

    Meaningful director evaluation has also become a key objective of institutional investors and a corporation is well advised to have it and talk to its investors about it. Regular board renewal and refreshment can be important evidence that meaningful director evaluation is occurring.

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.

To contact the author on this story:
Matt Levine at

To contact the editor on this story:
Zara Kessler at