This Tuesday, May 21, shareholders at JP Morgan Chase & Co. (“JPMorgan”) will vote on whether the bank should separate the roles of chairman and CEO.  Currently, Jamie Dimon holds both titles.  The impending vote is not binding on the board of directors, but has nonetheless caused significant controversy.  This post evaluates the arguments on both sides and asks how shareholders should vote.

The first claim made by advocates of separation is that combining the two roles creates a fundamental conflict of interest.  “How can the CEO be his own boss?”  “If the person leading the oversight is the overseen, it’s a fundamentally flawed system.”  Given that the primary duty of the board is to oversee management on behalf of the shareholders, it seems obvious that a board dominated by management would be conflicted.   For this reason, Andrew Ross Sorkin claims that the “knee-jerk response . . . to good governance is to separate the roles of chairman and chief executive.”

However, in the case of JPMorgan, this conflict may already be mitigated by a different structural separation:  the separation of dual chairman/CEO from the presiding director and independent board.  According to Sorkin, the presiding director – currently, Lee Raymond – is already, in effect, the chairman of the board, even though he does not hold this title.  This is because his duties are “virtually the same” as those he would have as chairman.  To support this position, Sorkin cites famed corporate governance champion, Ira Millstein:  “A strong leader with the same duties as a chair might serve the same purpose.”

But does Sorkin state the case accurately?  Not according to Professor Charles M. Elson, the Edgar S.Woolard, Jr. Chair in Corporate Governance and Director of the John L. Weinberg Center for Corporate Governance at the University of Delaware.  Elson told me on Thursday that “titles mean something.”  The Chairman “is setting the agenda and running the meetings.”  According to JPMorgan’s board, Lee Raymond only approves the meeting agenda and has the power to add agenda items.  In addition, Raymond only presides at the meetings when Dimon is not present.  This does not sound like the same duties that Raymond would have if he were the chairman.

JPMorgan’s board, however, also points out other structural mechanisms that weaken Dimon’s influence.  Among other things, only the independent directors (meaning everyone except Dimon) appoint the presiding director each year and the presiding director can call meetings of the independent directors without Dimon’s presence.  This suggests that the independent members of the board can act without Dimon unduly influencing them.  Furthermore, as Dimon is only one member on an 11-member board, the independent directors can theoretically discipline him, if need be.

While this may undermine the broad assertion that Dimon, as chairman, is effectively his own boss, Elson explains that this “doesn’t change the day to day dynamics of the board.”  Having the CEO set the agenda and run the meetings simply interferes with the board’s “critical oversight” function.   Moreover, it is “a waste of the CEO’s time,” says Elson.   Elson recently co-chaired a study group report on corporate boards that states the following:

running a business and running a board require two different skill sets and temperaments.  Individuals who become CEOs tend to have strong egos and high optimism.  Motivated by vision, they aim for high growth and tolerate high risk.  By contrast, the most effective Chairs tend to be consensus builders who try to balance the two.  Boards that can balance the “dynamic” CEO and the “wise” (and sometimes older) Chair can have highly effective governance.

CEOs like Dimon may focus too much on running the company instead of running the board.  For this reason, among others, the group, which also includes a former U.S. Secretary of the Treasury and former Chairman of the Securities and Exchange Commission (SEC), recommended splitting the roles of chairman and CEO as a default structure.  However, this recommendation was not unanimous, and it also recognized that the structure utilized by JPMorgan may be legitimate in some circumstances.  But the circumstances cited, such as responding to a catastrophic corporate event or a company founder who retains substantial equity ownership, do not apply to JPMorgan.  On balance, shareholders will have to decide for themselves whether splitting the two roles is really a matter of semantics or instead a matter of substantive governance.

The second claim made by advocates of separation is that studies show that an independent chair improves financial performance.  The shareholder proponent leading the charge cites two studies by Booz & Co.  One found that in 2006, “all of the underperforming North American CEOs with long tenure had either held the additional title of company chairman or served under a chairman who was the former CEO.”  This study, accordingly, argues that splitting the role of chairman and CEO is better for investors.  The study, however, does not consider the particular structural separation that JPMorgan uses.  The other from 2010 found that “worldwide, companies are now routinely separating the jobs of chair and CEO.”  Interestingly, however, the same study found that “the jury is still out on whether having a single-person chairman and CEO benefits or hurts company performance.”  It notes that “no one governance model consistently outperforms the other.”  Some more recent studies and papers also advocate the separation of the CEO and the chairman, but none present conclusive evidence that the separation of the chair and CEO improves financial performance.

Other studies conclude that the particular leadership structure chosen is less important than other factors such as the chairman’s industry knowledge, leadership skills, and influence on board process.  One study, in particular, cited by Professor Steven M. Davidoff, shows that “in more complex companies like JPMorgan, where it is harder to monitor the chief executive, separating the roles has had less effect.”  This may mean that JPMorgan is an exception to the general rule.  In the words of Ira Millstein, “one size may not necessarily fit all.”   Two JPMorgan directors currently make this claim in defense of preserving Dimon’s dual roles.  This suggests that more targeted research on complex firms employing JPMorgan’s form of structural separation may be needed before changing the status quo.  However, academic studies do not tell the whole story.  To the extent that Dimon is truly looking out for the interests of shareholders and not simply himself, it is hard to explain his recent threats to quit if he is prevented from keeping both titles.

The final claim that separation advocates raise specifically in the context of JPMorgan is Dimon’s handling of the London Whale incident.  According to the shareholder proponent, the resulting $5.8 billion loss “tainted Mr. Dimon’s reputation as one of Wall Street’s best risk managers, and raised questions about the board’s oversight.”  Economist Simon Johnson also cited the London Whale incident as “Exhibit A” in the case that Dimon and the board were not taking risk limits seriously.

On Thursday, I consulted corporate governance expert Professor Robert J. Jackson Jr. of Columbia Law School about this claim.  He told me that “in the wake of the London Whale debacle, shareholders are right to ask hard questions about JPMorgan’s leadership, and in particular its oversight of risk.”  However, Jackson expressed skepticism that separating the CEO from the chairman would address incidents like the London Whale in the future.  “While some evidence suggests that separating those roles might be desirable at some companies, it’s not clear how it would help make sure that risk is better managed at JPMorgan going forward.”   Others have also argued that the London Whale would not have been prevented by having an independent chairman.  Like Millstein, Professor Jackson warns against a “one-size-fits-all-solution” and emphasizes that “shareholders might be better advised to demand governance changes that will ensure that the JPMorgan board is better prepared to manage the firm’s risk in the future.” 

The London Whale trading loss, moreover, must be seen in the context of JPMorgan’s overall performance.  In spite of the incident, JPMorgan continues to show strong results.  It outperformed the broad S&P index for the past five years.   Shareholders must consider whether the board’s remedial actions, including docking Dimon’s pay by 50%, are sufficient means of deterrence.  Linking the separation of chairman and CEO to one bad incident in this context appears to be too attenuated.

So, how should shareholders vote? My sense is that a separation of powers model is always preferable, in the long run, to a model that encourages a benevolent dictator.  JPMorgan’s model appears to be closer to the former, though it is nonetheless a blend that leaves Dimon with more power than is likely ideal.  In any event, in the short-run, JPMorgan’s numbers speak for themselves.  Without better evidence that Dimon’s dual role is hurting shareholder value, it might be best, this year, for shareholders to follow Bill Tilden’s old adage, “never change a winning game.”  As a check and balance on Dimon, this same issue can be revisited next year.