Over the last decade, many large corporations in the United States have split the roles of Chairman and CEO. The direct drivers have been Sarbanes-Oxley and other regulations on corporate governance, increased SEC scrutiny, and activist investors, both institutional and individual. The indirect driver has been corporate scandals, centering around too-powerful CEOs, that negatively impact shareholder value.
Heightened attention to corporate governance has worthy goals – better oversight, checks and balances, transparency, and disclosure through adequate independence of boards of directors. Some see splitting the Chairman and CEO roles as a direct path to these goals, even the litmus test of good governance. The Millstein Center for Corporate Governance and Performance calls splitting roles “a logical next step in the development of an independent board.”
However, the answer is not that simple. There are sound arguments and examples on both sides. The consensus advice on whether to split or combine the roles is the classic “it depends.” Drivers and situations and appropriate actions vary. Companies with roles combined today typically elect independent Lead Directors to maintain checks and balances, facilitate good governance, and satisfy the analysts and regulators.
Before we cover our new research, here’s background on the trends in leadership structure, and the pros and cons of combining or splitting the Chairman and CEO roles.
Having one person serve as both Chairman and CEO has long been the norm in large U.S. companies, but the trend is toward splitting the roles. According to Russell Reynolds, only 21% of the S&P 500 had split roles in 2001; in 2012 it was up to 44%.2 Booz & Company estimates that half the incoming CEOs in the U.S. and Europe in 2000 also assumed the Chairman role; in 2009 it was down to 17% in the U.S. and 7% in Europe.3
Larger companies, however, tend to keep the roles combined. Equilar reports that only half of the S&P 1500 have the roles combined, but 58% of large-cap companies do.4 And the pattern varies by industry – utilities and consumer goods companies are most likely to have combined roles, technology and services companies least likely. Among the NASDAQ 100 (with heavy technology representation), 62% have the roles split.5
Lead Directors have become the norm. The “2012 Spencer Stuart Board Index” finds that 92% of the S&P 500 have either an individual designated as Lead Director or governance policy provisions for a Presiding Director to chair meetings of the independent directors (typically on a rotating basis). Among those, 58% have Lead Directors (up from 40% in 2007) and 42% Presiding Director arrangements.6
- The percentage of independent directors has risen from 79% in 2002 to 84% in 2012.
- The CEO is the only non-independent director on 59% of boards today, up from 31% in 2002.
- 23% of the boards have a Non-Executive Chairman who is truly independent (e.g., not the former CEO), up from 13% in 2007.
Note that having the roles combined is a North American, and to a lesser extent a European, phenomenon. Around most of the world, 90% of companies keep the roles split. 7 In Germany, a split is mandated by law.
Split versus Combine
Proponents of combining the Chairman and CEO roles offer basic and sound reasons. Most importantly, it enables unity of direction and speed of action, especially in times of business crisis or transformation. It also lowers the “agency costs” of communication and coordination, which can be especially important in industries where a company’s “insider information” is important and yet difficult for board members to gather and interpret. Corporations can maximize the leverage of the skills and experience of established and successful executives by placing them in dual roles.
This side cites examples where disagreement or even rivalry between CEO and Chairman hamstring a corporation. And they point to problems with a common form of split, where the former CEO retains the Chairman role. If the former CEO doesn’t really relinquish the reins, the company can end up with two people trying to run it.
CEOs themselves tend to be much in favor of holding both positions, as we’ve seen in response to recent calls for splitting roles at Goldman Sachs and J.P. Morgan Chase. Some executives will not accept CEO roles unless Chairmanship is included.
- Concentrated power. Holding both roles creates an inherent conflict of interest. Some dual-role CEOs have infamously used their power to enrich themselves rather than their shareholders, while others have taken ill-conceived and ego-driven actions including making ill-chosen acquisitions. Key among the board’s roles is selecting, overseeing, and evaluating the CEO. How can that happen when the CEO heads the board? How can a CEO serve as his or her own boss?
- Better governance. Boards need independence to exercise real oversight, and that starts with the Chairman. Corporations with split roles have more transparency, better disclosure, and more effective checks and balances. Shareholders have a stronger voice.
- Job’s too big. Day-to-day operational management and organizational leadership is very different from the responsibilities of board-level oversight and governance. Leading the company and leading the board is too big a job for even a seasoned executive. When trying to do both, the jobs are mutually distracting.
There is no activist organization behind combining the roles or keeping them combined (but who organizes in defense of the status quo?). A leading voice for splitting the roles in the name of good governance has been the Millstein Center for Corporate Governance and Performance at the Yale School of Management, together with its affiliated Chairmen’s Forum.8 The National Association of Corporate Directors (NACD) lists the pros and cons of each alternative; it leans toward encouraging splits while acknowledging that combined roles plus an independent Lead Director can be a successful model.9
Split-or-combine has not become an issue for the average shareholder. Proxy votes to split the roles fail when opposed by management. And the stock market tends not to react negatively to news of the expansion or contraction of a CEO’s role.
Variety of Stances
There has been a variety of academic research on the effects of corporate leadership structures, including the elusive question of what model correlates with higher business performance. A 2008 study (of companies in the 1999-2003 time period) found that having split or combined roles had no appreciable effect on performance.10 A recent study by GMI Ratings found that companies with split roles have higher 5-year shareholder returns – but lower 1-year and 3-year returns.11 We have yet to see convincing evidence on the performance question.
On the pro-combine side, a 2009 study of companies of all sizes from 1998 to 2005 (including the period of some of the most egregious corporate scandals) dispels the major concerns with one person’s holding both roles.12 The authors found that CEO-Chairmen overall do not use their positions for personal benefit, claim excessively high compensation, use their discretionary power to manage earnings, or go on acquisition sprees. Nor did they find evidence of lower dividends or market valuations. Interestingly, they did find conditions under which the roles tend to be combined: when the company is in a risk-intensive or highly concentrated industry, when the CEO has a strong track record, and when there is strong governance to maintain balance.
Recent research on the pro-split side paints a different picture. The GMI Ratings study of 180 North American mega-cap (over $20B) companies finds that executives in dual roles are paid significantly more than the combination of separate CEO and Chairman salaries.13 Corporations with combined roles are twice as likely to receive low ratings on GMI’s environmental, social and governance (ESG) scale, and they are twice as likely to have their accounting practices rated “Aggressive” rather than “Average” or “Conservative” (but companies with split roles are just as likely to have “Very Aggressive” practices).
Other GMI research points out that board turnover is low, directors are getting older, and they’re holding their seats longer.14 Entrenched boards exercise less oversight and are more likely to rubber-stamp the actions of a powerful dual-role CEO. Their conclusion is that companies with power combined in one individual are more subject to management and governance failures. (Keep in mind that GMI is in the business of rating corporate governance.)
Among those not wedded to either side, the consensus is that different models can work in different situations. In a February 2013 McKinsey Quarterly article, William George argues that “Board governance depends on where you sit.”15 He should know – the former Chairman and CEO of Medtronic, he’s also been in positions as CEO only and as Chairman only, and he’s served on ten corporate boards.
A 2005 MIT Sloan Management Review article compares the American model of combined roles with the British model of split roles and Non-Executive but often very active Chairmen. After weighing the pros (for example, there’s more attention to the functioning of the board, and the CEO has a close mentor and advisor) and cons (lines of responsibility can get confused, especially when the Chairman is a strong public face of the firm), the authors conclude: “No compelling argument exists for splitting the chairman and CEO jobs, and any U.S. company that does so should take into account the lessons of the British experience.”16
In a 2012 article in The Corporate Board, “Splitting the CEO And Chairman Roles – Yes or No?” Charles Tribbett of Russell Reynolds weighs the arguments for splitting versus combining and describes the merits of an independent Lead Director position.17 He also cites examples where splitting was appropriate (at Avon) and where remaining combined was appropriate (at Cardinal Health) despite shareholder pressure to the contrary.
In another 2012 article in The Corporate Board, “Dividing The Seat Of Power,” three authors from CTPartners focus on the goal of independent leadership in the boardroom and focus on the benefits of a Lead Director when the Chairman and CEO roles are combined.18
Given the trend toward splitting roles, people are studying how and when to do so. A 2013 article in the Academy of Management Journal looks at three types of splits – apprentice (executive relinquishes CEO role but remains Chairman), demotion (executive relinquishes Chairman role but remains CEO), and departures (new executive in each role).19 The least common, demotion, has the greatest effect on total shareholder return (TSR), but timing is critical. Demotion when the company is doing well drives TSR down; demotion when the company is underperforming drives TSR up. For some older and basic advice on splitting the roles, see the 2004 McKinsey Quarterly article, “How to separate the roles of chairman and CEO.”20