You have surely heard that two Boeing 737 MAX airplanes fell in recent months, killing hundreds of people, and leading to the grounding of most 737 MAX planes in active use. Boeing, as expected, suffered a huge reputational loss (President Trump suggested to rebrand the MAX), and several airlines cancelled their MAX orders. While we are awaiting the final conclusions of the various investigations of the disastrous air accidents, a major cause surfaced of nowhere: Boeing’s CEO/board chair duality. Boeing CEO, Dennis Muilenburg, holds the dual positions of CEO and board chairman, a situation anathema to shareholder activists and “good corporate governance” promoters. “CEOs who are also board chairs aren’t properly monitored by the board and consequently the company’s operations and performance are deficient.” Say the promoters of splitting the CEO and board chair positions.
Indeed, that’s the advice of the two major and influential proxy advisory firms (advising institutional investors how to vote on shareholder resolutions and proxy contests), Institutional Shareholder Services (ISS) and Glass Lewis, to Boeing’s shareholders: yank the board chairmanship from your CEO. As reported by the Wall Street Journal (April 17, 2019): “Two proxy advisory firms say that Boeing Co. shareholders would benefit from seeing the chairman and chief executive roles separated following two fatal crashes of its 737 MAX plane…. ‘Concerns raised within the last year about the company’s risk oversight and actions related to the development, certification and marketing of the 737 MAX aircraft are serious enough to suggest that shareholders would benefit from the most robust form of independent board oversight, in the form of an independent board chair.’ ISS said in the note.”
So, now we know what could have prevented the disastrous 737 MAX accidents: the separation of the CEO and board chair positions.
Proxy advisors, like ISS and Glass Lewis, wield substantial power in the U.S., since most institutional investors follow their recommendations in voting on shareholder resolutions, proxy contests, and “say on pay” (managers’ compensation) votes. Most investors don’t have the capacity to carefully examine shareholder resolutions and “say on pay” proposals, and therefore follow the advisors’ recommendations. Given the prominence of proxy advisors, we need to carefully examine the merits of their recommendation to split Boeing’s, and many other firms, CEO/board chair duality. As elsewhere on this blog, I now consider the relevant empirical evidence.
First, certain relevant U.S. statistics. (Apologies to my non-U.S. followers: I am not familiar with non-U.S. evidence on CEO/board chair duality. Perhaps you can enlighten us.) Slightly over half (52%) of S&P 500 companies (largest U.S. enterprises) are led by a dual chairman/CEO, whereas 29% have an independent board chair, and 19% an executive or other outside chair. The trend in recent years clearly moves toward separation: Ten years ago, 71% of S&P companies had a dual chairman/CO vs. 52% now (Spencer Stuart Board Index, 2015). Larker and Tayan’s (Stanford Closer Look Series, June 24, 2016) research indicates that: (1) large companies are more likely than small companies to have a dual (single) leadership, (2) most CEO/chairman separations occur during a CEO change process, (3) surprisingly, many separations are temporary, such as in 2012, IBM promoted Virginia Rometty to CEO with former CEO, Sam Palmisano, remaining as board chair. Ten months later Rometty got the board chairmanship too. (4) The frequency of change-back is unexpectedly high, note Larcker and Tayan: “Slightly over one-third (34%) of companies in our sample “permanently” separated the chairman and CEO roles and later combined them during the 20-year measurement period.”
So, judging from the actions of the 34% of companies which separated the CEO and board chair roles and later recombined them, the actual experience of having separate CEO and board chairs was far from satisfactory. The promoters of separation never mention this inconvenient fact. But this adverse experience is, in fact, corroborated by large sample evidence on the duality issue.
Would you believe that the persistent and widely-publicized calls to break the CEO/board chair duality fly in the face of the evidence? I surely was surprised. Thus, for example, Dalton et al. (Strategic Management Journal, 1998) conducted a meta-analysis (analyzing existing studies) of 31 studies and found no relationship between the corporate leadership structure (combined or separate CEO and board chair) and company performance (profits, stock returns). Perhaps, that’s an old study, so Krause et al. (Journal of Management, 2013) similarly reviewed 48 research papers and reported that the leadership structure (dual or not) had negligible impact on enterprise performance, managers’ entrenchment (difficulties to remove a CEO), organizational risk-taking, or executive compensation.
So, where are the great benefits touted by the promoters of the CEO/board chair separation? Definitely not in the data.
There is more: A comprehensive, and very careful study of 232 companies which switched from a dual to separate CEO/chair positions, compared with a control group of similar companies with dual CEO/board chair leadership reported: “…announcement returns [i.e., stock returns upon the public announcement of the leadership split] and future performance are significantly lower and investments contribute significantly less to shareholder wealth for firms that indicated that they separated the roles [of CEO and board chair] due to environmental [institutional] pressures… Our compensation analyses reveal that CEOs in firms that switched away from a dual leadership structure also had a significant decrease in the pay-performance sensitivity of their compensation [namely, the relation between executive compensation and firm, or market performance]…” (Dey et al., Journal of Corporate Finance, 2011, emphasis mine).
So, here we are: There is absolutely no supporting evidence to the claims that separating the CEO and board chair positions enhance enterprise performance, or reduces managerial entrenchment and compensation. Namely, there is no evidence of benefits to shareholders from the separation. So where are the widely-claimed benefits of separation? Definitely not in the data.
It may, of course, be that in special circumstances, such as a poorly-performing CEO, or an unusually tough competitive environment, it will make sense to separate the CEO and board chair positions, thereby enhancing the board’s influence on operations. In other cases, nominating a “lead director,” while retaining the CEO/chair duality will work best. In any case, the separation of CEO from board chair isn’t a “fit all” panacea, and it’s highly doubtful that it would have prevented Boeing’s 737 MAX mishap.