CEOs with COOs: Two heads are not necessarily better than one

Depending on the size of the corporation, the diversity of its products and the background of the CEO, a COO as second-in-command can be a help or a hindrance, according to a new study by professors from the W. P. Carey School of Business and Pennsylvania State University. The study, CEOs Who Have COOs: Contingency Analysis of an Unexplored Structural Form, reported that in many instances, "CEOs who have COOs deliver lower organizational performance than those who do not." The study probes this phenomenon and offers several possible explanations.

When Wind Mill Slatwall Products of Sheboygan Falls, Wis., recently appointed Steven J. Neumeyer to its newly created position of chief operating officer, President and CEO Jay Hogfeldt stated, "In this role, Neumeyer will guide the day-to-day operations and be responsible for all functions and departments of the company."

When Redwood City., Calif.-based Tumbleweed Communications Corp. President and Chief Operating Officer John Vigouroux resigned in the spring of 2004 after the software maker posted a loss of $531,000 for the most recent quarter, CEO and Chairman Jeffrey Smith assumed Vigouroux's titles and responsibilities.

When international minerals conglomerate AngloGold Ashanti Chief Operating Officer Dave Hodgson and Chief Financial Officer Jonathan Best recently announced their desire to retire within the next several months, a successor for Best was anointed, and the company said two COOs would take over Hodgson's duties. "In view of the company's size, with 21 operations in 10 countries on four continents, the board believes that it is necessary to appoint two chief operating officers in his place," the company stated.

Too many chiefs?

These three companies' experiences illustrate three very different results for the role of COOs in modern corporations. Indeed, although two heads generally are better than one, that is not necessarily true when it comes to a corporation having a chief executive officer and a chief operating officer.

Depending on the size of the corporation, the diversity of its products and the background of the CEO, a COO as second-in-command can be a help or a hindrance, according to a study by Albert A. Cannella of the W. P. Carey School of Business and co-author Donald C. Hambrick of Pennsylvania State University.

"Based on a large 10-year sample, we find some support for the contingency view in explaining the presence of COOs, but almost no support in explaining when COOs are most beneficial," states their study titled "CEOs Who Have COOs: Contingency Analysis of an Unexplored Structural Form."

"Instead, we find strong evidence of a very substantial negative main effect: CEOs who have COOs deliver lower organizational performance than those who do not." In other words, Cannella said, the success or failure of the CEO-COO setup depends on a company's specific situation.

The study traces the possible origins of the COO position to a 1954 book by Peter Drucker, whose 'The Practice of Management' suggested the duties of a modern CEO are too much for one person.

Cannella and Hambrick further note that "no mention of such a position exists" in major pre-1975 works about executive roles and responsibilities and that "only a minority of companies" today have a COO.

Internal vs. external focus

Based on interviews with eight CEOs who had COOs and five CEOs who did not have COOs, Cannella and Hambrick confirmed the widely held assumption that CEOs focus on external and long-term issues while their COOs handled internal operations. They also found that the decision whether or not to have a COO is usually the choice of the CEO.

The study found that the CEO-COO arrangement was more likely to occur when:

• There is 'industry dynamism,' in other words, high growth, volatile competition and a prevalence of technology-intensive industries.
• There is a need to interpret ambiguous information and make frequent strategic adjustments regarding product features and customer preferences.
• Large-scale mergers or other consolidations occupy a great deal of the CEO's attention.
• A company is highly diversified.

On a personal level, a CEO is more likely to have a COO if the CEO believes he or she has limitations that must be compensated for, the study says. Such limitations include a lack of expertise in operational matters and a dearth of detailed knowledge of the company's products, suppliers and customers.

Such limitations are more likely, the study says, when the CEO has a background in accounting, law or finance — and when the CEO was hired from outside the company instead of from within. "CEOs from the outside," the study notes, "are at a disadvantage in comprehending the inner workings of the firm — its people, culture and processes."

"One of our basic predictions is that only certain kinds of CEOs are likely to have COOs," Cannella said in an interview after the study. "Those who do not want to get their hands dirty with internal affairs are the ones most likely to delegate internal operating decisions to another. The COO in this setting is basically doing the work that the CEO finds distasteful."

The CEO-COO structure can hinder an organization's performance, Cannella and Hambrick said, when it adds a layer of sometimes costly and time-consuming bureaucracy that also may separate the formulation of strategy from its implementation.

Among the 404 firms in 21 industries involved in the study, there was a 20 percent incidence of COOs, and the use of a COO was gradually declining during the time frame of the study.

Abdication of leadership?

The study found that, not only was the incidence of a COO not linked to better return on assets and stock value, but that it has a negative effect.

"Our most notable result is the evidence that CEOs who have COOs deliver lower firm performance than those who do not," the authors state.

"We weren't as surprised as some might be, because we saw the delegation of half of the CEO's responsibilities, especially the internal half, as an abdication of leadership," Cannella said. "It is hard to motivate and inspire people when you put another person in between you and them. We don't like the COO deal, and though we do a decent job of staying unbiased in our discussion and presentation, we weren't surprised by the results."

Does this mean the COO has been overused, perhaps as a novelty, which is wearing off?

"That is hard to say. We do document a decline in the use of COOs, but honestly can't tell what the real cause of the decline is," Cannella said. "One possible explanation is the concept of leadership is changing to a more 'independent' model, or one in which the leader is expected to do everything well. Unrealistic expectations, yes, but they do tend to characterize thinking these days."