• Archive

Holes at the Top: Why CEO Firings Backfire - The Right Way to Replace a Fired CEO

Boards of Directors often botch up the job of finding a replacement CEO. This excerpt from Harvard Business Review explains that boards that do it right tend to have three things in common: They understand that selecting a CEO is a major responsibility; they help the CEO set realistic performance expectations; and they develop a deep understanding of the company's strategic position.

The Right Way to Replace a Fired CEO

Firing a CEO rarely pays off in terms of improved earnings or stock-price performance, says U.C. Irvine's Margarethe Wiersema. The reason: Boards of directors lack the strategic understanding of the business necessary to give due diligence to the CEO selection process. In this excerpt, Wiersema outlines the right way to find a replacement CEO.

Although many companies fail when it comes to replacing a CEO, not all do. Boards that manage the task effectively tend to have three things in common: They understand that selecting a CEO is a major responsibility, one that's theirs alone; they help the CEO set realistic performance expectations; and they develop a deep understanding of the company's strategic position.

Take the responsibility seriously
During the succession process, the board should let strategic needs dictate selection criteria for the new CEO. That's not always easy to do; board members have a tendency to become starstruck. This happens partly because they're trying to please the investment community, but it also happens because they, like investors, imagine that an outsider's charisma or past experience will trump his lack of knowledge about the company or the industry. To circumvent this problem, smart boards first identify the market, competitive, and technological factors that influence the company's performance. If members keep these factors in mind, they'll be better able to identify the skills and experiences that the new CEO will need.

Home Depot's recent CEO replacement process provides a good example of how this should be done. Ken Langone, the board's lead director, was acutely aware that the market and business had evolved in important ways. After two decades of spectacular growth, the company had reached a plateau. Big-box home improvement stores were beginning to look like any other mature business; slower growth and inroads by competitors had started to erode profit margins. Langone and other directors recognized that the founders had outgrown the business and that the company needed a new leader with experience improving efficiency and service. With this in mind, Langone suggested a candidate he knew from his service on GE's board. Bob Nardelli, the former CEO of GE Power Systems, was a talented executive with exactly the kind of experience Home Depot needed.

Boards that manage the task effectively tend to have three things in common.
— Margarethe Wiersema

Cofounders and directors Bernie Marcus and Arthur Blank quickly realized the match between Nardelli's credentials and the strategic challenges the company faced. Nardelli proved to be the right choice. Had Langone and the other directors failed to identify the specific attributes required of the new CEO, they likely would never have considered Nardelli—who had virtually no retail experience—for the job.

Even when companies use search firms to find candidates, board members still must first identify the company's competitive challenges and industry context, as well as the skills a replacement CEO will need. Doing so will allow them to guide (rather than be guided by) the recruiters. And it will ensure that recruiters look for a specific set of skills and experiences, rather than simply run through their Rolodexes of available executives.

Set realistic performance expectations
During the economic expansion of the 1980s and 1990s, the investment community came to expect constant improvement in earnings, and most executives and boards were willing to prop up their stock prices by setting targets that pleased Wall Street. Though it's now clear that this practice has created serious problems for corporate America, the pressure from investors lingers. New CEOs are often tempted to continue playing this earnings game by promising unrealistic turnaround numbers, but that's precisely the wrong thing to do. Instead, the CEO needs to restart the clock by deflating unrealistic expectations, and the board needs to support that shift.

James Kilts was quick to reset expectations when he was appointed CEO at Gillette in January 2001. In his first meeting with analysts in June of that year, Kilts lambasted the unrealistic earnings forecasts set by Michael Hawley, his predecessor. Hawley had done everything possible to return Gillette to the 15 percent to 20 percent earnings growth it delivered during the 1990s. "Everything possible" included shortsighted practices such as channel stuffing—that is, shipping inventory ahead of consumption to meet overblown sales and earnings goals. Despite these practices (or perhaps because of them), the company was unable to deliver on the numbers, and the investment community lost confidence in Hawley.

Kilts believed that Gillette's managers were making decisions based on those looming earnings forecast, rather than on long-term strategic thinking. So, with the support of the board, he refused to provide the investment community with any financial targets at all, except for that of long-term sales growth of 3 percent to 5 percent. He told analysts that he and the board had decided not to give them guidance going forward. His refusal to play the game left the investment community perplexed; three analysts downgraded the stock, and only five out of seventeen rated it a buy. Kilts stuck to his guns and focused his attention on the problems that had led to the earnings shortfalls. Analysts eventually noted that Gillette's product lines were rebuilding market share, and their confidence in the company's performance began to be restored. Recent earnings reports show that Gillette's strategy is beginning to pay off, and the stock has rebounded from its low.

Once board members have appointed a new CEO, their job is far from over.
— Margarethe Wiersema

The point isn't to starve the investment community of information—far from it. Though the wise board and CEO will back away from forecasts they can't meet, they'll also work extremely hard at communicating how they plan to address underlying competitive problems. After Jacques Nasser was dismissed from Ford, for example, William C. Ford, Jr., sent strong signals to investors. With advice and support from the board, he removed the executives perceived as responsible for Ford's problems. He also attempted to restore confidence in the company's finances by appointing former Wells Fargo Bank chairman and CEO Carl Reichardt as vice chairman, with the specific responsibility of overseeing Ford Financial. Then he announced a sweeping restructuring that trimmed production, rationalized car lines, reduced purchasing costs, and cut the labor force. These actions communicated the direction he and the board envisioned for the company more than any set of forecasted numbers could have done.

Provide more strategic oversight
And finally, good boards develop a deep understanding of the business and apply that understanding through active oversight of strategic performance. Once board members have appointed a new CEO, their job is far from over. They must resist the temptation to leave well enough alone for a while and remember that, since most replacement CEOs perform no better than their predecessors, they must be more vigilant than ever.

It's essential that boards provide strong strategic oversight following a dismissal because turnarounds aren't easy. They take subtlety, sophistication, and staying power. Consider General Motors. For nearly a decade under the leadership of Roger Smith, GM put profits ahead of sales—and watched its U.S. market share decline from 45 percent in 1980 to 36 percent in 1990. Eventually, the decline in market share eroded financial performance, and, despite the appointment of a new CEO, Robert Stempel, GM suffered losses in 1990 and 1991. The problems clearly went far beyond top leadership. GM offered too broad a range of product lines; the lines didn't share platforms or components; too many components were manufactured internally at higher costs; and the supply chain wasn't managed aggressively. Perhaps most important, central-office bureaucracy consistently stifled GM's ability to become a more innovative and efficient organization.

Following the back-to-back losses, Stempel, too, was dismissed, and the company finally got serious about turning itself around. The board appointed John F. Smith, Jr., who ultimately engineered GM's comeback. The board had to learn to be patient; the problems that had developed over a decade could not be solved overnight. Performance didn't improve much while Smith, with close oversight from the board, directed the much-needed overhaul: aggressively reducing costs through car platform rationalization, manufacturing consolidation, and outsourcing; spinning off the electrical components group, Delphi Automotive; selling aerospace business GM Hughes to Raytheon; and fundamentally restructuring the way the company conducted its business. Ultimately, GM improved its domestic and international operations and was rewarded with an increase in market share and a rebound in stock price. But that turnaround took almost a decade to accomplish—and it likely would never have happened at all if the board hadn't developed a sophisticated understanding of the complex, interconnected problems GM faced.

To ensure that they are providing proper oversight, board members should request from the CEO a strategic plan that identifies the options to consider, a timetable for a turnaround, most likely outcomes, and measures for evaluating whether the company is on track. Board members should closely question the assumptions underlying the strategic plan and how the proposed strategy alters the firm's risk factors and shapes its future direction. They should be willing to act if the new CEO does not improve the company's position in the marketplace. Despite the understandable desire to give the new CEO an extended honeymoon, members won't have fulfilled their primary fiduciary duty unless they formalize the responsibility to oversee strategy. End of Article

Excerpted with permission from "Holes at the Top: Why Ceo Firings Backfire," Harvard Business Review, Vol. 80, No. 12, December 2002.

[ Order the full article ]

Margarethe Wiersema is a professor of strategy at the University of California, Irvine.