Grow Your Own CEO
from
March/April 2002
by Julie Connelly
In case you hadn’t
noticed, it’s been open season on chief executives. Through the first
11 months of 2001, 871 of them left office, almost as many as the
record-setting 1,106 who moved on the previous year. And while some
thoroughbreds, like General Electric’s Jack Welch, cantered off to a
well-earned retirement, the carcasses of others—Ford Motor’s Jacques
Nasser, for example—were trussed up like dead game and dumped
ignominiously by the roadside. In many cases, departure came hard on
the heels of a CEO’s arrival in the corner office. Welch put in more
than two decades, but the average time served by departing CEOs was
just 5.9 years, according to the Chicago outplacement firm Challenger
Gray & Christmas Inc.
Some lasted a far shorter time than that. Richard Thoman stayed at Xerox Corp. for 13 months, and Durk Jager hung on for 17 at Procter & Gamble. Jeff Skilling was out after a scant six months at Enron Corp., but that was probably a special case. Humiliating headlines before a CEO has had much of a chance to get started don’t help longevity. “Time runs short for Terry Semel to save Yahoo!,” said The Wall Street Journal just five months after the former Warner Bros. co-head moved into his new job.
CEOs have become targets for a number of reasons. You know most of them, but they often boil down to board members’ failure to oversee the selection of the right corporate leader in the first place. This dereliction is compounded by another lapse. CEOs themselves are no longer putting the necessary time and resources into developing the next generation of leaders, and boards don’t even seem to be noticing. “The board should question, prod, and monitor the CEO about management development,” says Gertrude G. Michelson, an outside director of GE, which has an excellent CEO-succession system in place. “You don’t get succession candidates if you don’t start with a pipeline.”
Instead, directors at many companies dodge their responsibility by outsourcing the selection of CEO candidates to search consultants. But whose talent are you buying? “Why do you assume that other companies can develop leaders better than you can?” asks recruiter Frederick W. Wackerle, author of The Right CEO: Straight Talk About Making Tough CEO Selection Decisions (Jossey-Bass, 2001). Steven Currall, a professor of management and psychology at Rice University’s Jones Graduate School of Management, speaks of the “co-dependent relationship” between boards and search firms. As a result of it, he says, “boards ignore processes and structures to develop CEOs internally.”
It’s no secret that boards have fallen in love with the glamorous outsider, a breed that now accounts for about a third of the CEOs of the 700 largest U.S. companies, according to the search firm Spencer Stuart Inc. Jim C. Collins, author of Good to Great (HarperCollins, 2001), a book about how 11 mediocre companies became superlative ones, says of this worship of CEOs who’ve made their reputations elsewhere: “We are primitives dancing around a campfire in our understanding of management and leadership. The primal, primitive thing to do is go with the sun god. Occasionally, by chance, it works, which reinforces the belief.”
Rather than grow your own CEO, the thinking goes, it’s so much safer to buy the outsider who has already been one. AT&T found C. Michael Armstrong at Hughes Electronics Corp.; Eastman Kodak Co. spotted George Fisher at Motorola. When these stars don’t sparkle, boards can arrange for an encore by their companies’ former CEOs, a trend Corporate Board Member noted in its last issue. Lawrence Bossidy returned to Honeywell International, for example. Meanwhile, “the insider is penalized because he or she hasn’t been a CEO,” says Deborah Cornwall, managing director of the Corlund Group, a consulting firm in Boston. “And if the company hasn’t been performing well, the whole management team is tainted.”
Make no mistake: Employees and shareholders are being victimized by the cult of the celebrity CEO. He—or occasionally she—is paid like Bruce Willis and assumed to be just as much of an action hero. Both Lee Iacocca and Jack Welch have written best-selling autobiographies, to say nothing of The Donald and Chainsaw Al. And you’d have to live in another galaxy not to know about Andrew Grove’s prostate cancer or Gary Wendt’s messy divorce. “Some CEOs are better known than their companies,” says Rakesh Khurana, an assistant professor of management at Harvard Business School. “We suspend our disbelief, and when things don’t go right, we have someone to execute.”
It’s easy to be seduced by a charismatic messiah when you haven’t invested in the multiyear process of figuring out what you really need in the person who is going to run the company. But the transition from one CEO to another, difficult enough for an insider, becomes even more fraught when the new person has no history at your outfit. After all, if Carly Fiorina had been more familiar with Hewlett-Packard, would she have launched the Compaq merger without first testing the waters with the Hewlett and Packard families?
The 2001 Korn/Ferry International Annual Board of Directors Study reveals that only 36% of board members think most companies do an effective job of management succession. “It’s ironic,” says Patrick McGurn, director of corporate programs at Institutional Shareholder Services, a proxy advisory firm in Rockville, Maryland. “Shareholders used to complain that boards weren’t active enough in dealing with nonperforming CEOs. Now boards are pretty darn good at firing the CEO. But they’re faced with the issue of who’s next, and rarely do they have an answer.”
Failure costs a lot. Two University of Texas professors, Robert Parrino and Laura T. Starks, along with Richard W. Sias of Washington State University, have demonstrated that institutional investors decrease their holdings by an average of 12% in the year before a CEO is booted. And the shareholders are whacked another way: Companies hand out humongous severance payments so the guys will go quietly. It’s not always the guys, either. Even after the board of bankrupt Warnaco Corp. ousted her as CEO, Linda Wachner was holding out for a fat goodbye package, which press accounts valued at both $24 million and $44 million. Wachner herself declined to help narrow the number down. “I won’t comment on that,” she told Corporate Board Member.
Certainly there’s a risk to CEOs’ careers if they fail, but how serious is it when they will never have to work again, nor their children nor their children’s children? “You shouldn’t get nearly the same reward for failing as for succeeding,” says compensation consultant David Swinford, managing director of Pearl Meyer & Partners in New York City. “There has to be a cost of failing that includes something that seems like pain, and not the pain of making $50 million instead of $100 million.”
Occasionally someone puts a foot down. Kenneth Lay of Enron finally had second thoughts about taking a $60.6 million severance package after Enron’s energy traders, the company’s core assets, made it clear they wouldn’t stand for it. And the reason Michael Ovitz went to the Walt Disney Co., which he nailed for $90 million on the way out, was that Seagram Co. said no to the entire compensation package he wanted for running its Universal Studios.
Directors can do a lot to improve the odds of getting the right CEOs for their companies. Randall Tobias, chairman emeritus of Eli Lilly & Co. and a board member at Kimberly-Clark Corp., Phillips Petroleum Inc., and Knight Ridder Inc., believes that the first step is to elevate succession to a boardroom priority. “Let me give you an example,” he says. “DuPont historically has had better safety results than virtually anybody in corporate America. There’s one reason for that. Beginning with the CEO, they talk about safety every time they talk about anything. Things that get emphasis tend to be things that are done well, and I don’t think the succession process gets that kind of attention in most companies.”
Perhaps it’s about to. Peter Gleason, vice president of research and development at the National Association of Corporate Directors in Washington, D.C., reports that when directors ranked the issues they regarded as most important for their companies, management succession rose from fifth place in 1999 to second place in 2001. In first place, quite properly, came the board’s fiduciary responsibility to shareholders—though that may often be inseparable from the right (or wrong) choice of CEO.
In the 1960s and 1970s, when companies had a lot of jobs to fill, they needed to build management cadres. But then came the downsizings of the 1980s and 1990s, and rather than train their managers for increased responsibility, companies laid them off. Says Gertrude Michelson, who retired in 1992 from R.H. Macy & Co. as senior vice president in charge of external affairs: “I joined Macy in the company’s first postwar management-training class. There were 50 of us, and 10 were Harvard M.B.A.’s. In subsequent years, that got diluted. It’s one of the reasons there are so few homegrown leaders in retailing or in business generally.”
Today promising executives train themselves by leaping from company to company; only the stodgy stick with one employer. Sometimes that works out well. IBM’s Louis Gerstner is generally regarded as having engineered an incredibly successful turnaround for the stumbling computer giant. He started his career as a McKinsey & Co. consultant, went off to American Express Co. and rose to become its president, was recruited to RJR-Nabisco Inc. as CEO, and by 1993 was running IBM.
When Big Blue got him, Gerstner had had the kind of accelerated development that many companies can no longer provide for their managers. But it’s interesting to note that his likely successor is chief operating officer Samuel Palmisano, who has worked at IBM for 28 years.
The recent change at the helm of GE, a textbook illustration of succession by gladiatorial combat, illustrates the importance of management development. Many directors, CEOs, academics, and consultants are grudging about this, saying that only a company with GE’s deep bench could afford a succession in which, metaphorically speaking, the losers were taken out and shot. But they seem to miss the point that the three strong candidates who ended up as finalists in the competition to succeed Jack Welch were winnowed over six years from an original list of 23, and that the process was part of a painstaking system that has been in place since 1910.
Moreover, the two who didn’t get the job, W. James McNerney and Robert Nardelli, joined the very short list of corporate America’s most sought-after executives. McNerney went off to head 3M, and Nardelli Home Depot (see page 14 for “How Two GE Vets Make Out as CEOs”). Exults Kenneth Langone, a director at GE and Home Depot’s lead director, who recruited Nardelli: “Home Depot, 3M, and GE got the best that was in the pool.”
There is a troubling hint, however, that boards assumed that the three GE contestants were so well trained as to be completely fungible. Home Depot went after Nardelli as the only one of the three who had not yet been spoken for—not because the company’s directors decided early on that he was the man best suited to the job. “I’d have been thrilled to get any one of them,” says Langone. “They were all outstanding candidates, every one of them.”
You’ve doubtless read all you want to read about how Jack Welch spent 50% of his time on people issues. He showed up for Vince Lombardi-style pep talks at GE’s Crotonville, New York, training center (now the John F. Welch Leadership Development Center), personally evaluated the company’s top 500 executives, and demanded that managers cull their poor performers, encourage their average performers, and challenge their best people with jobs that would stretch them. He also made it clear that management development was to be a priority for the GE board.
“There are eight board meetings a year, and at breakfast before each one the management-development committee met for an hour and a half,” says Gertrude Michelson, who has clearly sat through her share of these conferences in 25 years as a GE director. “Then twice a year at the board meetings, we talked about management development. There were the plant visits that were usually a half-day or a whole day. There were lots of conversations on the phone. We’d talk about a new assignment for someone who was in the running. People’s eyes glaze over when they find out how much time it takes.” Adds Welch in his autobiography, Jack: Straight From the Gut (Warner Business Books, 2001): “Over the years we watched these guys like hawks. We kept throwing new tests in front of them. The eight who remained contenders by June 1998 had moved through 17 separate jobs.”
Helen Handfield-Jones, who works at McKinsey & Co. and co-wrote The War for Talent (Harvard Business School Press, 2001), says: “The vast majority of large U.S. companies are not very good at developing managers. Of 13,000 executives in the top 200 positions of 100 large U.S. companies, only 3% strongly agreed that their company develops managers quickly and effectively.”
Senior managers don’t have what Handfield-Jones calls a talent mind-set, which she defines as the belief that high-performing employees are a source of competitive advantage in any business. As a result, they think they have no particular responsibility for talent development and are therefore happy to delegate the job to human resources. “Senior leadership must constantly assess and identify the high performers and stretch them to maximize their growth,” says Handfield-Jones. “That just doesn’t happen in companies today.”
Directors have a responsibility to get to know who the top-performing managers in the company are. “You don’t create a successful management-succession program overnight,” says executive recruiter Thomas J. Neff, chairman of Spencer Stuart U.S. “It takes years to be fully effective. The program has really got to focus not just on the top job, which is obviously the most important part of it, but on the top 50 to 100 jobs, depending on the size of the company. It needs to be regularly visited by the CEO. Boards need to be actively engaged in it as well, and informed about what is going on to be sure there’s a program in place that is being taken seriously.”
Admittedly, most companies, however assiduous they become about management development, probably can’t field three potential CEOs from within their own ranks, as GE was able to do. That’s not necessarily a strike against them. Says William W. “Bill” George, chairman of the Minneapolis medical-technology company Medtronic Inc. and a director of Target Corp., Novartis AG, and Imation Corp.: “I don’t think GE’s situation would work well for most companies, because few are prepared to get rid of so many talented executives.”
But what excuse does a board have for not riding herd on a CEO to produce one succession candidate? “Succession is the board’s job, by corporate-governance rules,” says Peter Crist, vice chairman of Korn/Ferry and head of the firm’s board practice. “The CEO’s job is to train him.”
Bill George carefully picked his successor and groomed him patiently for nine years, and Arthur Collins took over as CEO last May. This May George, who by then will be a visiting professor at the International Institute for Management Development in Lausanne, Switzerland, as far away from Medtronic as he can reasonably get, will retire from the company.
As it happens, George himself was a relative outsider when he stepped into the company’s top job. Two years earlier he’d left Honeywell, where he had been president of the space-and-aviation division, to become chief operating officer of Medtronic. “I went to the board of directors shortly before I took office as CEO in 1991, and proposed that I would stay for 10 years and at the end of that time I’d have a successor not just identified, but ready to go,” he says.
In fact, he urged the directors to pass a motion limiting his term to 10 years, but they deadlocked six to six and the motion died. Why that length of time? “It was an arbitrary figure, but it seemed like the right time to have a full term to get the things done that you need to get done,” says George. “Sometimes strategies take a full 10 years to play out. And unless one sets a date certain for succession, it’s virtually impossible to groom a successor.”
Once he was CEO, George, who was 49, found that Medtronic had a number of capable operating executives running the company’s divisions. But he thought none of them was the right person to succeed him as CEO, so he engaged Fred Wackerle, the search consultant who had recruited him to Medtronic, to help him find a successor. Wackerle turned up Collins, then in his early forties and running the $2 billion-a-year diagnostics division of Abbott Laboratories. In 1992 George and Wackerle persuaded Collins—“twisted his arm,” George says—to come to the much smaller Medtronic as president of international operations, with the idea that he’d be groomed to become CEO. In the years since, Medtronic’s annual sales have grown from $1 billion to $5.4 billion.
What set Collins apart? “Art had all the qualities that one would look for in a CEO,” George says. “He has extremely good judgment; he had run major entities for Abbott Labs. He had lived in Germany and France, which I thought was very good broadening experience. And he had the presence of someone who was going to become a CEO. It would take a few years, but he had the presence all along of someone who could do the job.”
What George was also looking for was an executive who had the capacity to run Medtronic not just as the comparatively small company it then was but as the $25 billion corporation he thought it could become by 2010. But he had no idea what might produce those billions, and he needed someone who was comfortable with change.
At the annual reviews of succession planning with the board, George discussed Collins as his leading candidate, though he also kept the directors up-to-date on the progress of younger managers in their thirties and forties. He explained how he wanted to bring Collins along. “We gave him a variety of opportunities,” he recalls. “He took on international and did extremely well with it; then he became chief operating officer, and then president and chief operating officer. And as the business grew, Art and I worked closely together.” In 1997, George told the board that Collins was ready to become CEO at any time.
To ease him into the job, George formed an office of the CEO that shared the decision-making. It met every Monday morning and included Collins and vice chairman Glen Nelson, who was planning to retire soon. Gradually George started to turn the leadership of the company over to his successor. Collins got involved in strategic matters, especially acquisitions and integrating new businesses into the company. He began to represent Medtronic in its dealings with the U.S. Food and Drug Administration, the media, and Wall Street. “Art worked extremely hard at people selection and helped build many of the people who now make up his senior executive team,” George says. “As a result, when he took over there wasn’t very much turnover at all.” During Collins’s first six months as CEO, earnings increased 14% before nonrecurring charges on a 13% boost in revenues.
Randall Tobias faced a somewhat different situation when he was brought in to run Eli Lilly in 1993. The drug company’s board, of which he was a member, had unexpectedly ousted his predecessor after 18 months over differences in strategy. The outside directors offered the job to Tobias, who’d been considering early retirement from AT&T, where he was a vice chairman. Tobias said he’d need to serve as CEO for at least five years, the time it would take to prepare his successor.
He had already identified Sidney Taurel, Lilly’s chief operating officer, as his candidate for the job. “I think it was clear to all of us on the board that our first choice was to find someone inside the business,” he says. But “I think it was also pretty clear to everybody that Sidney just was not there yet. We concluded that we could ruin a very promising career by putting him into a difficult situation before he was ready.”
Taurel had been an extremely effective chief operating officer, but, says Tobias, he lacked experience in broader governance issues, especially those relating to Washington—Lilly is a drug company, after all—and Wall Street. He hadn’t yet demonstrated that he could handle the human-relations and communication issues that CEOs spend so much of their time on.
“Everybody says this, but it’s hard to understand the difference between being CEO and any other job in the company until you are a CEO,” Tobias says. So, for a year and a half from the time he became chief executive, he kept the board informed about Taurel’s progress, along with that of 15 to 20 other possible contenders. Watching Taurel, Tobias grew convinced that he had the skills the job required, and began to give him more responsibility in areas that were new to him. Gradually Tobias positioned Taurel as his partner.
In December 1997, he told the board that in six months he wanted to announce his intention to retire, and at that time “I’d like for the board to elect Sidney Taurel to the additional position of CEO.” Tobias would give up the CEO’s title but stay on as executive chairman for six months to smooth the transition. Then he’d leave the company in December 1998.
Was the board surprised? “I think if the CEO is having constant conversation with the board,” says Tobias, “it doesn’t become a news bulletin that you walk in one day and say, ‘Hey, I’ve decided that Sidney Taurel is the person who ought to be the next CEO.’” Taurel has been running Lilly solo for two years now and, since the Prozac patent expired in August 2001, has had to weather an 80% decline in the drug’s market share, the worst sales erosion the company has ever known. Third-quarter profits fell by 27% as a result. But Lilly retains its reputation for having the strongest new-product pipeline in the pharmaceutical industry, and sales of its other drugs are healthy.
At GE, Medtronic, and Lilly, boards oversaw the succession process, even though the CEOs took the lead in identifying strong candidates and getting them ready for prime time. That’s as it should be. As a director, you are responsible for ensuring that your company has a well-stocked managerial larder, just as you’re responsible for other governance activities like setting strategy, allocating capital, and overseeing financial reporting.
Moreover, now that the Internet silliness of the last few years has finally calmed down, all workers are less mobile. “Boards can now think about developing their managers internally without fearing that the investment will be wasted because the company will lose its good people,” says Robert Felton, a partner at McKinsey & Co. Some boards, of course, will stick with the same old same old, hiring glamour boys or girls and then paying through the nose to get rid of them a few months later. But what kind of governance is that?
