January, 2001

Level 5 Leadership
Harvard Business Review

by Jim Collins

In 1971, a seemingly ordinary man named Darwin E. Smith was named chief executive of Kimberly-Clark, a stodgy old paper company whose stock had fallen 36% behind the general market over the previous 20 years. Smith, the company’s mild-mannered in-house lawyer, wasn’t so sure the board had made the right choice—a feeling that was reinforced when a Kimberly-Clark director pulled him aside and reminded him that he lacked some of the qualifications for the position. But CEO he was, and CEO he remained for 20 years.

What a 20 years it was. In that period, Smith created a stunning transformation at Kimberly-Clark, turning it into the leading consumer paper-products company in the world. Under his stewardship, the company beat its rivals Scott Paper and Procter & Gamble. And in doing so, Kimberly-Clark generated cumulative stock returns that were 4.1 times greater than those of the general market, outperforming venerable companies such as Hewlett-Packard, 3M, Coca-Cola, and General Electric.

Smith’s turnaround of Kimberly-Clark is one the best examples in the twentieth century of a leader taking a company from merely good to truly great. And yet few people—even ardent students of business history—have heard of Darwin Smith. He probably would have liked it that way. Smith is a classic example of what we have come to call a Level 5 leader—an individual who blends extreme personal humility with intense professional will. According to our five-year research study, executives who possess this paradoxical combination of traits are catalysts for the statistically rare event of transforming a good company into a great one. (Our research is described in the sidebar “One Question, Five Years, Eleven Companies.”)

The term “Level 5” refers to the highest level in a hierarchy of executive capabilities that we identified during our research. Leaders at the other four levels in the hierarchy can produce high degrees of success, but not enough to elevate companies from mediocrity to sustained excellence. (For details on this concept, see the exhibit “The Level 5 Hierarchy.”) And while Level 5 leadership is not the only requirement for transforming a good company into a great one—other factors include getting the right people on the bus (and the wrong people off the bus) and creating a culture of discipline—our research shows it to be essential. Good-to-great transformations don’t happen without Level 5 leaders at the helm. They just don’t.

Not what you would expect
Our discovery of Level 5 leadership is counterintuitive. Indeed, it is counter-cultural. People generally assume that transforming companies from good to great requires larger-than-life leaders—big personalities like Iacocca, Dunlap, Welch, and Gault, who make headlines and become celebrities.

Compared to those CEOs, Darwin Smith seems to have come from Mars. Shy, unpretentious, even awkward, Smith shunned attention. When a journalist asked him to describe his management style, Smith just stared back at the scribe from the other side of his thick black-rimmed glasses. He was dressed unfashionably, like a farm boy wearing his first J.C. Penney suit. Finally, after a long and uncomfortable silence, he said: “Eccentric.” Needless to say, the Wall Street Journal did not publish a splashy feature on Darwin Smith.

But if you were to consider Smith to be soft or meek, you would be terribly mistaken. His lack of pretense was coupled with a fierce, even stoic, resolve toward life. Smith grew up on an Indiana farm and put himself through night school at Indiana University by working the day shift at International Harvester. One day, he lost a finger on the job. The story goes that he went to class that evening and returned to work the very next day. Eventually, this poor but determined Indiana farm boy earned admission to Harvard Law School.

He showed the same iron will as CEO of Kimberly-Clark. Indeed, two months after Smith became CEO, doctors diagnosed him with nose and throat cancer and told him he had less than a year to live. He duly informed the board of his illness but said he had no plans to die anytime soon. Smith held to his demanding work schedule while commuting weekly from Wisconsin to Houston for radiation therapy. He lived 25 more years, 20 of them as CEO.

Smith’s ferocious resolve was crucial to the rebuilding of Kimberly-Clark, especially when he made the most dramatic decision in the company’s history: sell the mills.

To explain: shortly after he took over, Smith and his team had concluded that the traditional core business—coated paper—was doomed to mediocrity. Its economics were bad and the competition weak. But, they reasoned, if Kimberly-Clark was thrust into the fire of the consumer paper-products business, better economics and world-class competition like Procter & Gamble would force it to achieve greatness or perish.

And so, like the general who burned the boats upon landing on enemy soil, leaving his troops to either succeed or die, Smith announced that Kimberly-Clark would sell its mills—even the namesake mill in Kimberly, Wisconsin. All proceeds would be thrown into the consumer business, with investments in brands like Huggies diapers and Kleenex tissues. The business media called the move stupid, and Wall Street analysts quickly downgraded the stock. But Smith never wavered. Twenty-five years later, Kimberly-Clark owned Scott Paper and beat Procter & Gamble in six of eight product categories. In retirement, Smith reflected on his exceptional performance by saying simply, “I never stopped trying to become qualified for the job.”

Not what we expected either
We’ll look in depth at Level 5 leadership, but first let’s set an important context for our findings: We were not looking for Level 5 or anything like it. Our original question was, “Can a good company become a great one, and if so, how?” In fact, I gave the research teams explicit instructions to downplay the role of top executives in their analyses of this question so we wouldn’t slip into the simplistic “credit the leader” or “blame the leader” thinking that is so common today.

But Level 5 found us. Over the course of the study, research teams kept saying, “We can’t ignore the top executives even if we want to. There is something consistently unusual about them.” I would push back, arguing, “The comparison companies also had leaders. So what’s different here?” Back and forth the debate raged. Finally, as should always be the case, the data won. The executives at companies that went from good to great, and sustained that performance for 15 years or more, were all cut from the same cloth—one remarkably different from that which was the fabric for executives from the comparison companies in our study. It didn’t matter whether the company was in crisis or steady state, consumer or industrial, services or products. It didn’t matter when the transition took place or how big the company. The successful organizations all had a Level 5 leader at the pivotal time of transition.

Furthermore, the absence of Level 5 showed up as a pattern of contrast across the set of comparison companies. The point: Level 5 is an empirical finding, not an ideological one. And that’s important to note, given how much the Level 5 finding contradicts not only conventional wisdom but much of management theory to date. (For more about our findings on good-to-great transformations, see the sidebar “Not by Level 5 Alone.”)

Humility + Will = Level 5
Level 5 leaders are a study in duality: modest and willful, shy and fearless. To quickly grasp this concept, consider Abraham Lincoln, who never let his ego get in the way of his ambition to create an enduring great nation. Author Henry Adams had called him “a quiet, peaceful, shy figure.” But those who thought Lincoln’s understated manner signaled weakness in the man found themselves terribly mistaken—to the scale of 250,000 Confederate and 360,000 Union lives, including Lincoln’s own.

While it might be a stretch to compare the 11 Level 5 CEOs in our research to Lincoln, they did display the same kind of duality. Take Colman M. Mockler, CEO of Gillette from 1975 to 1991. Mockler, who faced down three takeover attempts, was a reserved and gracious man with a gentle, almost patrician manner. Despite epic battles with the raiders—he took on Ronald Perelman twice and the former Coniston Partners once—he never lost his shy, courteous style. At the height of the crisis, he maintained a calm business-as-usual demeanor, dispensing first with ongoing business before turning to the takeover.

And yet, those who mistook Mockler’s outward modesty as a sign of inner weakness found themselves beaten in the end. In one proxy battle, Mockler and other senior executives together called thousands of investors, one by one, to win their votes. Mockler simply would not give in. He chose to fight for the future greatness of Gillette even though he could have pocketed millions by flipping his stock.

Consider the consequences had Mockler capitulated. If a share-flipper had accepted the full 44% price premium offered by Perelman, and then invested those shares in the general market for ten years, he still would have come out 64% behind a shareholder who stayed with Mockler and Gillette. If Mockler had given in, none of us would likely be shaving with Sensor, Lady Sensor, or the Mach III—and hundreds of millions of people would have a more painful battle with daily stubble.

Sadly, Mockler never had the chance to enjoy the full fruits of his efforts. In January 1991, Gillette received an advance copy of Forbes. The cover featured an artist’s rendition of the publicity-shy Mockler standing on a mountaintop, holding a giant razor above his head in a triumphal pose. Walking back to his office, just minutes after seeing this public acknowledgment of his 16 years of struggle, Mockler crumpled to the floor and died from a massive heart attack.

Even if Mockler had known he would die in office, he could not have changed his approach. His placid persona hid an inner intensity, a dedication to making anything he touched the best—not just because of what he would get but because he couldn’t imagine doing it any other way. Mockler could not give up the company to those who would destroy it, anymore than Lincoln could sue for peace and lose forever the chance to build an enduring great nation.

A compelling modesty
The Mockler story illustrates the compelling modesty typical of Level 5 leaders. (For a summary of Level 5 traits, see the exhibit “The Yin and Yang of Level 5.”) Indeed, throughout our interviews with such executives, we were struck by the way they talked about themselves—or rather, didn’t talk about themselves. Oh, they’d go on and on about the company and the contributions of other executives, but they would instinctively deflect discussion about their own role. When pressed to talk about themselves, they’d say things like, “I hope I’m not sounding like a big shot,” or “I don’t think I can take much credit for what happened. We were blessed with marvelous people.” One Level 5 leader even asserted, “There are lots of people in this company who could do my job better than I do.”

By contrast, consider the courtship of personal celebrity by the comparison CEOs. Scott Paper, the comparison company to Kimberly-Clark, hired Al Dunlap as CEO—a man who would tell anyone who would listen (and many who would prefer not to) about his accomplishments. After 19 months atop Scott Paper, Dunlap said in BusinessWeek: “The Scott story will go down in the annals of American business history as one of the most successful, quickest turnarounds ever. It makes other turnarounds pale by comparison.” He personally accrued $100 million for 603 days of work at Scott Paper—or about $165,000 per day—largely by slashing the workforce, halving the R&D budget, and putting the company on growth steroids in preparation for sale. After selling off the company and pocketing his quick millions, Dunlap wrote an autobiography, in which he boastfully dubbed himself “Rambo in pinstripes.” It’s hard to imagine Darwin Smith thinking, “Hey, that Rambo character reminds me of me,” let alone stating it publicly.

Granted, the Scott Paper story is one of the more dramatic in our study, but it’s not an isolated case. In more than two-thirds of the comparison cases, we noted the presence of a gargantuan ego that contributed to the demise or continued mediocrity of the company. We found this pattern particularly strong in the unsustained comparison companies—the companies that would show a shift in performance under a talented yet egocentric Level 4 leader, only to decline in later years.

Lee Iacocca, for example, saved Chrysler from the brink of catastrophe, performing one of the most celebrated (and deservedly so) turnarounds in American business history. The automaker’s stock rose 2.9 times higher than the general market about halfway through his tenure. But then Iacocca diverted his attention to making himself one of the most celebrated CEOs in business history. He appeared regularly on talk shows like the Today Show and Larry King Live, starred in more than 80 commercials, entertained the idea of running for president of the United States, and promoted his autobiography, which sold 7 million copies worldwide. Iacocca’s personal stock soared, but Chrysler’s stock fell 31% below the market in the second half of his tenure.

And once Iacocca had accumulated all the fame and perks, he found it difficult to leave center stage. He postponed his retirement so many times that Chrysler’s insiders began to joke that Iacocca stood for “I Am Chairman of Chrysler Corporation Always.” When he finally retired, he demanded that the board continue to provide a private jet and stock options. Later, he joined forces with noted takeover artist Kirk Kerkorian to launch a hostile bid for Chrysler. (It failed.) Fortunately, Iacocca did make one final brilliant decision: he picked a modest yet determined man—perhaps even a Level 5—as his successor. Bob Eaton rescued Chrysler from its second near-death crisis in a decade and set the foundation for a more enduring corporate transition.

An unwavering resolve
Besides extreme humility, Level 5 leaders also display tremendous professional will. When George Cain became CEO of Abbott Laboratories, it was a drowsy family-controlled business, sitting at the bottom quartile of the pharmaceutical industry, living off its cash cow, Erythromycin. Cain was a typical Level 5 leader in his lack of pretense; he didn’t have the kind of inspiring personality that would galvanize the company. But he had something much more powerful: inspired standards. He could not stand mediocrity in any form and was utterly intolerant of anyone who would accept the idea that good is good enough. For the next 14 years, he relentlessly imposed his will for greatness on Abbott Labs.

One of Cain’s first tasks was to destroy one of the root causes of Abbott’s mediocrity: nepotism. Systematically rebuilding both the board and the executive team with the best people he could find, Cain made it clear that family ties no longer mattered. If you couldn’t become the best executive in the industry, within your span of responsibility, you would lose your paycheck.

Such near-ruthless rebuilding might be expected from an outsider brought in to turn the company around, but Cain was an 18-year veteran insider—and a part of the family, the son of a previous president. Holiday gatherings were probably tense for a few years in the Cain clan—“Sorry I had to fire you. Want another slice of turkey?”—but in the end, family members were pleased with the performance of their stock. Cain had set in motion a profitable growth machine. From its transition date in 1974 to 2000, Abbott created shareholder returns that beat the market 4.5 to 1, outperforming industry superstars Merck and Pfizer by a factor of two.

Another good example of iron-willed Level 5 leadership comes from Charles R. “Cork” Walgreen III, who transformed dowdy Walgreens into a company that outperformed the stock market 16 to 1 from its transition date in 1975 to 2000. After years of dialogue and debate within his executive team about what to do with Walgreens’ food-service operations, this CEO sensed the team had finally reached a watershed: the company’s brightest future lay in convenient drugstores, not in food service. Dan Jorndt, who succeeded Walgreen in 1988, describes what happened next:

“Cork said at one of our planning committee meetings, ‘Okay, now I am going to draw the line in the sand. We are going to be out of the restaurant business completely in five years.’ At the time we had more than 500 restaurants. You could have heard a pin drop. He said, ‘I want to let everybody know the clock is ticking.’ ... Six months later we were at our next planning committee meeting and someone mentioned just in passing that we had only five years to be out of the restaurant business. Cork was not a real vociferous fellow. He sort of tapped on the table and said, ‘Listen, you now have four and a half years. I said you had five years six months ago. Now you’ve got four and a half years.’ Well, that next day things really clicked into gear for winding down our restaurant business. Cork never wavered. He never doubted. He never second-guessed.”

Like Darwin Smith selling the mills at Kimberly-Clark, Walgreen’s decision required stoic resolve. Not that food service was the largest part of the business, although it did add substantial profits to the bottom line. The real problem was more emotional. Walgreens had, after all, invented the malted milkshake, and food service had been a long-standing family tradition dating back to his grandfather. Not only that, some food-service outlets were even named after the CEO—for example, a restaurant chain named Corky’s. But no matter, if Walgreen had to fly in the face of long-standing family tradition in order to refocus on the one arena where Walgreens could be the best in the world, (convenient drugstores) and terminate everything else that would not produce great results—then Cork would do it. Quietly, doggedly, simply.

One final, yet critical, note on our findings about Level 5. Because Level 5 leaders have ambition not for themselves, but for their companies, they routinely select superb successors. Level 5 leaders want to see their companies become even more successful in the next generation, comfortable with the idea that most people won’t even know that the roots of that success trace back to them. As one Level 5 CEO said, “I want to look from my porch, see the company as one of the great companies in the world someday, and be able to say, ‘I used to work there.’ ” By contrast, Level 4 leaders often fail to set up the company for enduring success—after all, what better testament to your own personal greatness than that the place falls apart after you leave?

In more than three-quarters of the comparison companies, we found executives who set up their successors for failure, who chose weak successors, or both. Consider the case of Rubbermaid, an unsustained comparison company that grew from obscurity to become Fortune’s number one most admired company—and then, just as quickly, disintegrated into such sorry shape that it had to be acquired by Newell.

The architect of this remarkable story was a charismatic and brilliant leader named Stanley C. Gault, whose name became synonymous in the late 1980s with the success of the company. Across the 312 articles collected by our research team about Rubbermaid, Gault comes through as a hard-driving, egocentric executive. In one article, he responds to the accusation of being a tyrant with the statement, “Yes, but I’m a sincere tyrant.” In another, drawn directly from his own comments on leading change, the word “I” appears 44 times, while the word “we” appears only 16 times. Of course, Gault had every reason to be proud of his executive success: Rubbermaid generated 40 consecutive quarters of earnings growth under his leadership—an impressive performance, to be sure, and one that deserves respect.

But Gault did not leave behind a company that would be great without him. His chosen successor lasted a year on the job and the next in line faced a management team so shallow that he had to temporarily shoulder four jobs while scrambling to identify a new number-two executive. Gault’s successors struggled not only with a management void but also with strategic voids that would eventually bring the company to its knees.

Of course, you might say—as one Fortune article did—that the fact Rubbermaid fell apart after Gault left just proves his greatness as a leader. Gault was a tremendous Level 4 leader, perhaps one of the best in the last 50 years. But he was not a Level 5, and that is one crucial reason why Rubbermaid went from good to great for a brief, shining moment and then just as quickly went from great to irrelevant.

The window and the mirror
As part of our research, we interviewed Alan L. Wurtzel, the Level 5 leader responsible for turning Circuit City from a ramshackle company on the edge of bankruptcy into one of America’s most successful electronics retailers. In the 15 years after its transition date in 1982, Circuit City outperformed the market 18.5 to 1.

We asked Wurtzel to list the top five factors in his company’s transformation, ranked by importance. His number one factor? Luck. “We were in a great industry, with the wind at our backs.” But wait a minute, we retorted, Silo—your comparison company—was in the same industry, with the same wind, and bigger sails. The conversation went back and forth, with Wurtzel refusing to take much credit for the transition, preferring to attribute it largely to just being in the right place at the right time. Later, when we asked him to discuss the factors that can sustain a good-to-great transformation, he said, “The first thing that comes to mind is luck. I was lucky to find the right successor.”

Luck. What an odd factor to talk about. Yet the Level 5 leaders we identified invoked it frequently. We asked an executive at the steel company Nucor why the company had such a remarkable track record of making good decisions. His response? “I guess we were just lucky.” Joseph Cullman III, the Level 5 CEO of Philip Morris, flat-out refused to take credit for his company’s success, citing his good fortune to have great colleagues, successors, and predecessors. Even the book he wrote about his career—a book he wrote at the urging of his colleagues and which he never intended to distribute widely outside the company—had the unusual title I’m a Lucky Guy.

At first, we were puzzled by the Level 5 leaders’ emphasis on good luck. After all, there is no evidence that the companies that had progressed from good to great were blessed with more good luck (or more bad luck, for that matter) than the comparison companies. But then we began to notice an interesting pattern in the executives at the comparison companies: They often blamed their situations on bad luck, bemoaning the difficulties of the environment they faced.

Compare Bethlehem Steel and Nucor, for example. Both steel companies operated with hard to differentiate products, and both faced a competitive challenge from cheap imported steel. Both companies paid significantly higher wages to workers than most of their foreign competitors. And yet executives at the two companies held completely different views of the same environment. Bethlehem Steel’s CEO summed up the company’s problems in 1983 by blaming the imports: “Our first, second, and third problems are imports.” Meanwhile, Ken Iverson and his crew at Nucor saw the imports as a blessing: “Aren’t we lucky; steel is heavy, and they have to ship it all the way across the ocean, giving us a huge advantage.” Indeed, Iverson saw the first, second, and third problems facing the American steel industry not in imports but in management. He even went so far as to speak out publicly against government protection against imports, telling a gathering of stunned steel executives in 1977 that the real problems facing the industry lay in the fact that management had failed to keep pace with technology.

The emphasis on luck turns out to be part of a broader pattern that we came to call the window and the mirror. Level 5 leaders, inherently humble, look out the window to apportion credit—even undue credit—to factors outside themselves. If they can’t find a specific person or event to give credit to, they credit good luck. At the same time, they look in the mirror to assign responsibility, never citing bad luck or external factors when things go poorly. Conversely, the comparison executives frequently did the opposite. They looked out the window for factors to blame, but preened in the mirror to credit themselves when things went well.

The funny thing about the window-and-mirror concept is that it does not reflect objective reality. According to our research, the Level 5 leaders did make a difference. They were responsible for their companies’ transformations. But the Level 5 leaders would never admit that. We can’t climb inside their heads and assess whether they deeply believed what they saw in the window and the mirror. But it doesn’t really matter, because they acted as if they believed it, and they acted with such consistency that it produced exceptional results.

Born or bred?
Not long ago, I shared the Level 5 finding with a gathering of senior executives. A woman who had recently become chief executive of her company raised her hand. “I believe what you’ve told us about Level 5 leadership,” she said, “but I’m disturbed because I know that I’m not there yet, and maybe I never will be. Part of the reason I got this job is because of my strong ego. Are you telling me that I can’t make my company great if I’m not Level 5?”

“Let me return to the data,” I responded. “Of 1,435 companies that appeared on the Fortune 500 since 1965, only 11 made it into our study. In those 11, all of them had Level 5 leaders in key positions, including the CEO role, at the pivotal time of transition. Now, to reiterate, we’re not saying that Level 5 is the only element required for the move from good to great, but it appears to be essential.”

She sat there, quiet for a moment, and you could guess what many people in the room were thinking. Finally, she raised her hand again. “Can you learn to become Level 5?” I still do not know the answer to that question. Our research, frankly, did not delve into how Level 5 leaders come to be, nor did we attempt to explain or codify the nature of their emotional lives. We speculated on the unique psychology of Level 5 leaders. Were they “guilty” of displacement—shifting their own raw ambition onto something other than themselves? Were they sublimating their egos for dark and complex reasons rooted in childhood trauma? Who knows? And perhaps more important, do the psychological roots of Level 5 leadership matter any more than do the roots of charisma or intelligence? The question remains: Can Level 5 be developed?

My preliminary hypothesis is that there are two categories of people: those who don’t have the Level 5 seed within them, and those who do. The first category consists of people who could never in a million years bring themselves to subjugate their egoistic needs to the greater ambition of something larger and more lasting than themselves. For those people, work will always be first and foremost about what they get—fame, fortune, adulation, power, whatever—not what they build, create, and contribute. The great irony is that the animus and personal ambition that often drives people to become a Level 4 leader stands at odds with the humility required to rise to Level 5. When you combine that irony with the fact that boards of directors frequently operate under the false belief that you need a larger-than-life, egocentric leader to make a company great, you can quickly see why Level 5 rarely appears at the top of our institutions. We keep putting people in positions of power who lack the seed to become Level 5, and that is one key reason why there are so few companies that make a sustained and verifiable shift from good to great.

The second category consists of people who could evolve to Level 5; the capability resides within them, perhaps buried or ignored or simply nascent. Under the right circumstances—with self-reflection, a mentor, loving parents, a significant life experience, or any number of other factors—the seed can begin to develop. Some of the Level 5 leaders in our study had significant life experiences that might have sparked development of the seed. Darwin Smith fully blossomed as a Level 5 after his near-death experience with cancer. Joe Cullman was profoundly affected by his World War II experiences, particularly the last-minute change of orders that took him off a doomed ship on which he surely would have died; he considered the next 60-odd years a great gift. A strong religious belief or conversion might also nurture the seed. Colman Mockler, for example, converted to evangelical Christianity while getting his MBA at Harvard, and later, according to the book Cutting Edge, became a prime mover in a group of Boston business executives that met frequently over breakfast to discuss the carryover of religious values to corporate life.

We would love to be able to give you a list of steps for getting to Level 5—other than contracting cancer, going through a religious conversion, or getting different parents—but we have no solid research data that would support a credible list. Our research exposed Level 5 as a key component inside the black box of what it takes to shift a company from good to great. Yet inside that black box is another—the inner development of a person to Level 5. We could speculate on what might be inside that inner box, but it would mostly be just that, speculation.

In short, Level 5 is a very satisfying idea, a truthful idea, a powerful idea and, to go from good to great, very likely an essential idea. But to provide ”10 steps to Level 5 leadership” would trivialize the concept.

My best advice, based on the research, is to practice the other good-to-great disciplines we discovered. Since we found a tight symbiotic relationship between each of the other findings and Level 5, we suspect that conscientiously trying to lead using the other disciplines can help you move in the right direction. There’s no guarantee that doing so will turn you into a full-fledged Level 5 leader, but it gives you a tangible place to begin, especially if you have the seed within.

We cannot say for sure what percentage of people have the seed within, nor how many of those can nurture it enough to become Level 5. Even those of us on the research team who identified Level 5 do not know whether we will succeed in evolving to its heights. And yet all of us who worked on the finding have been inspired by the idea of trying to move toward Level 5. Darwin Smith, Colman Mockler, Alan Wurtzel, and all the other Level 5 leaders we learned about have become role models for us. Whether or not we make it all the way to Level 5, it is worth the effort. For like all basic truths about what is best in human beings, when we catch a glimpse of that truth, we know that our own lives and all that we touch will be the better for the effort.