In a 1999 interview in the academy of management Executive, former CompUSA CEO James Halpin boasted that he'd built a high-performance work culture at the company. His philosophy was that "you should consider your coworker your competition," although he insisted that his employees still had great trust in each other. Halpin turned this philosophy into action during quarterly meetings of his 20 regional managers. He drew a line down the center of the table and had the ten strongest performers sit behind the line. The ten weakest performers sat in front of the line, near the head of the table where the senior team stood, "because they have to listen to everything we've got to say." These regional managers also wore name tags displaying shrink numbers (lost and stolen inventory) for their stores, because Halpin believed that the reaction to those with poor numbers would be, " Look at that guy's shrink. It is terrible compared to the company average. I am not sitting next to him."
These methods probably did drive Halpin's people to work harder, but research suggests that it undermined employee cooperation. I suspect that this cutthroat culture was one reason that Comp-USA got into serious financial trouble shortly after this interview, and Halpin lost his job. If you were one of Halpin's top ten regional managers and someone in the bottom ten called you for help, why would you do it? It could mean that, at the next meeting, you would be sitting on the "losers" side of the line.
Halpin took it further than most, but internal competition makes sense to many managers. You hire the best people and assess them on regular intervals with sound measures. You rank them from best to worst, then oust the worst, shower the best with money, promotions and praise, and devote enough attention to the rest so that they don't leave and keep working hard. Or you breed competition by giving units of the same company virtually the same task, and wait to see which one does it better. The rhetoric of competition is appealing in our individualistic, sports-oriented society, and we revere executives who talk about tough competition, survival of the fittest and the virtues of winning.
Unfortunately, too many companies take this advice too far, forgetting that top organizational performance requires constant information sharing, so people can keep learning from each other and coordinate what they do. When you have an incentive to make your colleagues perform less well than you do, the whole system can suffer.
Indeed, after consulting and speaking to hundreds of managers at hundreds of firms for many years, I'm convinced that dysfunctional internal competition is the most prevalent and potent barrier to organizational action. It's a problem that arises when people within a company or team see each other as enemies.
There are some important things that organizations can do to minimize these problems. First, you need an incentive system that rewards people for supporting, not undermining, cooperation. Lincoln Electric rates employees from best to worst, and there are huge (200 percent to 300 percent) pay differences between the highest- and lowest-rated production workers. Yet, despite late cofounder James Lincoln's words that competition breeds excellence, the annual profit-sharing bonus that workers receive (typically a hefty 50 percent to 60 percent of their yearly income) is based largely on how well they contribute to the success of their coworkers. The key is that a "star" employee at Lincoln isn't someone who wins at the expense of others, but is someone who helps everyone else to succeed.
Another cause of dysfunctional internal competition is the star system, in which there is a huge spread between the best- and worst-paid employees. Companies like General Electric, Lincoln and Microsoft have flourished by showering far more rewards on top performers than on everyone else. But if you look at the empirical record, every shred of evidence shows that such companies are succeeding despite, rather than because of, such pay practices. As long as some cooperation is required to do the work, study after study shows that when organizations enforce a large spread between the best- and worst-paid, overall organizational performance suffers. Sure, one study of trucking firms showed that a large spread was linked to better overall performance, but truck drivers have a minimal need to get along with their coworkers. The opposite result is found everywhere else, including studies of baseball teams, academic institutions and top management teams in high-technology firms and in manufacturing settings.
Matt Bloom's study of 29 professional baseball teams over an eight-year period is especially interesting. Bloom is an Associate Professor at Notre Dame's Mendoza College of Business. After ruling out numerous alternative explanations, he discovered that teams with a wider pay gap (between the best- and worst-paid players) were plagued with substantially worse win-loss records, attendance and media income than teams with less dispersion. Bloom found an interesting twist: Paying top players more seems to improve individual performance, yet the inferior performance of their teams is driven, in large part, by the relatively poor performance of the average players. In other words, while so many management gurus and consultants talk about how important "A" players are, it was weak performance by the (apparently) underappreciated "B" players that explained which teams were winners and which were losers.
But don't get me wrong. Competition isn't always misguided. As the Lincoln case shows, when there is a balance between cooperation and competition, and "winners" are defined as people who help the organization and not simply themselves, competition can be quite constructive. And there's also something to be said for focusing the troops on a common enemy. Indeed, one of the best ways to spark performance is to find an "enemy" outside your company or group and declare war.
A few years ago, I was talking to a group of NASA project managers about these ideas, and when I mentioned the power of a common enemy, they talked about how much they missed the Soviet Union. They complained that now they were fighting over which people and programs in NASA deserved resources. In the good old days, the enemy was the Soviet space programand cooperation among NASA employees was intensive and all about achieving a common goal, winning the race to the moon.
Or consider the case of Herb Kelleher, the longtime CEO of Southwest Airlines. He has been renowned for declaring war on other airlines, from Braniff and American in the early days to Frontier Airlines in recent years. I am hearing lately, though, that there is more internal friction at Southwest than before. I wonder if part of the problem is that the rest of the U.S. airline industry (except for upstart JetBlue) is in so much trouble today that Southwest doesn't have any credible enemies, so its employees are now turning their battle swords on each other.
Consider, too, PeopleSoft, which, as I write this column, faces a hostile takeover bid from Oracle. I've had regular contact with PeopleSoft in recent months because I am serving as an Honorary Fellow at the company this year. I am struck by how hard everyone has been working to help the company fight off the bid and how everyone there has been working together more cooperatively to achieve this. An IT manager I spoke to a few weeks ago told me that the takeover bid, win or lose, has been the best thing that has ever happened to PeopleSoft, that people have only thought about "beating Larry Ellison, not about themselves."
Robert I. Sutton is co-author with Jeffrey Pfeffer of The Knowing-Doing Gap: How Smart Companies Turn Knowledge into Action. He co-leads Stanford University's Center for Work, Technology and Organization. Professor Sutton's next column will appear in February.