There is this widespread idea, made popular by McKinsey & Co., Fast Company, and others, that we are in a "war for talent." In an intellectual capital world, talent is what matters, and we are told that the companies that will win in the competitive arena are those that are the best at locating, assessing, recruiting, and keeping the most talented people. A typical statement of this position is this opening sentence from an article entitled "The Great Talent Caper" in the September, 2000 issue of Fast Company, the enormously successful business magazine for the new economy: "It's hard to argue with the idea that the company with the best talent wins."

Actually, it's quite easy to argue with that seemingly common-sense statement, while still granting the importance of intellectual capital and knowledge work in today's economy. As John Chambers, CEO of Cisco Systems, has noted, great teams outperform collections of individuals¯¯even when the individuals are more talented. Even in professional sports, the teams with the best talent, often nicely proxied by the highest salaries, don't always (or even usually) have the best records or invariably win championships. In business and nonprofit organizations, characterized by interdependence among individuals so that productivity is affected not only by one individual's skills and abilities but also by the capabilities and actions of others, individual talent matters even less in determining organizational success.

As W. Edwards Deming and the quality movement pointed out a long time ago¯¯a lesson that we clearly need to relearn¯¯what is important is not so much individual motivation or ability but the attributes of the system in which the person works. Some organizations have systems that bring out the best in their people, while others, filled with talented, motivated, hard working people, have practices and policies that interfere with the ability of these individuals to do their best and to make a difference. For example, does anyone really think that United Airlines has less capable or talented people than Southwest Airlines, even though United has much poorer service as assessed by on-time performance, lost bags, and customer complaints? To take another case, would the long-standing and pervasive business problems of Xerox Corp., a company that has been the source of business world-changing inventions such as the first personal computer, word processing software, and local area networks, be solved if Xerox somehow got more talented, brighter people?

It's bad enough that fighting the "war for talent" has companies fighting the wrong war, often using the wrong methods. But there is an even worse problem, namely the consequences that are unleashed by even waging the talent war in the first place. In this article, I describe the various organizational processes and dynamics that are frequently unleashed when companies adopt the "war for talent" mind-set. Considering these social psychological processes helps make the argument that not only should your company not necessarily try to win the war for talent, even adopting this image as a management metaphor can be quite hazardous to your organization's health.

What happens in a war for talent? There is: An invariable emphasis on individual performance (rewarding the individual stars), thereby diminishing teamwork, creating destructive internal competition, and retarding learning and the spread of best practices inside the company; A tendency to glorify the talents of those outside the company and downplay the skills and abilities of insiders, leading to a loss of motivation on the part of those inside the firm and to their turnover (thereby ensuring that the recruiting challenge will be even greater as the company tries to replace those it has inadvertently sent packing); The creation of a self-fulfilling prophecy where those labeled as less able become less able because they are asked to do less, given fewer resources, training, and mentoring, and become discouraged¯¯in the process ensuring that the organization has way too many people who are in the process of dropping out of the competitive fray; A de-emphasis on fixing the systemic, cultural, and business process issues that are invariably much more important for enhancing performance, as the company seeks success solely through getting the right people in the door; And finally, the development of an elitist, arrogant attitude¯¯once you have successfully competed in the war for talent, you have the best people¯¯an attitude that makes building a wise organization almost impossible. In wise organizations, people know what they know and they know what they don't know. Companies that think they are winning the war for talent think they are so full of smart people that they know everything!

Let's consider each of these processes and their consequences for the company.

One of the assumptions of the "talent" language is that there are individual stars. It only makes sense, then, to provide these stars differentiated rewards that recognize their stardom. Indeed, pay for performance is a common recommendation found in the literature on attracting and retaining talent. The assumption is that if you don't pay your stars inordinately well, someone else will and you'll lose them. Seems logical, but like much conventional wisdom, it ignores some important dynamics and facts.

Let's begin with one fact on which virtually everyone agrees¯¯sharing best practices and knowledge inside organizations is difficult and frequently not very successfully done. For instance, a study of oil refineries by the consulting firm Solomon Associates found that in a statistical analysis predicting refinery performance (uptime and maintenance costs), there was no effect of company on the results. What this means is that there was as much variation in performance across refineries within a single oil company as there was across refineries owned by different companies. The absence of a company effect on refinery performance is prima facie evidence for the absence of intraorganizational learning.

Nor is this case unique. A study of 42 food manufacturing plants in a single company doing essentially the same task found a difference in performance of 300% between the best and worst performing plant. Another study of a multinational food producer operating in seven countries and producing the same products using the same machines found a performance difference of 112% between the best and worst performing plant. An intensive study of an effort to make a Hewlett¯Packard Co. manufacturing unit more effective also reported that opportunities to share innovative process technologies or other sources of competitive advantage were overlooked. Indeed, Carla O'Dell and Jackson Grayson have written a book detailing the problems of spreading better practices and knowledge inside companies.

Why is it so hard to share best practices and knowledge? One of the answers must surely be the internal competition set up by the zero-sum reward systems that are part and parcel of the war for talent. Differentiated pay means that we want to pay the best more and the worst less. So what are the behavioral implications of this? If you are a plant manager having problems in your performance, are you going to ask others for help? Not likely, because when you do, you signal publicly to everyone in the company that you are not doing as well as they are. And why would you do that in a competition for salary and status¯¯in a competition to see who has the most "talent?" Moreover, even if someone has the audacity to ask for help, why would anyone else offer such help? Why would I help a competitor? We are competing for the zero-sum rewards of promotions, recognition as a winner, and raises parceled out only to those who excel. Self-interest dictates not being very helpful or forthcoming to those with whom someone competes for those rewards. Tom Lasorda, in charge of diffusing lean manufacturing when he worked for General Motors Corp., was quite articulate about how the internal competition inside GM hindered plant managers' learning from each other, in the process costing the company billions of dollars. Ironically, even as companies spend hundreds of millions of dollars on technologies to collect and distribute knowledge, they do things to create a culture in which knowledge sharing is unlikely to occur.

Logically, if internal competition retards the spread of best practices across a company, leaders who are concerned with improving internal knowledge transfer and subsequent organizational performance should try and stamp out excessive internal competition. And that's just what our research has found. Shortly after taking over IBM Corp., Lou Gerstner issued a number of edicts designed to ensure cooperation and stop internal competition that was hindering the performance of the company. Chris Galvin at Motorola Inc. has recently done the same thing, as did John Pizzey at Alcoa Inc.'s smelters. Guidant Cardiovascular some time ago began providing everyone the same percentage bonus, based on the division's achieving revenue, profit, and technological development goals. According to Peter McInnes, a vice president at Guidant, the company does this to encourage teamwork and to encourage people to focus on business results instead of maneuvering to be on the right project or work on the right team. Willamette Industries, by any measure the most successful integrated forest products company, does not pay short-term performance bonuses to anyone in the company. It certainly does not pay annual bonuses to division managers who need to cooperate rather than fight over transfer prices in this integrated wood products company. There are many other examples of companies that have decided excessive internal competition is getting in the way of achieving business results and have implemented management practices designed to curtail, rather than exacerbate, internal rivalry.

The point is that the emphasis on the individual rather than the team or the company is almost an inevitable outcome of a war for talent mind set. Not only are such individual, zero-sum reward practices encouraged by virtually all those who have written about talent wars, but even if they weren't, rewarding the individual is what logically follows from the belief that is individuals who make the difference.

Even though the talent-war mentality has led to an emphasis on rewarding "the best" to attract and retain those people, there is little evidence that simply paying people more is the most critical factor in recruiting or particularly in retaining employees. Most surveys show that money is not the most important reason why people take or leave jobs. In fact, in some surveys money is not even in the top ten. And money is the most available of all rewards¯¯any organization can offer it. That is why years ago Tandem Computers didn't even tell people while they were being recruited what their precise salaries would be. If the person asked, the statement that Tandem paid good, competitive salaries would be offered in return. If the person insisted on knowing the precise salary and negotiating over it, they would not be offered the job. Tandem understood something quite simple but profound: people who come for money leave for money. Why play that game? SAS Institute, the largest privately owned software company in the world¯¯with turnover less than four percent¯¯doesn't win the retention game by paying the best, but by treating its people the best. There is lesson there for us all.

In the search to find the best people, there is a tendency to see those people as existing primarily, although not exclusively, outside of the organization. Although theories of in-group favoritism suggest that people tend to like and identify more with those with whom they share a social identity, such as an organizational affiliation, outsiders have the advantages of mystery and scarcity value. In other words, there is more than a little truth to the adage that familiarity breeds contempt.

Why? First of all, there is the effect of behavioral commitment. Insiders are already there, coming to work and immediately available. Recruiting or even looking for someone from outside requires more effort. Initially, there is the effort of searching, hiring a search firm, running advertisements, tapping social networks, posting jobs on-line, and reviewing resumes. One way to justify and make sense of that extra effort is to convince oneself that the people uncovered through this effortful activity are, in fact, worth the effort¯¯that they are really better than those immediately at hand.

Once an outside candidate has been found, then comes the effort of interviewing and recruiting that person. Again, the expenditure of effort elicits a commitment response, in which there is tendency to justify the effort by thinking the person hired is better than those inside, for whom no such comparable effort has been recently expended.

Second and somewhat related, there is the effect of scarcity. As the play Romeo and Juliet illustrates, we always want what we can't or don't have. Research has shown, for instance, that cookies are rated as tasting better if there are fewer of them on the plate. Outsiders are scarcer and less available than the insiders working for the firm. Attracting them is invariably an uncertain process¯¯will they come, or won't they? The relative scarcity of outsiders means that they will probably be valued more highly simply because of their comparative unavailability.

Finally, there is the mystery and allure of the outsider. Someone who is not known as well, but who has a reputation for being an outstanding performer, looks larger than life. We don't see the inevitable mistakes, the efforts required to achieve that performance, the hard work, and the fact that the person is a person pretty much like everyone else. Once inside the organization, the individual's foibles and faults are more readily apparent. Moreover, once inside the organization the person becomes a competitor for status, and a tendency to derogate the individual's abilities in this status competition can emerge.

There are many examples of the process just described. In the 1980s, Japanese management practices were venerated by many American companies, and Japanese competitors were seen as larger than life. Ironically, one of those management practices that received so much acclaim and attention was total quality management, a management technique largely developed by an American, W. Edwards Deming, who was ignored until the practices he advocated were adopted by others outside the U.S. Or as another example, consider the many mergers that don't fulfill their expectations. Some of the reason for this is the postmerger integration process, but another part of the story is that the acquired firm was probably not very realistically appraised in the first place. Moreover, once inside the firm, those from the acquired organization are seen as less competent than they really are.

We saw this process in full force when Fresh Choice Inc., a publicly traded salad buffet restaurant chain headquartered in California, purchased Zoopa, a four-unit subsidiary of Restaurants Unlimited, located in Seattle. While they were competitors, Fresh Choice admired and even tried to copy Zoopa's recipes, look and feel, and service orientation. Once the merger was completed, Fresh Choice leaders told us that the Zoopa store managers weren't actually that great and that the acquisition had been mostly done for real estate, not organizational learning. The turnover in the Zoopa units increased after the merger, and much of the knowledge acquired walked out the door. Once inside Fresh Choice, Zoopa people looked neither so special nor so valuable as they did when they were outsiders.

The war-for-talent imagery has consistently emphasized identifying the top ten percent, not just for hiring but for retaining and cultivating once inside the company. Indeed, one of the suggestions coming out of some of the war-for-talent consulting goes like this: identify the top ten and the bottom ten percent. The top ten percent of your people should be lavished with rewards, interesting job assignments, fast-track opportunities, and special training and mentoring. The bottom ten percent should be either removed from the organization or helped to improve. Implicitly, such recommendations suggest ignoring everyone else.

There are two problems with this course of action. First, there is the question of whether or not there really are two (or three) kinds of people in your company¯¯the stars, the dogs, and everyone else. To presume that some people are better than others presumes that there are some reasonably stable attributes, such as talent and drive, which differentiate among people, and that such traits are largely not amenable to being changed. That's why the war for talent has such an emphasis on selection and retention¯¯there are better and worse people, and you'd better get more of the first kind and less of the second. But is this true?

Some American readers will recognize the name of Steve Young, the now-retired quarterback for the San Francisco 49ers, considered to be one of the better quarterbacks in professional football history. Fewer people may also remember that Young's career began outside of the National Football League, because he was deemed not good enough. A similar story holds for Kurt Warner, the quarterback who led the St. Louis Rams to the Super Bowl victory in 2000. More systematically, a study of professional baseball players in the early 1990s revealed that better managers were able to not only get their teams to perform better, but that players under those managers performed better than might have been expected given their lifetime records. And this is for a sport, baseball, in which there is relatively little interdependence, and where natural ability would appear to play an important role. The point is that although there are certainly differences across people, such differences are not invariably related to their performance, because people can perform above or below their natural level depending on a myriad of factors.

One of the most important of these factors affecting performance is the expectation for performance. And that's the second problem¯¯the very labeling of people will affect their performance. Labeling only a few as stars will cause the majority to perform way below their potential.

The self-fulfilling prophecy is one of the oldest and most established principles for understanding organizational behavior. Simply put, the self-fulfilling prophecy holds that high expectations increase performance, and low expectations will decrease performance. This effect has been found in early studies of intelligence and performance in the classroom, where measurable changes in intelligence quotient (IQ) were observed depending on the expectations for children's performance. It has been found in studies of performance in the Israeli defense forces, and in studies of sales force performance. There is, in fact, a vast literature on the effects of expectations on behavior.

Why might such effects be observed? For some fairly straightforward reasons. If a person confronts low expectations, one rational (and observed) response is something called "defensive effort"¯¯fancy ways of saying that people don't try very hard. If you aren't going to succeed anyway, why expend effort that will just wind up being wasted? Other things being equal, if you don't expend as much effort, you probably won't do as well. Conversely, if you think you have a good chance at success, you are quite likely to try harder and, as a consequence of this increased effort, probably do better. Another reason is anxiety. Most people don't find the prospect of failing very comforting, and indeed, failure or the prospect of failure can produce anxiety. As anyone who has ever taken a test can attest, beyond some level anxiety itself interferes with performance. To the extent that expectations of failure produce anxiety, that stress can itself induce poorer performance.

A third factor is that the resources you get depend on what people expect from you. Students labeled as not as bright tend to get less guidance and help from teachers. Bosses are less likely to spend time with people who aren't too good, saving their precious time for coaching and mentoring those who have a real chance of benefitting. Who is going to get sent to training and given more challenging job assignments, those that are labeled as stars or the others? These resource allocation choices have real consequences for performance. People who receive less coaching, mentoring, training, and fewer challenging job assignments will¯¯other things being equal¯¯learn less and be less able to perform at a higher level. In this way, labeling people produces an allocation of resources, including time and attention, that almost guarantees the initial labels will come to be true.

In a company obsessed with fighting the war for talent, such labeling will go on with a vengeance, as the company will be consumed with figuring out who is the best and who are the rest. But that very labeling process will produce many discouraged people¯¯those not accorded the most favorable labels. Two consequences can result. Either those labeled as less than star-like will leave, in which case the company will have to recruit even more people to replace them, or else they will simply give up and retire in place, depriving the company of their effort and ideas.

What happens when you don't fight the war for talent, thereby giving up on most of your people (only 10% of the people can be in the top 10%)? Consider the case of The Men's Wearhouse. The company operates in the difficult industry of selling tailored men's clothing, an industry that is not only competitive, but one that is declining. Nonetheless, the compound annual growth rate achieved by The Men's Wearhouse over the past five years has been almost 30% in sales and more than 25% in earnings per share. The Men's Wearhouse Inc. has an interesting operating philosophy. Its founder and chairman, George Zimmer, has stated that the company is in the people business, not the suit business. Charlie Bresler, in charge of human resources, has the interesting title of executive vice president for human development. The company aspires to help people become better than they (or probably anyone else) ever thought they could be¯¯a difficult challenge in retailing, a low-wage industry that has typically not attracted the best people. But the company has succeeded to an astonishing degree by giving people second, and even third chances, even when they have put a deposit in their pockets for a few days or ripped off a pair of socks. The company's success comes in large measure from its emphasis on training. Training not only imparts knowledge of clothing and sales techniques. Training enhances people's self-esteem, by signaling that the organization values them and is interested in investing in them. By helping people unleash their full potential, The Men's Wearhouse has lowered its turnover, has among the lowest losses because of theft in the retail industry even without electronic security, and has developed wardrobe consultants who provide a service experience that permits the company to achieve outstanding profit margins. If the company had adopted the talent-war mentality, it would have been finished before it even began.

The scarcest resource in most organizations is not money; it is time and attention. There are too many problems and issues competing for managers' time and too many things to think about. It is obvious that time spent on one set of issues or initiatives are time (and effort) that cannot be devoted to other things. Fighting the war for talent focuses the company on first grading or ranking and then seeking to select and retain the best people. It focuses the company, in other words, on individuals¯¯not groups, teams, or the entire company¯¯and it focuses on these individuals with the presumption that there are better and worse ones, and there's not much to be done to change anyone. It causes companies to ignore or downplay the importance of intervening to build cultures and systems that bring out the best in everyone.

The American automobile industry illustrates this problem quite nicely. In the 1980s, General Motors concluded that the problem with its automobile plants was its people. Many of them had worked in the industry for years and were, in the opinion of the company, old, tired, and burnt out. Most were unionized, which, the company thought, put them in an adversarial relationship. Many were not highly educated, therefore lacking the skills to help GM compete in an increasingly quality-conscious, technologically complex industry. GM's response was to invest in factory automation to eliminate as many of these autoworkers as possible. The outcome of that process was, as described by a number of observers, to leave the company with the highest fixed costs in the industry and with machines that didn't really work the way they were supposed to.

By contrast, Toyota Motor Corp. and a number of the other Japanese automobile manufacturers took a different approach, nicely illustrated by what happened in the Toyota-GM joint venture in Fremont, California, New United Motors Manufacturing, Inc. (NUMMI). At that plant, formerly operated by General Motors, Toyota neither tried to replace the people (85% of those who worked at NUMMI when it started production in 1985 had worked for the former GM plant, well known for absenteeism, poor quality, and drug abuse and alcoholism), nor did it substitute machines for employees. Rather, the company did those things that are well known as being associated with high performance or high commitment management practices: Established a policy of mutual commitment and employment security. NUMMI employees were promised that they would be laid off only under extreme conditions¯¯the viability of the enterprise was at stake. Instead, when demand was slack, employees used the time to rehabilitate and upgrade the plant and to take courses that enhanced their skills in quality, statistical process control, working in teams, and so forth. As Lincoln Electric Co. and its executives recognized a long time ago, no one will voluntarily contribute ideas for enhancing productivity if the result is that they or their colleagues will lose their jobs. In return for working cooperatively to enhance productivity, people in the NUMMI plant were promised security and were treated as real partners in the enterprise. Selective hiring. Although under the terms of the UAW contract people formerly employed at the GM plant that had been closed were promised recall rights, these individuals nonetheless went through interviews and screening. Even though few were actually culled during this process, the selectivity provided an opportunity for some who didn't like or agree with the new culture to self-select out. More important, it gave those who came back into the plant a feeling that the company cared about who they were, and a feeling of pride for having gone through the selection process. High pay. NUMMI offered the highest wage in the industry at that time. The deal was high pay and security in return for a new spirit of cooperation. Information sharing. At the NUMMI plant (and subsequently at Saturn which learned a lot from the experience of NUMMI), information on production, quality, and productivity was widely shared and readily available. People who were expected to take some responsibility for the operation of the plant were given the information necessary to do so. Decentralization of decision making. NUMMI really excelled at giving front-line people, who had years of experience and the wisdom and insight that comes from that experience, responsibility for improving performance. Virtually all industrial engineers were removed from the plant. Front-line people were trained in analyzing work methods and processes and were given the time and the freedom to implement productivity and quality-enhancing changes in the production process. Layers of management were eliminated, one reason why NUMMI was able to cut its costs. The idea of self-managing teams was one of the organizing principles that governed the operations of the plant. An egalitarian culture. When Toyota assumed control of the management of NUMMI, it abolished the executive dining room and reserved parking spaces. Everyone, including the president of NUMMI, wore the same smock, ate in the same place, and socialized with each other. The idea, conveyed symbolically, was that all employees shared the responsibility and the obligation to make the organization successful. An emphasis on training and skill development. As already noted, NUMMI expected its front-line people to contribute their ideas to making things better. They were given training in how to analyze jobs and work processes and when there was a downturn in demand, the slack permitted even more extensive training to occur.

The Toyota production system has been remarkably effective because it is a system that permits regular people to achieve world-class results on a consistent basis over long periods of time. It is a system of organizing that recognizes the connection between the social and the technical aspects of organization and that puts in place a culture that helps to ensure the motivation and performance of the employees. Why don't other automobile companies copy this system, acknowledged to be superior? One reason is that although other companies would like to implement flexible manufacturing, time and attention is diverted to other activities such as fighting with the union, implementing complicated robotics, engaging in internal politics and competing with each other. Time is indeed a scarce resource, and time spent fighting a war for talent is time diverted from building a culture and set of management practices that permits everyone to perform as if he or she were in the top 10%.

A corollary of the war for talent imagery is that what companies want is "smart" people. However, companies may be much better served by having "wise" people. As originally defined by Plato, wisdom is the attitude of knowing what you know and knowing also what you don't know. The attitude of wisdom, as characterized by a number of scholars, represents a midpoint between arrogance and insecurity. Having an attitude of wisdom permits an organization to take action even as it doubts what it knows, so that it continues to learn even as it acts.

IDEO Product Development, an award-winning product design firm headquartered in Palo Alto, nicely exemplifies a wise organization. When David Kelley, the CEO, announced a reorganization a while ago, he stated that all he knew was that the new structure was the best the company could come up with at the time, and that it was wrong. Not only does that statement illustrate that it is all right to make mistakes at the company, it shows that it is all right to act without having the complete right answer. Another motto at IDEO is that enlightened trial and error outperforms the planning of flawless intellects. This cultural norm means that people learn from doing, so the only mistake is not to take action based on what is known at the time. It is a norm that facilitates rapid prototyping, important not only for product design but for organizational learning more generally. A third aphorism is that the company believes in failing early and failing often¯¯which is better than failing once, failing at the end, and failing big.

In each of these statements, you see the attitude of wisdom. Each reflects having enough confidence to try something, be it a new structure or a prototype, while keeping open the idea that no one in the company or the company itself knows everything, so it needs to continue learning even as it does things.

We are surrounded by examples of companies that have gotten into trouble by being too arrogant. The recent books on the Microsoft Corp. antitrust trial show that Microsoft could have reached a settlement in the trial, but it was so sure of itself it delayed entering into serious negotiations. The company's arrogance during the trial offended the trial judge. Arrogance, however, may be a natural byproduct of operating in the software industry¯¯an industry that makes a practice of announcing "vaporware" and selling prototypes into the marketplace, letting customers find the bugs, and then selling upgrades that fix the problems the company shipped in the first place (this is called versions of the product).

To listen to your customers, you need to think you don't know everything. James Goodnight, a cofounder and CEO of SAS Institute, the largest privately owned software company in the world, is fond of saying that he is not a technological visionary and that he can't predict the future. Therefore, what SAS Institute must do is to listen to its customers, give them the software that they want and need, and be ready and able to change as market demands and conditions change. It is a formula that has produced more than 23 consecutive years of double-digit growth and a 98% license renewal rate. John Chambers of Cisco Systems is another person in another company that shows an attitude of wisdom. Cisco has no technology religion. The company is fond of saying it is agnostic with respect to specific technologies. Rather, it listens to the market and provides the technology that the market demands. It is not a company too proud to go out and purchase the technology it needs. Indeed, Cisco's ability to not only acquire companies but retain their people is one of its important sources of success. Chambers and other senior managers spend most of their time with customers and employees, listening. Watch John Chambers at a conference. In the world of high technology, typically populated by people with a surfeit of ego and self-confidence, he spends most of his time listening and asking questions. If you think about it, that is the best way to learn, and learning is important in a rapidly changing market environment.

What does this have to do with the "war for talent?" Just this¯¯if you hire the best people who think (or even know) they are the best, how likely are they to be willing to listen and learn? How likely are they to treat others not as "smart" as they are with respect, as opposed to the contempt more often seen? One of the geniuses of Toyota was recognizing that the people on the assembly line actually knew something about automobile assembly, regardless of formal degrees and the ability to talk smoothly. One of the marks of the wisdom of Jeff Hawkins, cofounder of Handspring, is his interest in watching people who use paper instead of personal digital assistants so he can see what they do, how they do it, and why. It would be easy for a technological guru (and billionaire) to think that those that don't use the latest technology are not only uncool but are not too smart. Hawkins wants to learn from them. The ability to learn from others depends importantly on not thinking yourself or your organization so superior to them that you have nothing to learn. Maybe that means not thinking that you have the smartest people, just the best listeners.

Is it possible to wage the war for talent and not fall into these problems and pitfalls? Of course. But it is not likely, nor is it easy. Any theory, implicit or explicit, causes us to see some things in specific ways and to ignore other things. The war for talent has embedded within it a theory that holds: 1) that organizational performance is essentially the aggregation of a bunch of individual performances (which is why if you get people who do well individually you can win the competitive battle); and 2) people are essentially unchanged, at least by the time they are adults in the work force, in terms of their abilities and capabilities, which is why selecting (and keeping) the right people is so crucial.

Once explained in this way, we can see that neither assumption can be accepted uncritically. There is a lot of evidence that suggests teams can outperform groups of more talented individuals, and that people can perform above or below their natural abilities depending on the situation, including the leadership they receive and the help they get from others in their immediate environment.

But the problem is not just that the war-for-talent imagery may be wrong. It is an image that almost naturally produces a set of management actions that cause problems. Fighting the war for talent can readily create self-fulfilling prophecies that leave a large portion of the work force demotivated or ready to quit, and produce an arrogant attitude that makes it hard to learn or listen. It can cause the company to focus always on getting better people, mostly from outside, instead of fixing the culture and system of management practices that research has shown are consequential for performance. It is for these reasons that fighting the war for talent can indeed be hazardous to an organization's health, and why great companies get the best out of their people instead of always searching for different people.

SELECTED BIBLIOGRAPHY

1. The war for talent has been described in Charles Fishman, "The War for Talent," Fast Company (August, 1998); Elizabeth G. Chambers, Mark Foulon, Helen Hanfield¯Jones, Steven M. Hankin and Edward G. Michaels III, "The War for Talent," McKinsey Quarterly (1998); and in a forthcoming book by Michaels and his McKinsey colleaques. Many of the examples of the importance of teamwork and of building great cultures found in this article come from Charles A. O'Reilly and Jeffrey Pfeffer, Hidden Value: How Great Companies Achieve Extraordinary Results with Ordinary People (Boston, MA: Harvard Business School Press, 2000).

2. Research evidence on the importance of how companies manage people for organizational success is summarized in several places, including Pfeffer, The Human Equation: Building Profits by Putting People First (Boston, MA: Harvard Business School Press, 1998) and Brian E. Becker and Mark S. Huselid, "High Performance Work Systems and Firm Performance: A Synthesis of Research and Managerial Implications," in Research in Personnel and Human Resources Management, Vol. 16 (Greewnwich, CT: JAI Press, 1998), 53¯101. Our discussion of the Toyota production system and flexible production is based on material from James Womack, Daniel Jones, and Daniel Roos, The Machine That Changed the World: The Story of Lean Production (New York: Harper Perennial, 1990) and from John Paul MacDuffie's research, including "Human Resource Bundles and Manufacturing Performance: Organizational Logic and Flexible Production Systems in the World Auto Industry," Industrial and Labor Relations Review (1995).

3. The discussion of the elements of high performance or high commitment work systems comes from The Human Equation. Discussions of organizational culture and its importance can be found in Michael L. Tushman and Charles A. O'Reilly III, Winning Through Innovation (Boston: Harvard Business School Press, 1997). Also see Carla O'Dell and C. Jackson Grayson, Jr., If Only We Knew What We Know (New York: Free Press, 1998).

4. The self-fulfilling prophecy, or the effects of expectations on behavior, is a powerful and important theory often overlooked in the current emphasis on rewards and incentives. Some good, classic articles on the phenomenon include, W. Peter Archibald, "Alternative Explanations for Self-Fulfilling Prophecy," Psychological Bulletin (1974): 74¯84; Dov Eden, "Self-Fulfilling Prophecy as a Management Tool: Harnessing Pygmalion," Academy of Management Review (1984): 64¯7; and J. Sterling Livingston, "Pygmalion in Management," Harvard Business Review (July-August, 1969): 81¯89.

Jeffrey Pfeffer is the Thomas D. Dee II Professor of Organizational Behavior at the Graduate School of Business, Stanford University. He is the author or coauthor of 10 books, including The Human Equation: Building Profits by Putting People First, Managing with Power, The Knowing-Doing Gap: How Smart Companies Turn Knowledge Into Action and Hidden Value: How Great Companies Achieve Extraordinary Results with Ordinary People, as well as more than 100 articles and book chapters. Pfeffer has taught at the business schools at the University of California, Berkeley and the University of Illinois and has been a visiting professor at Harvard Business School.

Pfeffer serves on the board of directors of Actify, AudibleMagic, Portola Packaging, SonoSite, and Unicru and on the board of advisors for MTW Corporation, Earlybird Venture Capital, and Connet and Company as well as on the visiting committee of the Harvard Business School. He has presented seminars in 26 countries throughout the world, as well as consulting and providing executive education for numerous companies, associations, and universities in the United States.