It might be the toughest corporate survivor of all. It has outlasted boom and bust, employee enmity, workforce managers’ opposition, media criticism, crippling lawsuits, Enron’s implosion and the de-deification of its biggest and most vocal champion, former General Electric CEO Jack Welch. For all that–partly by keeping its head down and going by a number of colorful aliases–it has remained very much at large. It is known as forced ranking and is variously called Topgrading, the Vitality Curve, Forced Distribution, Differentiation and Rank and Yank. And by some estimates, the controversial employee-rating system has taken root in as many as one in five Fortune 500 companies. It reached its peak of popularity just before Welch’s retirement in 2001, but the sour economy has kept its balloon from bursting.
It’s a workforce-management tool based on the premise that in order to develop and thrive, a corporation must identify its best and worst performers, then nurture the former and rehabilitate and/or discard the latter. It’s an elixir that in these slow-growth times has proved irresistible to scores of desperate corporate chieftains–but indigestible to a good many employees. It’s a rough-and-tumble evaluation technique practiced at least to some extent by such corporate heavyweights as GE (which didn’t respond to our requests for an interview on the subject), 3M (“We’re going to take a pass on this subject.”), Texas Instruments (“No comment.”), EDS (“The person you need to talk to is traveling and won’t be available.”), Microsoft (“We don’t use forced ranking.”) and Hewlett-Packard (“HP’s performance rating is not designed to drive out a certain percentage of the company. HP has been evaluating and measuring performance and results for a long time. It’s motivational, and the employee, the team and the company all benefit.”).
Forced ranking is a subject that makes many top managers cringe. “I believe that the reason for the great reluctance about talking about forced ranking,” says Dick Grote, founder and head of Grote Consulting Corporation in Addison, Texas, “is that in our culture we have a bone-deep belief in egalitarianism. That all people are essentially the same. And one of the great advantages of forced ranking is that it requires reluctant managers to actually identify the most and the least talented members of the work group.”
That’s a necessity because “all God’s children are not the same,” Grote adds. “And that is treated as management’s dirty little secret.” Grote is one of the country’s foremost advocates of the rating system and has helped implement it at half a dozen or so large companies, which he is contractually forbidden to name. “The benefits of forced rating, intelligently and ethically conducted, are numerous,” he wrote recently in an article published by the Conference Board. “More than any other process, the system creates and sustains a high-performance, high-talent culture.”
Byron Woollen, head of New York City-based Worklab Consulting, is one of many business consultants whose philosophy is diametrically opposed to Grote’s. Woollen has advised a number of corporations on how to avoid or extricate their organizations from the evaluation process. He says he can only speculate about why companies using forced ranking or its variants keep their heads down. “It’s really a hot-button issue these days.” Especially since attorneys specializing in employment law tend to aim for disgruntled forced-ranked employees like heat-seeking missiles.
The Ground Rules
For those experienced with forced ranking, no explanation of its attention-getting ground rules is needed. For the rest of us, the grading system is based on the premise that rigorous evaluation and routing of employees by their immediate supervisors on agreed-upon abilities, skills and attitudes are not only possible, it’s vital. Thus, everybody from the top down can be–and sometimes is–ranked and placed on a bell-shaped company-wide curve, or in one of four quartered-square “quartiles,” or “buckets.” Other companies use a 1-to-5 ranking scale, 5 being best. Employees who finish, say, in the first 15 percent of the curve or the top-left quartile or are rated as a 5 are marked as A players–corporate stars and future leaders–and showered with raises, stock options and training. Performers in, say, the middle 70 percent of the curve or top-left and bottom-right square are B’s, given lesser raises and encouraged to become A’s. Those in the bottom right-hand square or the bottom 10 percent–or rated as 1–are given no raises or bonuses and either (a) offered remedial tutoring and mentoring in the hope of turning them into B’s; (b) offered remedial tutoring and mentoring and asked whether they might not be happier at another company; or (c) fired. Thus, Rank and Yank.
The execution of a forced-ranking system demands intense yearly performance reviews, during which managers must place each of their underlings in their proper place. No exceptions are made. In theory at least, each round of ranking and/or yanking will ratchet up the total quality of the workforce one notch. In practice, that depends on whom you talk to.
The Pros and Cons
Generally speaking, the proponents and opponents of forced ranking agree on only one thing: that they are right and everybody else is wrong. However, their theoretical disagreements–which are played out inside real companies among real people and real lives–break down into five major categories.
• Statistical Validity. The very heart of forced ranking is the belief that if, say, 100 people selected randomly from the Bronx phone book are measured for height, weight or their time in the 100-yard dash, the results will invariably display themselves on a bell-shaped curve. Similarly shaped will be measurements of a company’s, division’s or department’s personnel on matters of “core competencies.”
Of course, the members of a typical work group are not selected randomly and seldom number 100. Not an insurmountable problem, says Helen Handfield-Jones, a former McKinsey & Co. consultant and co-author of The War for Talent, a 1999 book that was instrumental in popularizing what she and her co-writers call “differentiation.” “It doesn’t have to be an exact bell curve,” she says. “I agree entirely that this is not precise science.” But it is better than any alternative, she says. “Next time you have a promotion, are you going to make a random selection?” Handfield-Jones believes that 23 or 25 is the minimum number of employees that can be ranked. Dick Grote believes that it’s desirable to have at least 37.
Nonsense, says Worklab’s Byron Woollen. “You have managers taking a bastardized notion of that [mathematical] principle and then just kind of monkeying with an idea that doesn’t really apply.” And this mathematical mirage, says a former employee for Metropolitan Life who never quite made it into the top bucket (and who, like all other employees interviewed, we opted to keep anonymous for obvious reasons), produced results that might seem surreal if they weren’t so frustrating.
“There were 300 account specialists, all with the same caseload, on one floor doing the same thing, and 25 managers,” he says. “Now, my manager knew my work ethic, but what about the other 24? They didn’t have a clue.” He strongly suspects that when the managers went into a conference room to rank him and his associates, they had nowhere near the information needed to do it fairly. And, he adds, the “top 10 percent” of each 25-person group is exactly 2.5. “I believe in the science of the bell curve,” he says. “It would be great if it could be used properly. Also, ideally communism is a good idea. But in reality it doesn’t really work.”
Says another observer from the trenches: “I worked in HR for HP several years ago and never met a manager who didn’t loathe the forced-ranking system. The point they consistently made was that all their employees could be performing at acceptable or even above-average levels, yet they had to place a certain number in the bottom two performance levels.”
• Objectivity. Perhaps the most admirable aim of forced ranking is to liberate lagging organizations from lethargy and paternalism by forcing managers, most of whom have the all-too-human tendency to be lenient in evaluating their subordinates, to face hard facts. To do that, they have to rate their underlings honestly. And to do that, they need as objective a set of criteria as possible. Finding them is a major problem in instituting any performance-review system. Forced ranking, though, raises the stakes.
Jack Welch of General Electric, for instance, instituted “the four E’s of GE leadership.” They were: “very high Energy levels, the ability to Energize others around common goals, the Edge to make tough yes-and-no decisions, and finally the ability to consistently Execute and deliver on their promises.”
| “I never saw a CEO’s brother-in-law who was a C. Funny, isn’t it?” |
Each of these E’s and standards like them, say his critics, are “E-Z” on-ramps to varying degrees of interpretation, subjectivity and favoritism. “I was in the business world for more than 25 years,” says a former executive who recently became a private-school headmaster. “And I never saw a CEO’s brother-in-law who was a C. Funny, isn’t it?”
Also, forced-ranking systems usually leave out “softer” qualities that some consider essential to any organization–like teamwork, honesty, dedication and cheerfulness. Well, sure, says Helen Handfield-Jones, but what’s the alternative? “Yes, these are human beings. Again, this is not science. There are no steel calipers you can use to make measurements. But you still have to put people in jobs. And people make judgments all the time anyway. So the best thing to do is be as objective as you can and never leave the decision to one person. Different people see different things. Bring multiple senior people into the decision. Give everybody high-quality objectives to aim for. And have them written down to be as objective as possible.”
• Morale. For most people–especially those with outmoded concepts of loyalty and job security–the prospect of Darwinian struggle at their workplace is not a happy one. (Jack Welch noted: “This is hard stuff. No leader enjoys making the tough decisions. We constantly faced severe resistance from even the best people in the organization. I’ve struggled with this problem myself and have often been guilty of not being rigorous enough.”)
“It’s like cardiac arrest for an organization,” says Woollen, who adds that at one old-line paternalistic company he assisted, the mere (true) rumor of an impending switch to rank and yank sowed fear, loathing and preparations for a class-action lawsuit. On the other hand, he concedes, at companies where forced ranking is well established, the effects may be different.
“When you have people entering GE, you have people who are self-selecting out in some ways,” he says. “They say, ‘I want to go to GE because it’s cutthroat. And I’m badass enough so I can do that.’ And once in, there’s that culture of ‘We’re all badasses, and we’re all gonna take over the world, and this is the way we work.’ You sort of sign on for that.”
• Politicization. Armed with ample evidence, critics say that during the evaluation process, managers will often ignore the facts. To protect and advance their own people, they haggle, horse-trade, call in markers and even use threats, emotional appeals and executive-suite connections. The most extreme and well-documented example is Enron. The company’s “Performance Review Committee” became a snake pit and catalyst of naked power plays and greed. Forced-ranking advocates don’t have much to say these days about Enron. But they do contend that a well-run review process can squeeze out most or all of these aberrations.
• Cannibalization. Almost everybody agrees that the typical corporation has a certain number of under- and non-performers, and that a year or two of good, fair forced ranking can weed them out. After that, though–especially if a company is cutting its workforce–something else happens. Former A’s will become B’s and former B’s will become C’s. Upwardly mobile B’s will displace A’s; recovering C’s will merely replace shaky B’s. Not the healthiest of situations, admits Dick Grote. “I think that after about three iterations, forced ranking loses its effectiveness,” he says. “I think the best thing for companies to do is to wait three or four years, bring in some people from outside the company and start over again.”
• Diversification. An alleged bias against women, minorities and older workers–indeed, any class of workers that upper management allegedly wants to weed out–is the most contentious argument against forced ranking. In 2000, the Ford Motor Company instituted a forced-ranking system that attracted two class-action lawsuits alleging discrimination on the basis of age, gender and race. In 2002, the suits were settled for $10.5 million and Ford dropped forced ranking. At the energy company Conoco, two forced-fired geophysicists who were replaced by citizens of the U.K. under special visas alleged that they were discriminated against because they were Americans. A confidential settlement was reached in 2001.
Last year Goodyear was sued for age discrimination by several parties. It too dropped forced ranking. Microsoft, where employees are rated from 1 to 5, was sued by several African-American men and women for race and gender discrimination. Most of the suits were dismissed by a federal judge in Seattle, but the company reached a confidential settlement with one plaintiff.
These lawsuits are almost inevitable, says Byron Woollen, especially if vague, ill-considered criteria such as “fits in with others” are used in performance reviews. Most forced-ranking advocates contend that lawsuit traps can be avoided by conscientious managers and consultants. Dick Grote has a slightly different take. “The reason that people sue is not because they have been discriminated against. They sue because they don’t feel like they have been treated fairly.
“For example, if you have a person who is a member of a protected class, a black female, [with a] Spanish name and who is handicapped, it takes a lot of courage to walk up to her and say, ‘Susan, your performance isn’t very good.’ It takes a lot of courage to do that, so managers are quite likely to avoid having those tough conversations. Forced ranking is forcing them to do so.”
Workforce, July 2003, pp. 44-49 — Subscribe Now!