Within the professions there are abundant illustrations that the skills on which credentials are granted are different from the performance that matters most. For example, in 1979 Daniel Hogan, a lawyer and social psychologist at Harvard, published a four-volume study called The Regulation of Psychotherapists. Its ambition was to examine the day-by-day workings of psychotherapy at every level, from social worker to licensed psychoanalyst.
Hogan devoted his first several hundred pages to an analysis of the traits and qualities that distinguish effective psychotherapists from ineffective ones. In judging effectiveness he concentrated on "output"--changes in the patient's condition--rather than "input," such as how much effort the therapist applies, how much he charges, or how long he spent in school. Then, in the second half of that volume, and with the same painstaking thoroughness, Hogan went through the qualities demanded of those who want to be certified as psychotherapists. There was little overlap between the two lists.
"Contrary to much professional opinion ...," he said, "the effectiveness of therapists is more determined by the presence or absence of certain personality characteristics and interpersonal skills than technical abilities and theoretical knowledge." The skills that make a superb psychotherapist are mainly common-sense human skills--warmth, empathy, reliability, a lack of pretentiousness or defensiveness, an alertness to human subtlety, an ability to draw people out. "The necessary qualities are very similar to those one looks for in a good friend." These are not traits that can be detected on a multiple-choice exam, but they are real, and can be measured in creative ways. In half of the "effectiveness" studies that Hogan surveyed, non-professional therapists did better than professionals in helping patients, despite their lack of formal education. In one study conducted in 1965, for example, five laymen (only one of whom had finished college) were given less than 100 hours of training in therapy skills. Then they were put in charge of patients who had been hospitalized, on average, for more than thirteen years. Under their treatment more than half the patients improved.
Hogan contrasted such subjective skills with the traits the profession considered essential before issuing a license, most of which were based on academic proficiency. "For traditional psychotherapy, psychiatrists stress an understanding of human biology, neurology, and psychopharmacology; psychologists stress personality dynamics and interpersonal behavior; and social workers believe that a theoretical understanding of environmental influences on behavior is essential." As Hogan pointed out, such "hard" scientific preparation was necessary in some cases, to be sure that the patient's complaint did not arise from chemical imbalance, from injury, or from a tumor. But once those possibilities had been eliminated, Hogan's findings showed, advanced technical training counted for nothing in restoring most mental patients to health.
If psychotherapy seems too "soft" a discipline to provide a fair test of meritocratic standards, what about air-traffic control? In 1970 Ivar Berg reported on a study conducted by the Federal Aviation Administration, which wanted to understand what made 507 highly competent air-traffic controllers good at their jobs. The question was whether advanced educational requirements would produce competent controllers; the answer was no. As Berg explained,
This complicated job ... might well reuire, not merely the details of engineering or management science or mathematics, but all the supposed "correlates" of education--a disciplined mind, for example--and the more personal qualities that education is supposed to produce--reliability, steadfastness, responsibility, ability to think quickly, motivation, etc.
Common sense might suggest that the better controllers would be more educated--but the FAA found that fully half the top-ranked controllers had no formal education beyond high school. Many of them had come directly to the FAA for rigorous technical training specifically related to the jobs they were expected to do. Berg said,
Because it was "stuck with" less educated men ... the FAA became a little laboratory in which the relevance of education for attainment of, and achievement in, important managerial and technical positions would be examined. Education proves not to be a factor in the daily performance of one of the most demanding decision-making jobs in America.
The implication of examples such as these is not that talent is equally distributed or that minds are limitlessly malleable or that advanced training is always destructive. A liberal education is good for its own sake, and schooling of any sort can impart a broad perspective that can help in any job. Rather, the charge against credential requirements is that they are simultaneously too restrictive and too lax. They are too restrictive in giving a huge advantage to those who booked early passage on the IQ train and too lax in their sloppy relation to the skills that truly make for competence. No nurse is allowed to hang out a shingle and collect professional fees for the many medical functions she can competently perform; any psychiatrist is legally entitled to perform open-heart surgery or read x-rays of your knee. If sports were run like the meritocracy, the Miami Dolphins would choose their starting lineup on the basis of high-school times in the forty-yard dash and analyses of the players' muscle tissues to see who had the highest proportion of "quick-twitch" fibers. If the Dolphins actually did this, they'd face a long losing season: the coach cares about speed but finally chooses the players who have proved they can catch the ball or stop the run.
Nearly fifteen years ago David McClelland wrote an article called "Testing for Competence Rather Than Intelligence." It said, in effect, that what Don Shula does for the Dolphins the testing and licensing system should do for the professions. While some people are brighter than others, and while the variations in their abilities matter in some jobs, differences in IQ scores should not be the central concern of professional licensing. The proper function of licenses is to ensure that when passengers enter an airplane, they can count on the pilot's knowing how to fly, and that anyone who offers to argue a case in court or prepare a tax return is competent in those tasks. Designing tests of these specific skills might be slightly harder than drawing up yet another IQ test, McClelland said, but the obstacles would hardly be insuperable. Social competition would be more open, the economy would be more flexible, and standards of performance would be higher if credential requirements gave way to tests of specific skills.
In business the companies that are growing and changing the fastest, and where flexibility and performance are presumably more crucial than anywhere elese, already tend to overlook credentials and behave like armies in wartime, rewarding people for what they can do today, not for their background or what their theoretical potential might be. "We do a lot of college recruiting to find our new people," says Steven Ballmer, a twenty-nine-year-old vice-president of Microsoft, the phenomenally successful software firm that Ballmer's contemporary and college classmate, Bill Gates, founded after dropping out of Harvard. Ballmer dropped out of Stanford Business School to join him. "We go to colleges not so much because we give a damn about the credential but because it's hard to find other places where you have large concentrations of smart people and somebody will arrange the interviews for you. But we also have a lot of walk-on talent. We're looking for programming talent, and the degree is in no way, shape, or form very important. We ask them to send us a program they've written that they're proud of. One of our superstars here is a guy who literally walked in off the street. We talked him out of going to college and he's been here ever since."
Such established firms as General Electric and AT&T have long been known for recruiting college graduates and then offering management training, as necessary, inside the firm. Of the 4,500 entry-level professionals General Electric hires each year, only fifty are new M.B.A.s. Most of the others have technical backgrounds; as they move up, they are given brief courses inside the company rather than being formally sent back to school. "As far as we're concerned, there's no broad incentive for technical companies to go out and get M.B.A.s," says James Baughman, who formerly taught at Harvard Business School and now supervises management training at GE. "It's a heck of a lot easier to change a technical person into a businessman over the years than the other way around."
As an alternative or supplement to judging academic credentials, many firms have developed "assessment centers," in which employees handle simulated business problems, in a setting as close to real life as possible, to demonstrate their competence or indicate the need for training. Candidates for administrative jobs, for example, might work their way through a sample in-box. "Bosses find those promoted because of their assessment center scores to be competent, the candidates feel the system is fair, and assessors believe that the process has given them the chance to measure important characteristics," wrote Robert Klitgaard in his recent book, Choosing Elites.
A number of firms, from McDonnell Douglas to Mobil to Digital Equipment, have turned to McBer for its "competency" analyses of specific jobs. The results are sometimes surprising. To manage its new-product-development lab, for example, one firm had habitually looked for freewheeling, creative types; the lab's researchers were innovators, so naturally their boss should be too. "It turned out that those with the best performance were actually less creative and risk-taking than others," Richard Boyatzis says. "The most creative people held onto ideas way too long. What distinguished the superior performers were other traits, like being able to informally steer people and to get engineers, market researchers, and scientists to pull together."
Equipped with such knowledge, the company was able to select more-competent directors; more important, it was able to train a broader range of people to succeed. McBer's view of "competencies" is very similar to Binet's view of intelligence: after the illness, the remedy. Boyatzis says, "The most positive message we consistently get is that people do want to improve themselves, but usually they don't know exactly what to work on. When you can give them good feedback on specific goals, that relates the natural internal inclination to improve."
IT IS POSSIBLE TO COMBINE THAT BASIC DESIRE FOR IMPROVEMENT and upward mobility with standards that ensure high performance? Can a society be both efficient and open? One of the most successful, and least credentialed, assessment procedures suggests that it is.
Among lawyers, accountants, an M.B.A.s incompetence may be a nuisance, but in airline pilots it is a catastrophe. In the early days of commercial flight the airlines bore the responsibility for training and certifying their pilots, but the soon begged for government regulation, so as to spread the responsibility when crashes occurred. Like the licensing procedures for doctors, lawyers, and engineers, these standards were supposed to protect the public from incompetence, but they were of a very different nature from those of the professional guilds. The pilot-licensing system was built on the premise that competence was divisible: people can be good at one thing without being being good at others, and they should be allowed to do only what they have mastered. As opposed to receiving a blanket license, the way members of other professions do, pilots must work their way up through four certificate levels, from student to air-transport pilot, and be specifically qualified on each kind of aircraft they want to fly. What's more, a pilot must demonstrate at regular intervals that he is still competent. To keep his license a pilot must take a review flight with an instructor every two years, and the pilots for commercial airlines must pass a battery of requalification tests every six months. "A small but regular percentage is washed out each time," John Mazor, of the Air Line Pilots Association, says. It is reassuring to know they are gone, but what about their tenured counterparts in the other professions?
The results of this licensing scheme are a high level of proficiency and a profession more open socially than the rest. Most pilots of big jets learned to fly in the military, since that is the least expensive way to put in the 1,500 hours of flight time necessary for an air-transport license. But the remainder slowly worked their way up, putting in flight time on their own or working for small air-taxi outfits until they could move to the next level of licensure. Imagine what other professions would be like if they operated this way. The sociologist Randall Collins's prescription for medical training follows a similar pattern:
All medical careers would begin with a position as orderly, which would be transformed into the first stage of a possible apprenticeship for physicians. After a given number of years, successful candidates could leave for a few years of medical school (2 years seems sufficient background for most practitioners ...) and then return to the hospital for advanced apprenticeship training of the sort now given in internship and residency programs. ... Advanced specialties could continue to be taught as they now are--through further on-the-job training; only medical researchers would be involved in lenghty schooling.
In theory business is better positioned than the professions to resist the worst effects of a meritocracy. The professions depended for their creation and growth on credential barriers that kept people out; business depends for its survival on making the best and most flexible use of all its resources, including talent. Even dominant firms must face the possibility that somebody who may not have gone to the right school and may not have the right degree might still come to market with a better, cheaper product.
Because successful business practice depends to some extent on apperances, business may never be as completely open as America's one true meritocracy--sports. (It didn't matter that Babe Ruth was fat, slovenly, and ungrammatical, so long as he could hit the ball.) But why shouldn't sports, rather than the professions, epitomize the meritocracy to which we aspire? American professional sports have their sins and excesses, to be sure. But with their relative openness to newcomers and disregard for background (most teams have hired no-name free agents and waived famous first-round draft choices) and their faith that ruthless and continuing judgments of performance will finally lead to equal opportunity, sports seem more admirably meritocratic than the system of early selection and later tenure that meritocracy has been perverted to mean.
Perhaps the cultural changes that have professionalized America are irreversible. The economist Mancur Olson has gloomily hypothesized that most societies tend to separate into inflexible castes, except when warfare or other cataclysms disrupt the social order an unleash new talent. The United States has renewed itself in less traumatic fashion--by continually populating new regions, by absorbing varied immigrant groups, and by taking deliberate steps, such as the GI Bill, to give more of its people a chance. As we drift toward a neater and more predictable social order, we might reflect on the rough-and-ready adaptation to experience that brought us this far, and ask ourselves whether we need it still.