Peter Drucker's Legacy
Includes Simple Advice:
It's All About the People

Peter Drucker was the most influential management thinker of the past century. But his most crucial insights were about workers.

Mr. Drucker, who died Friday at age 95, was among the first to see the limits of large industrial organizations and their authoritarian hierarchies. Long before the Internet, before even the first computer chips, he foresaw the arrival of "knowledge workers" motivated by personal pride as much as by fear and a paycheck. Harnessing their talents, he argued, required a new approach to management.

He dispensed this advice in simple prose in 39 books over a remarkable 60-year career, and in probing conversations with scores of executives. Along the way, he developed a loyal following among many of the world's most-famous corporate chieftains, and became the model of the modern management guru, a craft he plied far more modestly than many of his successors.

While Mr. Drucker's eclectic interests ran from European history to Japanese art, his management teachings centered on ways to make workers more effective.

David A. Jones, co-founder and retired chairman and chief executive of Humana Inc., a Louisville, Ky., health insurer, recalls the core of Mr. Drucker's advice this way: "Successful enterprises create the conditions to allow their employees to do their best work."

Mr. Drucker offered plenty of other lessons, of course. He believed organizations should articulate a clear purpose, with specific, measurable goals; he developed the concept of "management by objective," to keep managers in step with those goals; he encouraged managers to ask unspoken questions and consider ignored issues.

His interests weren't limited to profit-seeking corporations. Mr. Drucker viewed nonprofit organizations as social linchpins, and devoted entire books to management of these groups.

"He had the anthropologist's insight into this strange tribe [of managers] that had these formal rituals and strange practices," says Michael Useem, a professor at the University of Pennsylvania's Wharton School, and an author of several books on management and leadership. "Peter Drucker was able to see behind them, and also see what could be changed and made for the better."

Mr. Drucker contributed much to the modern cult of the chief executive. Yet as an emigrant from Nazi Europe, he retained a lifelong distrust of charismatic leaders. "He was skeptical of hero worship," says John Alexander, president of the Center for Creative Leadership in Greensboro, N.C. "He saw management as an activity rather than a heroic venture."

Mr. Drucker's varied interests led him to predictions that gave him a reputation of a visionary in some circles. Warren Bennis, a University of Southern California business professor and author of more than two dozen books about leadership and related subjects, recalls Mr. Drucker warning him 15 years ago about coming social disruptions because of shrinking populations in Western Europe. In 1987, when Japan's roaring economy was the envy of the world, Mr. Drucker saw trouble ahead. "The pillar of their success -- lifetime employment -- is becoming an almost insurmountable barrier to flexibility," he said.

Skeptics, and there were few, who studied his record said that Mr. Drucker was wrong as often as he was right, and had a penchant for twisting anecdotes in the retelling. But that did little to shake the faith of several generations of CEOs. Mr. Drucker's impact was so profound that most of them still remember the first time they read, or met, him.

For Humana's Mr. Jones, it was in 1974, when his colleague Wendell Cherry bought one of Mr. Drucker's books at an airport to help pass the time of a flight delay. "Wendell called me and said, 'Some guy wrote a book about us," Mr. Jones recalls. The two finished the book, "Management, Tasks, Responsibilities, Practices," in a weekend, then called Mr. Drucker on Monday morning.

A few weeks later, the pair flew to Mr. Drucker's home in Claremont, Calif. There, Mr. Jones quickly learned the cornerstone of Mr. Drucker's style: "He never really answered questions. He always asked them." Still, Mr. Jones was sufficiently impressed that he repeated the pilgrimage annually for more than a decade. He recalls two preachings: That profit is a requirement for a company, but should not be a goal in itself; and that productivity and quality are effectively the same thing. "A day or two spent with Peter was the most valuable way I could spend my time," Mr. Jones says.

Dan W. Lufkin, co-founder of the Wall Street investment firm Donaldson Lufkin & Jenrette, encountered Mr. Drucker's unusual style during their first meeting, in the early 1960s, just as DLJ was just getting off the ground. Mr. Drucker, who spoke with an Austrian accent, initially seemed "formal and authoritarian," Mr. Lufkin recalls. "I asked him if he thought we should sell a certain product or do a certain strategy, but all he said was 'I don't know' to every question I posed," he recalls. "So finally I asked, 'what am I hiring you for?' " Mr. Lufkin says.

In response, Mr. Drucker said, "I'm not going to give you any answers, because there are always many different ways to approach problems, but I'm going to give you the questions you should ask,' " Mr. Lufkin says. "So we started talking in great length and depth about who we were and what we wanted to do -- and I can't tell you how important he was to the development of the firm," Mr. Lufkin says.

For Andrew Grove, the retired chairman and chief executive of Intel Corp., the Peter Drucker moment came in the late 1970s, when he ran across a book that Mr. Drucker had published about a decade earlier. The book included a chapter on the multiple roles of a CEO: the public face of the company, a strategist and an operational manager.

Mr. Grove says the descriptions echoed the way that he and two other Intel co-founders, Robert Noyce and Gordon Moore, had unconsciously divided the duties at the top of the semiconductor maker: Mr. Noyce as the public face, Mr. Moore as "a man of thought," and Mr. Grove as "a man of action." Mr. Grove says he ran to a copy machine and distributed copies of the chapter to Messrs. Noyce and Moore.

At the time, Mr. Grove says, he was a "young manager, very skeptical about management gurus and consultants." Nonetheless, he says Mr. Drucker's writings "spoke to me." He was so impressed that Mr. Grove drove an hour from Intel's Silicon Valley offices to San Francisco to watch that era's motivational video -- a three-hour movie of Mr. Drucker speaking about management.

Mr. Grove singles out two of Mr. Drucker's precepts that have stuck with him: That managers should never promote an employee on the basis of his or her potential, but based only on performance; and that managers should make a decision "no later than you need it, but as late as possible, because you always have more information."

Mr. Drucker's lessons still resonate with a younger generation of managers. Mike Zafirovski, 51, who begins work tomorrow as chief executive of Nortel Networks Corp., never met Mr. Drucker, but says the author had a "huge influence" on him. Reading Mr. Drucker's books, Mr. Zafirovski says he was persuaded by the argument that companies should "treat employees like their most valuable resources, including pushing decision making to the lowest levels."

Theory & Practice is a weekly look at people and ideas influencing managers.

Write to Scott Thurm at scott.thurm@wsj.com and Joann S. Lublin at joann.lublin@wsj.com

By STEVE FORBES
November 15, 2005; 

What made Peter Drucker, who died last Friday just shy of his 96th birthday, the most influential management guru of the modern era?

Mr. Drucker's genius for extraordinarily farsighted insights came from a combination of intense curiosity, right principles and deep understanding of the perfections and imperfections of human nature. He never went stale intellectually, which is why business journalists, executives, entrepreneurs, leaders of nonprofit institutions, students and the occasionally wise politician eagerly sought to pick his brains right up to the time he died.

* * *

What helped make Mr. Drucker so insightful was a profound understanding of economics, an understanding that still eludes most economists today. Not for him was the notion of "macroeconomics," of seeing the economy as something of a machine that can achieve steady, stable growth. To him, traditional economic notions of "equilibrium" or Keynesian ideas of "aggregate demand" were nonsense. Innovation, constant change, and turmoil were the true constants of a progressing economy.

No surprise that the economist fellow-Austrian (at least by birth) Joseph Schumpeter was Mr. Drucker's hero. In 1983, at the centennial of both Schumpeter and the then-legendary John Maynard Keynes, Mr. Drucker wrote in Forbes that Schumpeter's centenary birthday would hardly be noticed. Yet "Schumpeter it is who will shape the thinking and inform the questions on economic theory and economy policy for the rest of this century, if not for the next 30 or 50 years." Today Schumpeter's emphasis on the crucial importance of entrepreneurship and "creative destruction" are now commonplaces.

As Mr. Drucker wrote over two decades ago, "The economy is forever going to change and is biological rather than mechanistic in nature. The innovator is the true subject of economics. Entrepreneurs that move resources from old and obsolescent to new and more productive employments are the very essence of economics and certainly of a modern economy. Innovation makes obsolete yesterday's capital earnings and capital investment. The more an economy progresses the more capital formation -- profits -- will it therefore need." These two men saw profits as a moral imperative, a genuine "cost" in the cost of staying in business because "Nothing is predictable except that today's profitable business will become tomorrow's white elephant."

B.C. Forbes, our company's founder, who came to this country 100 years ago with little education and even less money, liked to say that you learn more about a company's prospects from observing its "head knocker" (what he called CEOs) than you will from its balance sheet. Mr. Drucker spent a lifetime hammering home the point that people are key. For instance, a leader who looks at workers as a cost instead of a resource is fatally flawed.

No surprise he long recognized the importance of entrepreneurs: "All great change in business has come from outside the firm, not from inside."

Mr. Drucker's ability to prophesy -- almost always correctly -- was uncanny. All of this is why he could come up with innovations that now seem commonplace, such as management by objective. He continued to admonish executives to carve out time to think and make careful decisions, to focus on one or two tasks, to delegate to others what you can't do well yourself. That's why, for example, Mr. Drucker remained a one-man shop, a soloist; he could easily have founded a large consulting firm and gotten immensely rich. But that would have gone against his profoundest instincts. He was at his best as a teacher -- gathering information, gaining insights and then getting others to gain understanding. Schumpeter believed asking the right questions was more important than the answers. Mr. Drucker agreed -- to a point, anyway.

Decades ago, Mr. Drucker foresaw the rise of "knowledge workers." After World War II, he realized the far-reaching consequences of the GI Bill of Rights, which enabled millions of veterans to go to college, thus leading him to predict long before computer chips and the Internet that "knowledge workers" would replace manual workers. Mr. Drucker also prophesied the breakdown of the traditional, thoroughly integrated, hierarchal industrial corporation. In the 1950s, he predicted the rise of Japan as a major economy, an astonishing insight when many experts thought the country would forever be a nation of small farmers and manufacturers of cheap, shoddy goods. He also saw Japan's subsequent troubles -- an aging population and lack of vigorous entrepreneurship and worker flexibility.

Mr. Drucker long ago warned of the consequences of the rise of corporate and government pension funds, and the impact these vast accumulations of money -- and thus power -- would have on corporate governance, years before anyone had heard of Calpers. He also warned of a backlash from the extraordinary rise in CEO pay. "In the next economic downturn," he told Forbes readers nearly a decade ago, "there will be an outbreak of bitterness and contempt for these super corporate chieftains who pay themselves millions. In every major economic downturn in U.S. history, the villains have been the heroes during the preceding book."

Mr. Drucker also told us to expect enormous changes that will come in higher education, thanks to the rise of satellites and the Internet. "Thirty years from now big universities will be relics. Universities won't survive. It is as large a change as when we first got the printed book." He believed "High school graduates should work for at least five years before going on to college." It will be news to most college presidents and a lot of alumni that "higher education is in deep crisis. Colleges won't survive as residential institutions. Today's buildings are hopelessly unsuited and totally unneeded." All this from a life-long academic.

He brooked no nonsense about some of the topics that obsess Chicken Little today. Outsourcing? He told Fortune in 2002 that "We import two to three times as many jobs as we export. Wage costs are of primary importance for very few industries. The industries that are losing jobs out of the U.S. are the more backward industries." He never tired of pointing out the huge advantage the U.S. has over Europe and Japan and other countries with American workers' flexibility, not only for changing jobs but physically moving from one area to another to pursue opportunities.

In fact, outsourcing is a necessity, Mr. Drucker said. Companies should have others do what is not their prime task. Outsourcing is not so much about cost cutting ("illusory") as it is about improving the quality of work that others can do better than you: "You should outsource everything for which there is no career track that can lead to senior management."

* * *

How higher education is managed did not impress Mr. Drucker; but what did is our continuing education system, whether in community colleges or by computers. Also: "Our most important education system is in the employees' own organization." That is where most Americans learn the most. Mr. Drucker also came up with the admonition of pursuing your opportunities and cutting your losses: "A critical question for leaders is 'When do you stop pouring resources into things that have achieved their purpose?'" As he repeatedly told Pastor Rick Warren, founder of the 15,000 member Saddleback Community Church in Lake Forest, Calif., and who has helped start another 60 churches around the world, "Don't tell me what you are doing, Rick, tell me what you stopped doing."

Until his last breath, Mr. Drucker himself never stopped doing and doing.

Mr. Forbes is president & CEO of Forbes, Inc. and editor-in-chief of Forbes magazine.

By PETER F. DRUCKER
November 15, 2005

(This article originally appeared in The Wall Street Journal on March 24, 1987)

The policy decisions that led to "Irangate" have been proved by hindsight to have been mistakes. But the management methods that turned Irangate into a scandal should have been seen from the beginning as likely -- indeed almost certain -- to lead to trouble, if not to disaster. They were violations of well-known and amply tested management principles. The Reagan administration violated not just one of these principles -- it violated four.

First -- in one of the most common but also most unforgivable management mistakes -- it confused delegation of authority with abdication of responsibility. A chief executive officer must delegate. Otherwise, he'll end up like Gulliver in Lilliput, ineffectual and ensnared in details, as were Lyndon Johnson and Jimmy Carter. But delegation requires greater accountability and tighter control. Delegation requires clear assignment of a specific task, clear definition of the expected results and a deadline. Above all it requires that the subordinate to whom a task is delegated keep the boss fully informed. It is the subordinate's job to alert the boss immediately to any possible "surprise" -- rather than to try to "protect" the boss against surprises, as Mr. Reagan's subordinates apparently did. If they keep surprises away from the boss, they invariably will end up making him look incompetent or not in control or a liar -- or all three.

The greatest delegator in recent American political history was not Ronald Reagan; he is apparently quite immersed in all kinds of operational matters. The greatest delegator was Franklin D. Roosevelt, who "did" an absolute minimum. Yet FDR always delegated a specific task, defined the desired results, and stipulated when and how the subordinate -- a cabinet member as a rule -- would report back. And he demanded upward responsibility from his subordinates. In particular, they had to inform him immediately if the project deviated from the plan even in the smallest detail. He knew, as every chief executive officer learns sooner or later, that there are no "pleasant surprises."

The second major management lesson of Irangate is not to confuse, as the Reagan administration did (and apparently still does), two different and actually incompatible roles -- that of the chief operating officer and that of the chief of staff.

Whether a chief operating officer can work in the American constitutional system is by no means clear. The system centralizes all executive authority in the president, after all. None of the cabinet ministers, for instance, have any political or constitutional authority of their own as they have in a cabinet-government, like that of Britain. And the concept of chief operating officer has never worked when tried in Washington. It always gets the president into trouble.

It did not work when Franklin D. Roosevelt put in James Byrnes as the chief domestic operating officer during World War II. And Dwight Eisenhower did not become an effective president until a scandal forced him to get rid of Sherman Adams, his chief operating officer. But if there is a chief operating officer, the chief executive must retain some direct operating responsibility. Otherwise he soon becomes isolated, and loses "feel" and "touch."

FDR, for all his delegating, never relinquished direct, day-to-day control of congressional relations or of relations with the press. Alfred Sloan at General Motors always had a president and chief operating officer. But he himself kept day-to-day operating responsibility for what he considered the two true "controls" of the corporation: personnel decisions down four levels -- that is, for all appointments to a senior management position even in the smallest accessory division; and allocation of capital, with the chief financial officer reporting directly to him. Both Harry Truman and John F. Kennedy trusted their respective secretaries of state. But both kept day-to-day control of foreign affairs.

It will be argued that Donald Regan was not "chief operating officer"; he was "chief of staff." If so, he was set up to do exactly what a chief of staff must never do: to keep information from the president. The first job of a chief of staff is to make sure that the chief executive officer gets all dissents, conflicting points of view and alternatives. To do this, he must not himself be a "part of the problem," must not himself represent one of the contending interests or ideologies, and must, above all, not be in the "line of command." In the military the chief of staff is never allowed to get between the senior commander and his subordinate commanders -- they always outrank the chief of staff. His job is to make sure that the boss gets all the information he needs to make a decision, rather than only what the chief of staff thinks the boss should hear.

If the two roles are mixed, the chief of staff invariably cuts the boss off from vital information, invariably tries to monopolize access to the boss, invariably ends up making the boss look out of touch. And he also encourages subordinates to do what President Reagan's subordinates at the National Security Council apparently did: keep information away from "upstairs" and bootleg their own policies.

Confusing delegation and abdication, and the chief operating officer with the chief of staff, are structural mistakes. But the Reagan administration also made two elementary mistakes in how it did things. It asked Vice Adm. John Poindexter and Lt. Col. Oliver North to carry out the official policy toward Nicaragua sanctioned by Congress -- which was not to give additional aid to the contras. At the same time, these two men also apparently were charged to raise private money for the contras. It does not matter that this hypocrisy was exactly what most members of Congress clearly wanted (including, probably, many who had voted against aid to the contras). If for whatever reason two conflicting policies have to be carried out at the same time -- and that, of course, does happen -- they must be carried out by different people and in separate organizations. The right hand must not know what the left hand is doing, if the two are working at cross-purposes. Otherwise there will always be a scandal and both policies will miscarry, as they did in Nicaragua.

Finally, the Reagan administration violated the simplest rule: There ain't no secrets. If more than one person knows it, it won't stay a secret. And then the only thing to do with bad news -- such as the miscarriage of the McFarlane-Poindexter mission to Tehran -- is for the executive himself to make it public. This way he has control. If he tries to cover up, he will make sure only that the "secret" is published at the worst possible moment, by someone who tries to hurt him, and in a form that gives it the worst possible interpretation.

American presidents in this century have tended toward secret actions in foreign affairs. Only one of these worked: the Kissinger-Nixon rapprochement with China. The other three ended in disaster. President Wilson, in 1916, sent his personal confidant, Colonel House, to Berlin to get the Germans to stop sinking civilian shipping. House succeeded only in convincing the Germans that Wilson was profoundly isolationist and would never go to war -- which then encouraged them, after Wilson's reelection a year later, to launch the unrestricted submarine warfare that forced America into World War I.

President Roosevelt sent Col. Charles Lindbergh to Berlin 20 years later to "establish contact with Nazi moderates." Lindbergh convinced the Germans only that the U.S. was unprepared, and could not mobilize fast enough to turn the balance in Europe. And the McFarlane-Poindexter mission to Tehran clearly also sent the wrong signals to the Iranians.

Yet, the dilemma created by kidnapping and hostage-taking is a real one, and if -- a big if, of course -- the mission had resulted in the release of the remaining hostages, President Reagan would have been a hero. But when such an endeavor fails, then one does what Franklin D. Roosevelt was particularly good at: make sure that the news is leaked by the president (or, in a business, the chief executive officer), and in a form in which it deflects blame to someone who is sympathetic. But above all one makes sure that there is no secret -- for the only secret no one pays any attention to (as Edgar Allen Poe showed 150 years ago in "The Purloined Letter") is something that is out in the open.

These are elementary lessons, and obvious ones. But it is not only the politicians in Washington who seem not to know them. There are far too many chief executives in American business who make the same mistakes. Let's hope that they will learn the management lessons of Irangate.

Mr. Drucker is a professor of social sciences at the Claremont Graduate School in California.

By PETER F. DRUCKER
November 15, 2005; Page B2

Peter F. Drucker died on Friday. The following article ran in The Wall Street Journal on July 25, 1989.

More and more people working in and for organizations will actually be on the payroll of an independent outside contractor. Businesses, hospitals, schools, governments, labor unions -- all kinds of organizations, large and small -- are increasingly "unbundling" clerical, maintenance and support work.

Of course, the trend is not altogether new. A great many American hospitals -- and European and Japanese hospitals as well -- now farm out maintenance and patient feeding; 40 years ago none did. "Temporary help" firms go back more than 30 years; but while in the beginning they handled file clerks and typists, they now provide computer programmers, accountants, engineers, nurses and even plant managers. Cities farm out "waste management" (once known as street cleaning and garbage disposal); even prisons are being run by private contractors. . .

Support work is rapidly becoming capital-intensive. In many manufacturing companies, the investment in information technology for each office employee now equals the investment in machinery for each production worker. Yet the productivity of clerical, maintenance and support work is dismally low, and is improving only at snail's pace, if at all. Unbundling will not by itself make this work more productive. But without it the productivity of clerical, maintenance and support work is unlikely to be tackled seriously.

In-house service and support activities are de facto monopolies. They have little incentive to improve their productivity. There is, after all, no competition. In fact, they have considerable disincentive to improve their productivity. In the typical organization, business or government, the standard and prestige of an activity is judged by its size and budget -- particularly in the case of activities that, like clerical, maintenance and support work, do not make a direct and measurable contribution to the bottom line. To improve the productivity of such an activity is thus hardly the way to advancement and success.

When in-house support staff are criticized for doing a poor job, their managers are likely to respond by hiring more people. An outside contractor knows that he will be tossed out and replaced by a better-performing competitor unless he improves quality and cuts costs.

The people running in-house support services are also unlikely to do the hard, innovative and often costly work that is required to make service work productive. Systematic innovation in service work is as desperately needed as it was in machine in the 50 years between Frederick Winslow Taylor in the 1870s and Henry Ford in the 1920s. Each task, each job, has to be analyzed and then reconfigured. Practically every tool has to be re-designed. . . .

The most important reason for unbundling the organization, however, is one that economists and engineers are likely to dismiss as "intangible": The productivity of support work is not likely to go up until it is possible to be promoted into senior management for doing a good job at it. And that will happen in support work only when such work is done by separate, free-standing enterprises. Until then, ambitious and able people will not go into support work; and if they find themselves in it, will soon get out of it.

It is hardly coincidental that the productivity decline in American factories set in as soon as finance and marketing were taking over from manufacturing in the early '60s as the main avenues of advancement into senior management. Nor is it coincidence that stock brokers have been plagued by recurrent "back office" crises despite steadily increasing employment and increasing investment in clerical and support work. Until very recently even the head of the back office (though responsible for half the firm's expenses), was at best a "titular" partner. Promotions, bonuses, but equally the time available on the part of top management were reserved by and large for traders, analysts and sales people.

They are "we"; the back office is "they." And one explanation why non-instructional costs in colleges and universities have risen twice as fast as instructional ones since World War II -- to the point where they now account for almost two-fifths of the total bill -- is surely that the people who run the dorms or the business office don't have Ph.D.s and are therefore non-persons in the value system of academia.

Forty years ago, service and support costs accounted for no more than 10% or 15% of total costs. So long as they were so marginal, their low productivity did not matter. Now that they are more likely to take 40 cents out of every dollar they can no longer be brushed aside. But value systems are unlikely to change. The business of the college, after all, is not to feed kids; it is teaching and research.

However, if clerical, maintenance and support work is done by an outside independent contractor it can offer opportunities, respect and visibility. As employees of a college, managers of student dining will never be anything but subordinates. In an independent catering company they can rise to be vice president in charge of feeding the students in a dozen schools; they might even become CEOs of their firms. If they have a problem there is a knowledgeable person in their own firm to get help from. If they discover how to do the job better or how to improve the equipment they are welcomed and listened to. The same is true in the independent firm that takes over customer accounting in the mutual-fund company.

In one large hospital-maintenance company, some of the women who started 12 or 15 years ago pushing vacuum cleaners are now division heads or vice presidents and own substantial blocks of company stock. As hospital employees, most of them would still be pushing vacuum cleaners.

Of course there is a price for unbundling. If large numbers of people cease to be employees of the organization for which they actually work, there are bound to be substantial social repercussions. And yet there is so far no other option in sight for giving us a chance to tackle what is fast becoming a central productivity problem of developed societies.

Japan: New Strategies for a New Reality

(This article originally appeared in The Wall Street Journal on Oct. 2, 1991)

Quietly, and with a minimum of discussion, the leading Japanese companies are moving to new business strategies. They are embracing two radically new theories: To do blue-collar manufacturing work in Japan is a gross misallocation of resources that weakens both the company and the national economy. And leadership throughout the developed world no longer rests on financial control or traditional cost advantages. It rests on control of brain power.

These companies are also fast restructuring their organizations on the assumption that the winner in a competitive world economy is going to be the firm that best organizes the systematic abandonment of its own products. And they are moving from Total Quality Management toward Zero-Defects Management based on drastically different principles and methods.

The Japanese now hold about 30% of the U.S. automobile market and expect to increase this share substantially in the next few years. Yet they also expect to stop exporting Japanese-made cars to the American market within the next three to five years; by 1995 or so, most Japanese marques sold in the U.S. should be manufactured in North American plants.

Similarly, the Japanese expect to have something like one-third of the automobile market of the European Economic Community by the year 2000 (whatever their present promises to the EC to the contrary), but again without exporting many cars from Japan. And Japanese multinationals -- Toyota, Honda, Sony, Matsushita, Fujitsu, the ceramics leader Kyocera, and the Mitsubishi companies -- are pouring staggering amounts of money into manufacturing plants in developing countries. They are in Tijuana on the U.S.-Mexican border, throughout South America, in Southern Europe, and in Southeast Asia.

The standard explanations for moving manufacturing out of Japan are "foreign protectionism" and "Japan's growing labor shortage." Both explanations are legitimate, but they are also smoke screens. The real reason is the growing conviction among Japan's business leaders and influential bureaucrats that manufacturing work does not belong in a developed country such as Japan.

Before youngsters can go to work on the assembly line, my Japanese friends say again and again, Japan pours $100,000 in school expenses into them. And then they have to get a middle-class income, lifetime security, a pension and health care. In Bangkok or in Tijuana, youngsters require very little capital investment in their educations; and they are "middle class" if paid a 10th the wages of the U.S. or Japan. Yet their productivity after two or three years of training is as high in Tijuana or in Bangkok as it is in Nagoya or Detroit. When you figure the enormous social capital invested in them, my friends say, the return that blue-collar workers make to society in developed countries is at most 1% or 2%; in Latin America or Indonesia, it's 20 times that.

Whenever I then argue that a country is highly vulnerable without a strong manufacturing base, they respond that the supply of young people in the developing world will be so large in the next 30 years that it's absurd to worry about the "manufacturing base," the way Americans do. Indeed it's my friends' social responsibility to Japan, they say, to make sure that as few as possible of its high-investment, high-cost young people are being misused for low-yield manufacturing work.

Instead, the new Japanese strategies call for total control of what now matters. To be competitive, the argument goes, Japan requires leadership in technology, marketing and management, and firm control of what my Japanese friends are beginning to call "brain capital."

The Japanese are willing to pay large sums to gain access to knowledge -- through a minority participation in a Silicon Valley computer specialist; through similar investments in U.S. and European pharmaceutical or genetics entrepreneurships; above all, through financing research in Western (mainly U.S.) universities. The direct financial return is usually zero. But the Japanese are paying not for dividends but access to the knowledge their partners will produce, and control over it -- or at least priority in using it.

Increasingly Japanese companies employ foreigners in their international operations, both as professionals and as executives. The large Japanese auto makers now all have design studios in Southern California and Westerners running their international marketing. But the use of the knowledge these foreigners produce is "proprietary" and tightly held within the Japanese management team. And while in the past some Japanese companies granted licenses on their knowledge to Western companies -- e.g., on some Japanese-developed cardiac drugs -- they are now revoking or not renewing them.

Every major Japanese industrial group now has its own research institute, whose main function is to bring to the group awareness of any important new knowledge -- in technology, in management and organization, in marketing, in finance, in training -- developed world-wide. On my last trip to Japan, a few months ago, I spoke at the 20th anniversary of one of these think tanks, that of the Mitsubishi Group. At lunch after my talk, one of the most respected elders of the Mitsubishi clan said to me: "In another 20 years the entire Mitsubishi Group will be organized around this research institute."

Everybody now knows that the Japanese can bring out a new product in half the time it takes their American competitors and in one-third the time it takes the Europeans. And everybody also knows that major U.S. companies are reorganizing their research and development work on the Japanese model, along cross-functional lines. But the Japanese are already moving to the next stage.

They are reorganizing R&D so that it simultaneously produces three new products with the effort traditionally needed to produce one. And they do this by starting out with a deadline for abandoning today's new product on the very day it is first sold. "The faster we can abandon today's new product, the stronger and the more profitable we'll be" is the new motto.

To most Western businessmen, this is madness. They believe that a product becomes more profitable the longer its product life -- for then the money spent on developing it has been written off. But "writing off" to the Japanese is useful to cut taxes but otherwise self-delusion.

Money spent on developing a product or a process is not "investment" to the Japanese; it is "sunk cost." But the main reason the leading Japanese businesses are now shifting the life cycle of their products is their conviction that the only alternative is for a competitor to do so -- and then the competitor will have not only the profits but the market.

My Japanese friends acknowledge that some Western companies -- 3M, for example -- have long operated on the policy that 70% of their sales five years hence will have to come from products that do not exist today. But these companies rely on a spontaneous upswelling of entrepreneurship from within.

By deciding in advance that they will abandon a new product within a given period of time, the Japanese force themselves to go to work immediately on replacing it, and to do so on three tracks:

One track ("kaizen") is organized work on improvement of the product with specific goals and deadlines -- e.g., a 10% reduction in cost within 15 months and/or a 10% improvement in reliability within the same time, and/or a 15% increase in performance characteristics -- and enough in any event to result in a truly different product. The second track is "leaping" -- developing a new product out of the old. The best example is still the earliest one: Sony's development of the Walkman out of the newly developed portable tape recorder. And finally there is genuine innovation.

Increasingly, the leading Japanese companies organize themselves so that all three tracks are pursued simultaneously and under the direction of the same crossfunctional team. The idea is to produce three new products to replace each present product, with the same investment of time and money -- with one of the three then becoming the new market leader and producing the "innovator's profit."

Finally, the leading Japanese companies are moving from Total Quality Management to Zero Defects Management. "We can't use TQM," one of the top manufacturing people at Toyota recently said. "At its very best -- and no one has reached that yet -- it cuts defects to 10%. But we turn out four million cars, and a 10% defect rate means that 400,000 Toyota buyers get a 100% defective car. But Zero-Defects Management is now possible and actually not too difficult."