CMR143
Winter   1999    I    Vol.41.   N □ . 2    i    REPRINT SERIES
\/I QT1 O (V,
IVlolJLaH1
Knowledge-Worker Productivity: The Biggest Chalienge
Peter F. Drucker
©1999 by Tile Regent: of the Jrwersity of California
Berkeley
Knowledge-Worker Productivity:
The Biggest Challenge
Feter f, Drucker
The must important, and indeed the iruiy unique, contribution ot management in the 2.0th century was the fiity-lold increase in the productivity of the manmt\^rkerin manufacturing. The most im-porlant contribution management needs to make in ihe 21*1 remury is similarly to increase the productivity of knowledge work and knowledge, workers. The most valuable assets of a 20'"-century company was its production equipment. The most valuable asset nf a 2 lucent ury institution (whether business or nonbusiness) will be its knowledge workers ami their productivity.
The Productivity of the Manual Worker
First, We. ritusl take a look at where we are. K was only a littte over a hundred years ago that for the first time fin educated person actually looked at manual work and manual workers, and then began to study bothr The Greek poet Hesiod (eighth eenlury B.C) and Lhe Roman poet Virgil (700 years later) sang about the work oJ the farmer. Theirs are still among Lhe lineal poems tn any language, but neither the work they sang about nor their farmers; bear even the most remote resemblance tn reality, nor were they meant to have any. Neither Hesiud nor virjnl ever held a sickle in their hands, ever herded sheep, or even looked at the people who did either. When Karl Marx, 1900 years after Virgil, came to write abouL manual work and manual workers, he too never looked at either, uor had he ever as much as touched a machine. The first man to do both—that is, to work as a manual worker and then to study manual work—was Frederick Winslow Taylor {1676-1913),
Throughout history there have been steady advances in what we today call "productivity" (the term itself is barely HIty years old). They were the result
CAJFkRNtV^NA^rr.N""R.SV1EW  VOL4LNC12 WINTER
ínc^dks-Wbrter Product^ Tŕ* Biggfeť Oaflefije
of new tools, new methods, and new technologies; they were advances in what the economist calls "capital." There were lew advances throughout the ages in what the economist calls "labor"—that is, in the productivity of the worker. It was axiomatic throughout history that workers could product* more only by working harder or by working longer hours. The 19"'-cemury economists disagreed about must things as much as economists do today. However, they all agreed—from David Rieardo through Karl Marx—that there are enormous differences In skill between workers, but there are none in respect 10 productivity other than between hard workers and lazy ones, or between physically strong workers and weak ones. Productivity did noi exist. It still is an "extraneous factor" and not part of ihe equation in most contemporary economic theory' (e.g., in Keynes, but also in that of the Austrian School!.
In the decade after Taylor first looked at work and studied it, the productivity of the manual worker began its unprecedented rise. Since iheu, it lias been going up steadily at the rale of 3% per annum compound—which means it has been risen fifty-fold since Taylor On this achievement resl allot the economic and social gains of the 20lh century. The productivity of the manual worker has created what we now call "developed" economies. Before Taylor, there was no such thing—all economies were equally "underdeveloped." An underdeveloped economy today—or even an "emerging" one—is one that has notr or at least lias not yet, made the manual worker more productive.
The Principles of Manual-Work Productivity
Taylor's principles sound deceptively simple. The first step in making the manual worker more productive is to look at the task and to analyze its constituent motions. The next step is to record each motion, the physical elfort it takes, and the time it takes. Then moiions that are not needed can be eliminated; and whenever we have looked at manual work, we have found that a great many of The traditionally most-hallowed procedures turn out to be waste and do not add anything. Then, each ol the motions that remain as essential to obtaining the finished product is set up so as to he done the simplest way, the easiest way. the way that puts the least physical and mental strain on the operator, and the way that requires the least time. Next, these motions are put together again into a "job" that is in a logical sequence Finally, the tools needed to do the motions are redesigned. Whenever we have looked at any job—no matter for how many thousands of years it lias been performed—we have found that the traditional tools are wrong for the task. This was the case, for instance, with the shovel used to carry sand in a foundry (the first task Taylor studied 1, It was the wrong shape, the wrong size, and had the wrong handle, We found tliis to he equally true of the surgeon's traditional tools. Taylor's principles sound obvious—effective methods always do, However, it took Taylor twenty years of experimentation to work them out.
ho       CA= JFOftNlA MANAGEMENT' AťiVlfcW  VCt,4I.NQ? WINTER
Kricrwledge-V'/artŕJ" R^dixfcvtr/'i'h<s fiigges: Challenge
Over these last hundred years, there have been countless further changes, revisions, and refinements. The name by which the methodology goes has also changed over the past century. Taylor himself first called his method "Task Analysis" or Tiisk Management." Twenty years later it was re-christened "Scientific Management." Another twenty years later, after the First World War, it eame to be kilOWS as "Industrial Liighieeriug" hi the U.S. and Japan, and as " Katión-ajtsaäoťi* in Germany.
To proclaim that one's method "rejects" Taylor or "replaces" him is almost standard "public relations." For what made Taylor and his method so powerful has also made íl unpopular. What Taylor saw when he actually looked at work violated everything poets and philosophers had said about work from Hesiod and Virgil to Karl Marx. They all celebrated "skill." Taylor showed that in manual work there is no such thing. There are only simple, repetitive motions. What makes them mure productive is knowkdgt, that is. the way the simple, unskilled motions are put together, organized, and executed. In fact, Taylor was the first person to apply knowledge to work.1
This also earned Taylor the undying enmity of the labor unions at his time, all of which were craft unions and based on the mystique of craft skill and their monopoly on it. Moreover. Taylor advocated—and this is still anathema to a labor union—that workers be paid according to their productivity—that is, for their output, rather than for their input je.g,r for hours worked). However, Taylor's definition of work as a series of operations also largely explains his rejection by the people who themselves do not do any manual work; the descendants of the poets and philosophers ol old, the Literati and Intellectuals, Taylor destroyed the romance of work. Instead of a "noble skill," it becomes a series of simple motions.
Nevertheless, every method during these past hundred years that has had the slightest success in raising the productivity of manual workers—and with it their real wages—has been based on Taylor's principles, no matter how loudly his antagonists proclaimed their differences with Taylor. This Ls true of "work enlargement," "work enrichment/ and "job rotation"—all of which use Taylor's methods to lessen the worker's fatigue and thereby increase the worker's productivity, It is also true of such extensions of Taylor's principles ol task analysis and industrial engineering as Henry Ford's assembly fine {developed alter 1914, when Taylor himself was already sick, old, and retired). It is just as true of the Japanese "Quality Circle," "Continuous Improvement "(AdRsntJ* and "Jusi-In-Time Delivery."
The best example, however, Is W. Edward Deming's "Total Quality Management/ What Demiitg did—and wiiat makes Total Quality Management effective—is to analyze and organize the job exactly the way Taylor did. However, he also added Quality Control (around 1940) that was based on a statistical theory that was only developed ten years after Taylor's death. Finally, in the 1970s, Denting substituted dosed-circuit television and computer simulation lor
CA.. IFOSžsaA ^AfMAGfrME? JT fi EVlEW  VOL II.NO 2  WIN T W 1999 tí I
Knowtedge-VVbrter Productivity. The Bigg«t Qdnp
Taylor's stopwatch and motion photos. Deming's Quality Control Analysis arc Che spiuiag image of Taylor's Efficiency Engineers and function the same way.
Whatever his limitations and shortcomings—and he had many—no other American, not even Henry Pont has had anything like Taylor's impact. "Scientific Management" (and its successor "Industrial Engineering"] is the one American philosophy that has swept the world—more so even than the Constitution and the Federalist Papers. In the past century there has been only otic worldwide philosophy that could compete with Taylor's: namely, Marxism. In the end, Taylor has triumphed over Marx.
During the First World War, Scientific Management swept through the U.S. together with Ford's Taylor-based assembly'line. In she 1320s, Scientific Management swept, through Western Europe and began to be adopted in Japan.
During the Second World War, both the German achievement and the American achievement were squarely based on applying Taylor's principles to Training. The German General Staff, after having lost the First World War, applied "Rationalization" (i.e., Taylor's Scientilic Management) to the job of the soldier and to military training. This enabled Hitler to create a superb fighting machine in the six short years between his coming to power and 1939. In the U.S., the same principles were applied to the training of an indusirial work force, first tentatively during the First World War and then with full power, during the Second World War. This enabled the Americans to outproduce the Germans, even though a larger proportion of the U.S. than die German male population was in uniform and thus not in industrial production. Then, training-based Scientific Management gave the U.S. civilian work forte more than twice—if not three times—the productivity of the workers in Hitler's Germany and in Hitler-dominated Europe. S'cieniific Management thus gave the U.S. the capacity to outnumber both Germans and Japanese on the battlefield and yet soil outproduce both by several orders nf magnitude.
Since 1950, economic development outside the Western World has largely been based on copying what the U.S. did in the Second World War, i.e., on applying Scientific Managcmeni to making the manual worker more productive. All earlier economic development had been based on lechnological innovation—first in France in the 18,lL century, then En Great Britain from 1760 until IB50, and finally in the new economic Great Powers, Germany find the U.S., In the second half of the J9"11 century. The non-Western countries that developed after the Second World War, beginning with Japan, eschewed technological innovation. Instead, they imported the training thai the U.S. had developed during the Second World War based on Taylor's principles and they used it to make highly produclive, almost overnight, a Slill largely unskilled and pre-industrial work force. (In Japan, for instance, almost two-thirds of the working population were still, in 1950, living on the land and unskilled in any work except cultivating rice). However, while highly productive, this new work force was still—for a decade or more—paid pre-industrial wages so thai ihese countries—first Japan, then Korea, then Taiwan and Singapore—could produce the
hi        CAJFORNiA MANAGt^tiNT ftfcVlKW   UOLfl.NCM   WINTER 1999
Kj^cw1edg*-VvbH<*f" ProductfutyThe BiggeE Challenge
same manufactured products as the developed countries, but at a fraction of their labor costs.
The Future of Manual-Worker Productivity
Taylor's approach was designed for manual work in manufacturing, and at first applied only to it. Nevertheless, even within these traditional limitations, Taylor's approach sit!! lias enormous scope. It is still going to be ihe organizing principle in countries in which manual work, and especially manual work in : . ii. : .i        is ihe growth sector of the society and economy—that is, "Third World* countries with very large and still growing numbers ol young people with little education and little skill.
However, there is equal—or even greater—opportunity In the tfewhped countries 10 organize non-rnanulaeturing production (Lc.f production work in services) on ihe production principles now being developed in manufacturing —and that means applying Industrial Engineering to the jab and work of the individual service worker. There is equally a tremendous amount of knowledge work—including work requiring highly advanced and thoroughly iheorciieal knowledge—that includes manual operations. The productivity of these operations also requires Industrial Engineering.
Still, in developed countries, ihe central challenge is no longer to make manual work more productive—after all, we know how to do it. The central challenge will be to make knowledge workers more productive. Knowledge workers are rapidly becoming the largest single group in the work force of every developed country. They may already compose two-fifths of the U.S. work force —and a still smaller but rapidly growing proportion of the work force of all other developed countries, IE is on iheir productivity, above all, that the future prosperity—and indeed ihe future survival—of the developed economies will increasingly depend.
What We Know About Knowledge-Worker Productivity
Work on the productivity of the knowledge worker has barely begun, hi lerms of actual work on knowledge-worker productivity, we will be in the year 2000 rouglily where we were in ihe year 1900 in lerms of ihe productivity of ihe manual worker. Nevertheless, we already know infinitely more aboui the productivity of the knowledge worker than we did then about that of the manual worker. We even know a good many of the answers. We also know the challenges to which we do not yet know ihe answers, and on which we need to go to work-Six major factors determine knowledge-worker productivity. ■ Knowledge-worker productivity demands that we ask the question: "What is the task?"
CA..IfORMA1ANAdffMffNT^£W   VOL 41. NQ 2   WINTER 1999 83
■ It demands thai we impose the responsibility for their productivity on the individual knowledge workers themselves. Knowledge Workers hu\p to manage themselves. They have to have autonomy.
■ Continuing innovation has to be part ol the work, the task and the responsibility ol knowledge worked.
■ Knowledge work requires continuous learning on the pan ol the knowledge worker, but equally continuous teaching on the part of ihv. knowledge worker.
■ Productivity ol the knowledge worker is not—at least not primarily— a matter of the quantity ol output. Qualify is at least as important.
* Finally, knowledge-worker productivity requires thai die knowledge worker is both seen and treated as an "asset" rather than a "con." It requires that knowledge workers want to work for the organization in preference to ah other opportunities.
Each ol these requirements (except perhaps the last one} is almost the exact opposite ol what Is needed to increase rfie productivity of the manual worker. In manual work, ol course, quality also matters. However, lack of quality is a restraint. There has to be a certain minimum quaifty standard. The achievement of Total Quality Management—that is, of the application ul century Statistical Theory to manual work—is the ability to cut (though not entirely to eliminate) production thai falls below this minimum standard,
ui most knowledge work, quality is not a minimum and a restraint. Quality is the essence of the output. In judging the performance of a teacher, we do not ask how many students there can be in his or her class, we ask how many students learn anything—and that's a quality question. In appraising the performance of a medical laboratory, the question of how many tests it can run through its machines is quite secondary to the question of how many tests re-suhs are valid and reliable. Tins Is true even for the work ol the file clerk.
Productivity of knowledge work therefore has to aim iirst at obtaining quality—and nol minimum quality but optimum if not maximum quality. Only l hen can one ask: "What is the vo!umer the quantity of work?'This not only means that we approach the (ask of making more productive the knowledge worker from the quality of the work rather than the quantity, it also means that we will have to Icaro to define quality.
What Is the Task?
The crucial question in knowledge-worker productivity is: What is the task? It is also the one most at odds with manual-worker productivity. In manual work, the key question is always: How should the wt?rk he dene? hi manual work, the task is always given. None of the: people who work on manual-worker productivity ever asked: "What is the manual worker supposed to do?" Then only question was: "How does the manual worker best do the job?" This was just as true of Frederick W. Taylor's Scientific Management as it was true of the people
M        CALJfOftNlA !*IAWGMNT WMW   VOL -II. NCI 2 WINl'fcW
Kr^ledge-YVbrkfr' h^duetivity: i    Biggest Challenge
at Sears Roebuck or the Ford Motor Company who first designed she assembly line, and as it is true ol W. Edward Deming's Total Quality Control.
Again, in knowledge work the key question is: What is the task? One reason for this is thai knowledge work, unlike manual work, does not program the worker. The worker on the atuomohile assembly line who puts on a wheel is programmed by Lhe simultaneous arrival of the car's chassis on one line acid the wheel on Lite other, line. The farmer who ptbwsa field in preparation for planting does not climb out of his tractor to take a telephone call, to attend a meeting, or to write a memo. Whatxs to be done is always obvious in manual work.
However, in knowledge work the task does not program the worker. A major crisis in a hospital, such as when a patient suddenly goes into coma, does oi course control the nurse's task and programs her; but otherwise, ii Is largely the nurse's decision whether to spend time at the patient bed or whether to spend lime fillins om papers, Engineers are constantly being pulled off their task hy having to write a report or rewrite it, by being asked to at Lend a meeting, and so on. The job of the salesperson In the department store is to serve the customer and to provide the merchandise the customer is interested fn or should become Interested in. Instead, the salesperson spends an enormous amount of time on paperwork, on checking whether merchandise is in stock, on checking when and how it can be delivered, and so on—ail things that take salespeople away from the customer and do tiot add anyiliing to their productivity in dnittg what salespeople are being paid for, which is to sell and to satisfy the customer.
The first requirement in tackling knowledge work is to find out what the task is so as to make it possible so concentrate knowledge workers on the lask and to eliminate everything else—at: least as far as it can possibly be eliminated. This requires that the knowledge workers themselves define what the task is or shoofd be—and only the knowledge workers themselves can do that. Work on knowledge-worker productivity therefore begins with asking the knowledge workers themselves: What in your task? What should it be? What should y$u be expected to eoaMbute? and what hampers yov in doing your task andtkauM be eliminated?
Knowledge workers themselves almost always have thought through these questions and can answer them. Still, it then usually takes time and hard work to restructure their jobs so that they can actually make the contribution they are already being paid fur. However, asking the questions and taking action on the answers usually doubles or triples knowledge-worker productivity, and quite List,
Nurses in a major hospital were asked these questions. They were sharply divided as to what their Lask was, with one group saying "patient care" and another saying "satisfying the physicians." However, they were in complete agreement on the things (hat made them unproductive. They called I hem "chores'7 —paperwork, arranging flowers, answering the phone calls of patients' relatives, answering lite padems' bells, and so on. Ail—or nearly all—ol these could be
CA-.tfORNlA MANAGE MSN"!' ft tVffW  VCX ^ I, NQ 2   VtfNTfR 1999 ttS
turned over to a non-nurse floor clerk, paid a fraction of a nurse's pay. The productivity ;if the nurses on the flour immediately more: than doubled, as measured by the lime nurses spent at the patients' beds. Patient satisfaction more than doubled and turnover of nurses (which had been catastrophically high) almost disappeared—all within four months.
Once the task has been defined, the next requirements can be tackled, and they will be tackled by the knowledge workers themselves. These re<pure-ments are:
• Knowledge workers' rtupjtisibility for their own contribution. It is the knowledge worker's decision what he or she should be held accountable for In terms of quality and ipiantity with respect to time and with respect to cost. Knowledge workers have to have autonomy and that emails responsibility.
» Continuous innovation has w be built into ike hwwkiige worker's jcb-' C&iitirtttQttf tctitiiittg and tontimnyus leaching have to be built into the job. One central requirement dif knowledge-worker productivity remains. We have to answer the question: What is quality? In some knowledge work—and especially in some work requiring a high degree- of knowledge—we already measure quality. Surgeons, lor example, are routinely measuredr especially by their colleagues by thuir miccl-ss rales in difficult atul dcrn^erou?. procedures (e.g,r by the survival rates of their open-heart surgical patients or the full recovery rates of their orthopedic-Surgery patients). By and large, we mainly have judgments rather than measures regarding the quality ot a great deal of knowledge work. The main trouble is. however, noi ihe difficulty of measuring quality-It is the difficulty—and more particularly the sharp disagreements—in defining what the task is and what it should be.
The best example of this is the American school system. As every one knowsr public schools in the American Inner city liavc become disaster areas. Next to them—in the same location and serving the same kind of children—are private i mostly Christian] schools in which the kids behave well and learn well-There is endless speculation to explain these enormous quality differences. A, major reason is surely that the two kinds of school define their tasks differently. The typical public school defines its lask as "helping the underprivileged/ while the typical private school {and especially the Parochial Schools of the Catholic Church J delinc their task as "enabling those who want to learu. to learn." One therefore is governed by its scholastic failures, the other one by its scholastic successes.
Similarly the research departments at two major pharmaceutical companies have totally different results because ihey define their tasks differently. One-sees its task as not having failures, that is, in working steadily on fairly minor but predictable improvements in existing products and for established markets. The other one defines its task as producing "breakthroughs" and therefore courts risks. Both are considered fairly successful—hy themselves, by their own top
B6        CALIFORNIA H^NAGEhlEN^ REVIEW  TOl.4l.Na7   VWTtf* |99¥
fc^QwIedgfrWorker rVodurtiwtyzThe Bigg^^ Chattel
managements, and by outside analysis. Yet each operates quite differeuily and quite differently defines its own productivity and tltat of lis research scientists.
To define quality in knowledge work and to convert the definition into knowledge-worker productivity is thus to a large extent a matter of defining the task. It requires the difficult, risk-takutg, and always controversial definition as to what "results" are for a given enterprise and a given activity. We therefore actually knttw how to do iir Nevertheless, the question is a completely new one for most organizations and also for most knowledge workers. To answer it tvqitires controversy, ttquitw dissent.
The Knowledge Worker as Capital Asset
In no other area Is the difference greater between manual-worker productivity and knowledge-worker productivity than hi their respective eetmotttidi. Economic theory and most business prattfire sees manual workers as a owf. To be productive, knowledge workers must be considered a aipitol asset. Costs need to be controlled and reduced. Assets need to be made to grow.
To be sure, in managing manual workers we learned fairly early that high turnover {i.e., losing workers) Is very costly. The Ford Motor Company as is well known, increased the pay of skilled workers from eighty cents a day to (5.00 a day on January ]r 1P14. It did so because its turnover had been so excessive as to make its labor costs prohibitively high; it had to hire 60,000 people a year to keep 10,000. Evui so. everybody (including Henry Ford IdmsclL who had at first been bitterly opposed to this increase) was convinced that the higher wages would greatly reduce the company's profits, instead, in the very first year, profits almost doubled. Paid £5.00 a day, practically no workers left—in fact, the Ford Motor Company soon had a waiting list.
However, short of the costs of turnover, rehiring, retraining, and so on, the manual worket is stifi being seen as a cost. This is true even in Japan, despire the emphasis on lifetime employment and on building a "loyal," permanent work force. The management ol people at work, based on millemrfa ot work being almost totally manual work, still assumes that with few exceptions (erg,r higldy skilled people) one manual worker is like any other manual worker.
This fs definitely not true for knowledge work. Employees who do manual work do not own the means of production. They may, and often do, have a lot of valuable experience, but that experience is valuable only at the place where they work, ft is not portable. Knowledge workers, however, ewn the means of production. That knowledge between their ears is a totally jwirtable and enormous capital asset. Because knowledge workers own their means of production, they are mobile. H may not be true for most ol thetn that the orga* nization needs them more than they need Lhe organisation. For most of them it is a symbiotic relationship in which they need each other in equal measure. It is
CAJFORNIA ""lANAGFrMfrN? HfrVlffW   VCX.1l.Ntt".*   WiNT^ |W9 A/
Knwfedge-Workfir iVoducWtyThe Biggefr Challenge
not true, as it was ior the manual worker hi modern industry, that they need (hi job much Htorc than the job needs them.
Management's job is to preserve the assets of thr institution in its carft. What does this mean when the knowledge ol the individual knowledge worker becomes an asset—and, in more and mure casts, the main asset—ol an institution? What does this mean for personnel policy? what i*i needed to attract and to hold the highest producing knowledge workers? What is needed to increase [heir producJivily and to convert their increased productivity into performance capacity ior the organization?
The Tech nologtsfcs
A very large number of knowledge workers do both knowledge work ami manual work. 1 call them "teehnoJogists." litis group includes oeople who apply knowledge of die highest order.
Surgeons preparing Tor an operation id correct a brain aneurysm before it produces a lethal brain hemorrhage, spend hours in diagnosis before they cut —and that requires specialized knowledge of the highest order. Again, during the surgery, an unexpected complication may occur which calls for theoretical knowledge and judgment, both oi the very highest order. However, ihe surgery itself is manual work—and manual work consisting of repetitive, manual operations hi which the emphasis is on speed, accuracy, and uniformity. These operations are studied, organized, learned, and practiced exactly like any manual work—That isr Ivy the same methods Taylor first developed ior factory' work.
The technologist group also contains large numbers of people in whose work knowledge is relatively subordinate—though it is always crucial. The file clerk's job—and that ol the clerk's computer-opera tor successor—requires a knowledge of the alphabei ihat no experience can Lciaeh. This knowledge is a small part of an otherwise manual task, but It is its foundation and is absolutely crucial.
Technologists may be ihe single biggest group of knowledge workers. They may also be the fastest-growing group. They include the great majority of htialth-care workers: !ab-technicians; rchabilitaiion technicians; technicians in imaging such as X-ray ultrasound, magnetic-resonance imaging; and so on. They include dentists and all dental-support people. They include automobile mechanics and all kinds of repair and installation people, in factr die technologist may he tlie true successor to the 19iil and SO1*1 century skilled workers.
Technologists arc- also the one group in which developed countries ran have a true and long-lasting competitive advantage. When It comes 10 truly high knowledge, no country can any longer have much of a lead the way 19,ft century Germany had through its University. Among theoretical physicists, mathematicians, economic theorists, and the like, there is no "nationallty," Any country can, at fairly low cost, train a substantial number of high-knowiedge
CAUFORMA MANAGEMENT REVIEW  VOL 41, NQ 2  WtNTTft IW?
people. India, lor ins Lance r despite her poverty, has been training fairly large numbers ol first-rate physicians and first-rate computer programers. Similarly, there is no "nationality'' In respect to the productivity of manual labor. Training based on Scientific Management Jws made all countries capable of attaining —overnight—the manual-worker productivity of the most advanced country, industry, or company. Only by educating technologists can the developed countries still have a meaningful and lasting competitive edge.
The U.S. is so far the only country thai has developed this advantage through its unique nationwide systems of community colleges. The community college was actually designed {beginning in the 1920s} to educate technologists who have bjth the needed theoretical knowledge and the manual skill. On this, I am convinced, rests both The still huge productivity advantage of the American economy and the (so far unique) American ability lo create, almost overnight, new and different Industries.
Currently, nothing quite tike the American Community College exists in any other nation. The famous Japanese school system products cither people prepared only for manual work or people prepared only For knowledge work. Not until the year 2003 is the first Japanese institution devoted to train technologists supposed to get started. The even more famous German apprenticeship system (started lb the 1930s) was otic of the main factors in Germany's becoming the world's leading manufacturer. However, it focused—and still focuses— primarily on manual skills and slights theoretical knowledge, ft is thus in danger of becoming rapidly obsolete.
Other developed countries should be expected to catch up with the U.S, fairly fast. "Emerging" or 'Tliird World' couritries are, however, likely to be decades behind—in part because educating technologists is expensive, in part because in these countries people of knowledge still look down with disdain, if not with contempt, on working with one's hands. "That's what we have servants lor" is still their prevailing attitude. However, in developed countries—and again foremast in the U.S.—more and more manual workers are going ro be technologists, to increase knowledge-worker productivity increasing the productivity of technologists deserves to be given high priority
The job was actually done more than seventy years ago by the American Telephone Company fATfiT) ior its technologists, the people who install, maintain, and replace telephones. By the early 1920s, The technologists working outside the telephone office and at the customer's location had become a major cost center—and at the same lime a major cause oi customer uuhappiness and dissatisfaction. It took about five years or so {from 1920 until 1925) for AT&T—which had by that time acquired a near monopoly on providing telephone service in the United States and in parts of Canada—to realize that the Fask was not installing, maintaining, repairing, and replacing telephones and telephone connections. The ta&wKitp&eQtt tttetisf&diMUetiteT* Once they realized this, it became fairly easy to organize ihe job- it meant, first, ihat the technician* ihemselves
CALIFORNIA MANAGEMENT REVIEW  UGMl»f«a  WNTEft |9^9 89
^ncwtecge-Wbrfcer ^rodudwit/ltie Bi^est Challenge
had id define what "satisfaction" meant. The results were standards Lhat established that every order for a new telephone or an additional telephone flmneo tion would have to be satisfie d within 4« hours, and ihat every request tor repair would have to be satisfied the same day if made before noon, or by noon the following day. Then it became dear lhat the individual service people—in those days all menh ol cuiirse—wutild have to be active participants in such Je-cisions as whether to have one person installing and replacing telephones and another one maintaining and repairing them or whether the same people had to be able to do all jobs (which in the end turned out to be the right answer). Then these people had to be taught a very substantial amount ot theoretical knowledge—and in those days few of them had more than six years ot schilling. They had to understand how a Telephone work4;. They had to understand how a switchboard works. They had to understand how the telephone system works. These people were not qualified engineers nor skilled craftsmen, but they had to know enough electronics lo diagnose unexpected problems and be able to eope with them. Then they were trained in the repetitive manual opera Li on or in the "one rlghf way' {that is, through the methods of Scientific Management) and they made the decisions (e.g., where and how to connect the individual telephone to the system and wliat particular kind of telephone and service would be the most suitable for a given home or a given office). They had lo become salesmen In addition to being servicemen.
Finally, lhe telephone company faced the problem how tn define quality. The technologist had lo work by himself. He could not be supervised. He, therefore, had lo define quality, and he had lo deliver it. ll took another several years before: ihat was answered- Al first the telephone company thought that ihis meant a sample cheek, which had supervisors go om and look at a sample (maybe every 2Cth or 30th job done by an individual service person) and check it for quality. This very soon turned out to be the wrong way of doing the job, annoying servieemen ami customers afike. Then the telephone company defined quality as "no complaints"—and they soon found out that only extremely tin-happy customers complained. It then had to redefine quality as "positive customer satisfaction." In the end, this then meant thai the serviceman himself controlled quality {e.g., by calling up a week or ten days after he had done a job and asking the customer whetlicr the work was satisfactory and whether there was anything more The technician could possibly do to give the customer the besl possible and most satisfactory service).
1 have intentionally gone into considerable detail in describing this early example because it exemplifies the three elements for making the worker who is both a knowledge worker and a manual worker both effective and productive, ■ First, there is the answer to the question "What is the task?"—the key question in making every knowledge worker more productive. As the example of the Bell System shows, this Is not an obvious answer. As lhe Bell System people learned, the only people who knew the answer to This were the technologists themselves. In fact, until They asked the
90        C-Ai.JfORMA MAMACsEMEN^ REVIEW  VOMt.NC^   Wfffi l?9?
'Qnowledge-Wc^ker productivity The Biggest DiaBenge
technologists, they floundered. However, as soon as the technologists were asked, the answer came back loud and clean "a satisfied customer/'
* Then, the technologists bad to take full responsibility for giving customer satisfaction, thaT is, for delivering quality. This showed what Jfr wal knowledge the technologist needed. Only Then could The n.r-rr:! of the job he organized for manual-worker productiviTy.
■ Above alL this example shows that technologists have to be treated as knowledge wark£i$. No matter how important the manual part of their work—and it may take as much time as it did in the case of the AT&T installers—the focus has to be on making the technologist knowledgeable, responsible, and prodiuTive as a knowledge worker.
Knowledge Work as a System
Productivity of the knowledge worker will almost always require that the work itsetfbe reslmctured and be made part of a system, One example is servicing expensive equipment, such as huge and expensive earth-moving machines. Traditionally litis had been seen as distinct and separate from the job of making and selling the machines. However, when the VS. Caterpillar Company, the world's largest producer ol such equipment, asked "What are we getting paid for?" the answer was "We are not getting paid for machinery. We are getting paid for what the machinery does ai the customer's place of business. That means keeping the equipment running, since even one hour during which the equipment is out of operation may cost The customer far more titan the equipment itself," in oilier words, the answer lo "What is our business?" was "Serviced This then led to a total restructuring of operations all the way hack to the faciory in order that the customer could be guaranteed continuing operations anil Immediate repairs or replacements. The service representative, usually a technologist, has become the true "decision maker."
As another example, a group of about 21 orthopedic surgeons in a Midwestern U.S. city, have organized themselves as a "system" to; produce the highest quality work; make optimal use of the limited and expensive resources of operating and reetrvery rooms; make optimal use of the supporting knowledge people such as anesthesiologists or surgical nurses; build continuous learning and continuous innovation into the work of the entire group and of every member thereof; and. finally, minimize costs. Each of the surgeons retains full control of his or her practice. He or she is fully responsible for obtaining and treating the individual patient, Traditionally, surgeom schedules surgeries early in ihe mom* ing. Hence, operating moms and recovery moms are standing empty most of the time. The gtoup now schedules the use of operating and recovery rooms for the entire group so that this scarce and extremely expensive resource is utilized ten hours a day. Tfie group, as a group, decides on the standardization ot tools and equipment so as to obtain tire highest quality at the lowest cose. Finally, the group has also built quality control Into its system. Every three months three
CAI.FORNIA MANAGEMENT r\£VlEVY  VQL4I.NQJ  WINTER IW9 91
Krvw^dga-Wufka1 HTOductJvity:Ttw Biggest Ctolerij|e
surgeons are designated to scrutinize every operation done by each of the members—Lhe diagnosis, the surgery, the a fter-treatment. They [hen Mil down with the individual surgeons and discuss their performance. They suggest where there is need for improvement and they also may recommend that a certain surgeon be asked to Leave the group when his or her work is not. satisfactory. Each year, the quality standards that these supervising committees apply are discussed with the whole group arid are raised, often substantially,    a result; this group now does: almost four limes as much work as it did before* It has cul the costs by 30%, hall of it by cutting back on the waste of operating and recovery rooms and half by standardizing tools and equipment. In such measurable areas as success rate*: in knee or shoulder replacements and in recovery a iter sports injuries, it has greatly improved its results,
Whtit to do about kncwlcdgc-workcr productively is thus largely known. So is hůwtti do it.
How to Begin?
Making knowledge workers more productive requires changes in basic altitude, whereas making lhe manual worker more productive only required telling the worker how to do the iob. Furthermore, making knowledge workers more productive requires changes in attitude not only on the pair of the individual knowledge worker, but on the pan ní the whole organization. Il therefore has to be "piloted," as any major change should be. lhe first step is to find an area In tin: organization where there b a group of knowledge workers who are receptive. (The orthopedic surgeons, for instance, first had their new ideas tried out by lour physicians who haJ long argued lor radical changes.} The next step is to work consistently, patiently, and for a considerable length oi time with this small group. The lirsl attemptsr even if greeted with great enihusiasiit, will almost certainly run into all kinds of unexpected problems. It is only after the productivity of this small group oi knowledge workers has been substantially increased thai the new ways of doing the work can be extended to a larger area, ilnot to lhe entire organization. At this jxiint, the main problems will be known, ■iudi as where resistance can be expected {e.g., from middle management) or what changes in taskr organization, measurements, or fittiiudes are needed tor full effectiveness. To bypass the pilot stage—and there is always pressure to do so—only means thai the mistakes become public while the successes slay hidden. It only means discrediting the entire enterprise. 1Í properly piloted, a great deal can he done to improve knowledge-worker productivity.
Knowledge-worker productivity is the biggest of the 2T-century management challenges, hi the developed coimiries, it is their first stovřvái requirement. In no other way can the developed countries hope to maintain themselves, let alone maintain their leadership and their standards of living. In the 20<]t century, this leadership very largely depended on making the manual worker more productive. Any country, any industry, any business can do that today using the
97        CAUFOsiMA HAN^GfMt^" ftfcVlEW  VOL.Il.NCU WINTEBI5W
methods that the developed countries have worked out and put into practice in lhe 120 years since Frederick Window Taylor first looked at manual work* Anybody today, any place, can apply those policies to training, the organization of work, and the productivity of workers—even if they are barely literate, if not illiterate, and totally unskilled.
Above all, the supply of young people available for manual work will be rapidly shrinking in the developed countries—in tin: West and in Japan very fast, in rbe USr somewhat more slowly—whereas the supply oi such people will still grow fast in the emerging and developing countries for at least another thirty or forty years. The only possible advantage developed countries can hope u> have Is in the supply ol people prepared, educated, and trained for knowledge work- There, for another fifty years, the developed countries can expect to have substantial advantages, both in quality and in quantity. Whether this advantage will translate into performance depends on the ability of the developed countries —and ol every Industry in it, of every company in it, ol every institution in it— to raise the productivity ol the knowledge worker and to raise it as fast as the developed countries have raised the productivity of the manual worker in the last hundred years.
The countries and the industries that have emerged as the leaders in the last hundred years in the world are the countries and the industries thai have led in raising lite productivity of the manual worker—the U.S. first, Japan and Germany second. Flhy years from now, if not much si>onerr leadership in the world economy will have moved to the countries and to the industries ihat have most systematically and most successfully raised knowledge-worker productivity,
The Governance of the Corporation
What does the emergence ol the knowledge worker and of knowledge-worker productivity mean for the govswancr fffiki corporation? What does it mean for the future and structure of the economic system?
In the ]ast ten or fifteen years, pension funds and other institutional investors became lhe main share owners of the equity capital of publicly owned companies in all developed countries. In the U,S„ this has triggered a furious debate on the governance of corporations. With the emergence of pension funds and mutual funds as the owners of publicly owned companies, power has shifted to these new owners. Similar slilfts in both the definition oi the purpose of economic organisations {such as tht: business corporation) and their governance tan be expected to occur in ail developed countries.
Within a fairly short period of time, we will face the problem of the governance of corporations again. We will have to redefine the purpose oi the employing organization and oi Us management as both satisfying the legal owners (such as sltareholdcrs) and satisfying the owners of lite human capital thai gives the organization its weaIth-producing power—that is, satisfying the knowledge
CALIFORNIA ^lANAC^f MENT fifcVtfrW  VOL A I,MO-     VtfNTEP. I9y9 ?S
Knowtedg^Worker ProdurtwityiTht Biggest Chalenge
workers. Increasingly, the ability or organizations—and not only of businesses— to survive will come to depend on their "comparative advantage" in making the knowledge worker more productive. The ability to attract and hold the best of the knowledge workers is the first and most fundamental precondition.
However can this he measured or is it purely an "iniangihte"? Tills will surely he a central problem lor management, lor invcslurs, and for capital markets. What dties "capitalism'* mean when knowledge governs rather than money? And what do "free markets" mean when knowledge workers—and no one else can "own" knowledge—are the true assets? Knowledge workers can neither be bought nor be sold. They do not come with a merger or an acquisition. In fact, although they are the greatest "value," ihey have no "market value"—that means, of course, that they are not an "asset" in any sense of the term.
These questions go far beyond the scope ol this anicie. However, it is curtain thai the emergence as key questions ol the knowledge worker and of the knowledge-worker's productivity will, within a lew decades, bring about fundamental changes in the very structure and nature of the economic system.
Note*
1.   For work in the oldest knowledge profession-—that is, in Medicine-—Taylor's close contemporary William Osier (19) did what "ftwlpr did and m the same time in his 1S92 1* N* Tft? Prinayhs and Practice of Medicine (arguably the best texi-book since Euclid's Gtemttiy in the I bird century B.C-i. Osier's work has rightly been called the application of Scientific Management to Medical Diagnosis, Like Tjjlor, Osier preached that there is no "skill," there is only ntetfted-
<H        CAJFC«USllA^W*V^KNTftfiV|ilW   VJX4ltNQ3  WINTER (995