(FORTUNE Magazine) – THE YEAR was 1963, the occasion was a prestigious series of lectures at Columbia University's business school, and the speaker was arguably America's most famous chief executive. He talked about the difficulty successful corporations face in maintaining strength and might. Of the 25 largest U.S. industrial corporations in 1900, he said, only two remained in that select company. He named them not. But by his description, one appears to have been General Electric and the other a predecessor of U.S. Steel (now USX). The rest had failed, been merged out of existence, or simply fallen in size. ''Figures like these,'' said the executive, ''help to remind us that corporations are expendable and that success -- at best -- is an impermanent achievement which can always slip out of hand.'' The speaker was Thomas J. Watson Jr., chairman of IBM.

IBM was at that point a monarch, about to launch its famed 360 family of computers. Watson could not have dreamed he was foreseeing his own company's future. Neither could he have imagined the fading of two other giants of the day, General Motors and Sears Roebuck. GM was then widely suspected of trying to hold its market share down, so as not to rouse the dogs of antitrust. Sears absolutely ruled retailing; the company was bigger in merchandise sales than its next four rivals combined. Had a ranking been compiled in the 1960s of the large companies most likely to succeed, these three might well have headed the list. Yet the three lost a total of $32.4 billion in 1992, more than four times what the 1990-91 Gulf war cost U.S. taxpayers. In the tough-talking world of big business, moreover, these companies are often called dinosaurs. The word is a stretch: The three are not extinct, only painfully and wheezingly gasping for breath. Nonetheless, in capturing an image of broad and coincident decline among several species of onetime dominant creatures, the word works beautifully. It is truly amazing to see these legends struggling all at once, as if they had simultaneously suffered an industrial accident. Fundamentally, that's what they did, in a very particularized fashion that caused their fortunes to deviate from those of their industries, autos and retailing and computers. These industries have not staggered in distress. All, in varying degrees, have grown and prospered. But their onetime leaders? They just didn't keep up. Additionally, they went lame in remarkably similar ways and in a manner that supplies pointed lessons for all of business. The companies acquired the usual encumbrances of success -- among them arrogance and bureaucracy -- and they devised new ways to fail as well. Or, precisely, their executives did. Companies don't stumble; people do. As Peter Drucker has said, ''Every failure is a failure of a manager.'' On the leading edge of this topic is the question of whether enormous success, such as these three companies once enjoyed, inevitably breeds failure. The normal and ruinous problems of success, so it is contended, are complacency and size, the latter of which adds layers of authority that slow reaction time and generally stand in the way of good management. Over the years these dangers have inspired many a business bromide, among them ''Nothing fails like success'' and ''If it ain't broke, fix it.'' Thomas J. Watson Sr., the patriarch of IBM and a lover of sayings, himself admired ''It is harder to keep a business great than it is to build it.'' By instinct and by the plentiful evidence at hand, Corporate America recognizes much of this conventional wisdom as true. On the other hand, any fool knows there are advantages to size, among them purchasing muscle, staying power, and depth of talent. So logic says that adept management ought to be able, at least on occasion, to cope with success. The table on the preceding page proves that can happen, though it also confirms the impermanence of which Watson the Younger warned. The table lists the 20 largest companies in stock market value, worldwide, at three points in the last 20 years. At the end of 1972, which was an era of sky-high valuations, the Big Eight, American companies all, lined up like this: IBM, AT&T, Eastman Kodak, GM, Exxon, Sears, General Electric, and Xerox. By the end of 1992, five of those -- IBM, Kodak, GM, Sears, and Xerox -- were gone from stardom. Gone not only from the top eight, but from the top 20 as well. Nonetheless, that leaves three survivors, Exxon, AT&T, and GE, which together convey the unmistakable message that success doesn't necessarily kill. From 1972 to 1992, Exxon vaulted from fifth place to first. AT&T, slimmed down by the Bell breakup in 1984, nonetheless stayed in the top seven. And GE shot up to second. A case can be made that this duke of diversity, GE, is the standout survivor among the big guys. Formed in 1892, GE has ridden the rapids with eight chief executives, and in the past decade alone, under changemaster John F. Welch Jr., has gained $52 billion in market value. Jack Welch definitely draws some criticism: Tom Peters, the management consultant turned author (In Search of Excellence and other books), doubts that the strong-minded Welch is doing enough to develop managers capable of standing up to and succeeding him. But a more generally accepted view comes from Noel Tichy, a business professor at the University of Michigan and co-author (with FORTUNE editor Stratford Sherman) of a new book about GE, Control Your Destiny or Someone Else Will. Tichy claims Welch's bureaucracy-busting management kept GE itself from veering into Dinosaurland. In the ruin department, the slouching of Eastman Kodak and Xerox right off the big-value list reminds us that there are more than three species of huge, humbled creatures around. Kodak executives recoil at being lumped with GM, IBM, and Sears, but the company's fall in prestige is comparable -- as its | decline in market value, from $23.9 billion in 1972 to a recent $17 billion, signals. Xerox is a classic case of a company that couldn't cope with prosperity. It rode a rocket of success in the 1960s and then sputtered in the 1970s, done in by arrogance, poor products, and Japanese competition that it did not at first take seriously. Concluding when he took over as CEO in 1982 that Xerox was literally in danger of going out of business -- ''we were becoming a dinosaur that couldn't get out of its own way'' -- David Kearns engineered a rebound in copiers by focusing intensely on quality. Even today, however, Xerox's market value is only about $8 billion, far less than its $11.8 billion in 1972. The most visible hulks in Dinosaurland -- Sears, IBM, and GM -- got their original stature in ways that make them seem brothers under the skin and perhaps even carriers of the same genetic defects. Sears, born in 1886, is the oldest; GM was formed in 1908 and IBM in 1911. All ultimately got to the top by besting important competitors: Sears outdid Montgomery Ward; GM left Ford in the dust; IBM overtook Sperry Rand in computers. Likewise, all were driven to success by strong executives with long tenures. General Robert E. Wood ran Sears from 1928 to 1954. The genius of GM was Alfred P. Sloan Jr. (CEO from 1923 to 1946), and the viziers of IBM -- for 57 years, until 1971 -- were the Thomas J. Watsons, father and son. Perhaps as night follows day, the CEOs who came next were neither as visionary nor as forceful. Some did not even perceive a need to be. All had inherited strong, dominant companies that seemed competitively secure. Each company had built a moat that guarded its castle: Sears had the best store sites, many in flourishing malls; GM's dealers were the class of the industry; IBM had installed so much equipment in the shops of big customers that they were virtually forced to go for the compatibility of other IBM machines when they next stepped up to buy. Above all, GM and Sears thought themselves protected from competitive harm by economies of scale. GM's came from skills in mass production, Sears' from purchasing might. In time IBM also began talking economies of scale, gained by billions of dollars of capital investments it made in the late 1970s and early 1980s. In the end, none of this mattered. What swept over these companies was profound change in their markets, to which they were required to adapt. None did, neither fast enough nor fully enough, in part because the erosion of ) their positions was so gradual as to leave them unaware that they were descending into a state of crisis. They could have profitably brooded over an opinion stated recently by Andrew S. Grove, chairman of Intel: ''There is at least one point in the history of any company when you have to change dramatically to rise to the next performance level. Miss the moment, and you start to decline.'' At bottom, these three troubled companies were constricted by their cultures. In a scolding internal memo written in 1988 to GM's brass, Elmer W. Johnson, then executive vice president of the company and before and now a partner at the Chicago law firm Kirkland & Ellis, called for the culture to change so that GM could rise above its maddening ''inability to execute'' -- but he doubted that radical improvement was possible. NOR COULD the culture be surmounted at Sears, says a former senior executive: ''I've talked to a friend who worked at IBM, and we agree that when a company gets to the top, the processes of how decisions are made become all- consuming. You start focusing on how decisions are made rather than what you decide.'' At Sears, he recounts, the problem was compounded by a whole library of ''bulletins'' that spell out procedures for dealing with almost any problem. Says he: ''God forbid there should be a problem that comes up for which there isn't a bulletin. That means the problem's new!'' General Electric's Welch has defined management as ''looking reality straight in the eye and then acting upon it with as much speed as you can.'' The three big companies that GE has outrun never seemed to take off the blinders where reality was concerned. Neither did these companies fix their gaze on the customer. Rather, they looked both inward and backward, remembering how it used to be and somehow expecting those glories to reappear. As a result, they were always behind the curve. At Sears, the challenges were intensified by the way in which General Wood passed the baton of management. Not wanting to retire at age 65, as the rules called for, Wood persuaded his board to give him extensions that lasted more than nine years, until 1954. At that point there was lined up at Sears a string of top executives deemed by Wood and the board to deserve their moments in the sun before they themselves reached retirement age. The next five CEOs served an average of 3 3/4 years each, which is hardly time enough to figure out where the sun's coming from. Donald R. Katz, author of a 1987 book about Sears, The Big Store, describes this roster of CEOs as having a ''caretaker'' mentality, dedicated to maintaining the majesty of Sears, which they naturally thought unassailable anyway. It is always a mistake to think this in retailing, whose customers are notoriously fickle. Before long, Sears' anchor stores in malls were getting tough competition from the specialty shops that fed off the Sears traffic and sluiced away every dollar of business they could. But the critical change for Sears was the surge, in the 1960s, of the discount stores, especially Kmart, later followed by Wal-Mart. In general, the caretaking crew at Sears (reacting just like Xerox's managers) did not take this new competition seriously. Without exactly articulating the theme, Sears thought of itself as America's discount store, and it was inclined to look down on Kmart's customers as an inferior, unmonied lot. Even when Sears' growth slowed in the early 1970s and Kmart kept clipping along, Sears did not fully realize there was a kind of sea change going on in customer preferences, against which it had no workable strategy. It did, though, have a new place in Chicago to not worry about such things: the Sears Tower, an opulent monument to greatness, completed in 1973.

Later Sears scrambled into financial services, adding Dean Witter and Coldwell Banker to its existing Allstate insurance business. In retailing, it rolled out one strategy after another. But try as it might, Sears could not get back its merchandising magic. A certain reason for this failure is that the company always stayed ''encased in time'' -- to use the description of a consultant who once studied its workings -- and never truly caught up with all that was going on. For example, well into the 1980s Sears was still producing position papers that, though replete with talk of competitors, included no mention of Wal-Mart. In such respects, Sears seems the corporate equivalent of a laboratory ''boiled frog'' -- the hapless creature that sits calmly and unsuspiciously in water whose temperature is gradually raised to the killing point. From his bare-bones headquarters in Bentonville, Arkansas, 500 miles and a world away from the tower, Sam Walton watched the boiling exercise disdainfully. He was an admirer and student of Kmart but no respecter of Sears. In his autobiography, Sam Walton: Made in America, written with FORTUNE editor John Huey, Walton summed up Sears' gravest miscalculation. ''One reason Sears fell so far off the pace is that they wouldn't admit for the longest time that Wal-Mart and Kmart were their real competition. They ignored both of us, and we both blew right by them.'' True. In 1992, Sears in its entirety -- including financial services and all manner of special charges -- had a loss of $3.9 billion, of which $3 billion came from the merchandising group. Sales for that group were $32 billion. Kmart had $38 billion in sales and $940 million in profits. And Wal-Mart was winging out of sight: $55 billion in sales, on which it earned just under $2 billion. The face of denial that Sears turned toward the discounters was, of course, the identical countenance that GM presented when Japan's automakers began to assault the U.S. in the 1960s. Seduced by the bountiful profits to be made on large cars, GM's executives could neither bring themselves to believe that Americans would buy small cars nor convince themselves that ''Made in Japan'' manufacturers could make salable cars of any size. In these opinions GM was joined by Ford and Chrysler. But the ability of this pair to duck reality was aborted by financial crises that both faced as the 1970s ended. Shocked into action, the two undertook rush programs that brought them back to competitive shape and took them in the direction of Japanese-style ''lean production,'' the antithesis of mass production. GM, historically rich, stayed wealthy enough to keep throwing money away. Some went to director Ross Perot, who hassled GM unmercifully about its committee-driven bureaucracy and the insularity of its executives. GM paid him $750 million to vamoose and shut up. That was small change compared with GM's spending on car models that seldom sizzled in the market. In the 1980s, singing one more chorus about economies of scale, the company spent a huge $7 billion on the so-called GM-10 program, a sweeping remake of its midsize cars and the largest new-model package ever. But the cars had to be redesigned and their manufacture replanned so that savings could be made; the bureaucracy grew one more notch because of a reorganization that added staff; and as the decade wore on, cash tightened. Ultimately GM's cars arrived late in a market that had rudely developed new tastes. Asked by FORTUNE to explain what went wrong, Roger Smith, GM's boss from 1981 to 1990, said, ''I don't know. It's a mysterious thing.'' In perhaps its largest gesture of big spending, GM capitulated repeatedly to the demands of the United Auto Workers, thereby burdening itself with enormous wage and benefit costs. One kind of cost -- health benefits for retirees that were to be paid in the future -- was for many years virtually invisible because accounting rules did not require that it be recorded as a current expense. The rules allowed all companies to bury their heads in the sand about a cost that was spiraling out of control. But the reprieve especially fit the temper of GM executives, who over the years have worked hard at presenting their financial results in the most favorable light possible. In her book about GM, Rude Awakening, security analyst Maryann Keller says that one way to get ahead at the company was to display a talent for sprucing up the numbers. Last year, however, new accounting rules brought the retiree health benefit problem home to roost, and GM took a monumental charge for the cost. Pretax it was $33 billion; aftertax, $21 billion. Do not be lulled by the reassuring statements of GM (or of other companies) that the retiree charge ''does not affect cash flow.'' The charge doesn't change cash flow, true. But it publicly exposes the size of the future cash flow burden -- this monumental liability of $21 billion -- for the first time, starkly revealing just how profligate GM was in its promises. For 1992, a year hit by other charges as well, GM reported a $23.5 billion loss, far and away the largest ocean of red ink that has ever engulfed a FORTUNE 500 company. As it began 1993, the company was down to stockholders' equity of $6.2 billion, vs. a peak of nearly $36 billion in 1988. In its share of the U.S. car market, GM was also down, to 34.9% in 1992, a planet away from the shares of 50% or higher that it sometimes registered in the 1960s. In those days imports had under 10% of the market. The milestone of 1992 was that foreign companies, counting the cars they produced in the U.S. and outside it, had precisely the same share as GM, 34.9%. IBM HAS always seemed younger in spirit than GM and Sears. Yet it built a bureaucracy that would rival either's. Trying to get action out of IBM, said one of its business partners a few years ago, is like swimming through ''giant pools of peanut butter.'' One theory of how the bureaucracy grew refers back to the System/360 family of computers, introduced in 1964. Audaciously ambitious and complex, the system included six models that were launched simultaneously, that incorporated new and radically different software suitable for large and small machines alike, and that required five new factories. Writing about the gamble in progress, FORTUNE put a price on the 360 that in today's dollars would be about $20 billion and called it ''the most crucial and portentous -- as well as perhaps the riskiest -- business judgment of recent time.'' After some horrendous growing pains, the 360 became a market sensation, cementing IBM's industry leadership and raising the Big Blue flag at hundreds of corporations. But one retired IBM-er, Malcolm Robinson, who rose to a senior post in IBM Europe and who earlier had worked as an assistant to senior vice president (and future CEO) Frank T. Cary, thinks the 360 may have been another kind of turning point as well: ''Triumph though it was, the scale of the project created a complexity in the business that almost couldn't be handled. It was chaos for a while. So an organization had to be created to bring things under control and make sure that kind of breakdown never happened again. And that really may have been what made the bureaucracy take off.'' IBM's figures support his theory: Between 1963 and 1966, while sales in constant dollars were rising about 97%, IBM's head count went up almost 130%, growing from 87,000 to 198,000. In the next 20 years the number of employees climbed, sedately but surely, to a peak in 1986 of 407,000. Now, mainly because crowds of employees have accepted retirement packages, the count is down around 300,000 and falling. After the 360's problems had been subdued, IBM confronted another personality-altering crisis. The trouble was antitrust suits, brought by Control Data and the Department of Justice. Toward the start of these, in 1970, Tom Watson, then 56, suffered a heart attack, and by 1971 he had decided to retire. He was succeeded first by Vincent Learson (who settled the Control Data suit) and then, in 1973, by Frank Cary. CEO for eight years, Cary was by nature a careful, stolid, singles-hitting kind of executive, whom the antitrust litigation made even more so. His strategy of moving cautiously worked in court: The government gave up in 1981 without having forced the breakup of IBM that it wanted. But the whole litigation atmosphere, says the chairman of a big IBM competitor, ''gave the lawyers too much power and made the company risk-averse.'' You can't be that, Sears' antagonist Harry Cunningham of Kmart once said, as he quoted Sir Hugh Walpole: ''Don't play for safety. It's the most dangerous game in the world.'' For IBM, trying to steer the safe course in those days meant, for one thing, keeping the Japanese computer companies in check, which it did skillfully. Above all, the company thought safety lay in the preservation of its prodigious mainframe profits (which were its version of GM's big-car bundles). Unhappily for this approach, a cascade of miniaturization miracles has proved mainframes to be incipient dinosaurs themselves and their profits not defensible. The years have also proved three steps that IBM did not take to be disastrous. The first was the failure to exploit the technology called RISC, for reduced instruction-set computing. IBM invented the RISC microprocessor in 1974 and knew its promise: simplified and faster computing, particularly well suited to the minicomputers then gaining fans. But RISC was also a rival to IBM's existing mainframe technology. So what do you do with this newcomer? The question was effectively answered by IBM's organizational structure, which had once included a separate sales force for small and midsize computers but by the mid-1970s had been reorganized into a monolith, with only one sales division. That meant there was no significant internal competition -- no hungry champion of midsize machines who might have said, ''Hey, let me take RISC and run with it.'' The technology therefore mainly sat on the shelf, as IBM lollygagged in deciding what RISC products to make or even whether to make them. Behind the delay -- at least partly, if not solely -- was management's knowledge that the smaller, cheaper machines that RISC promised might steal business from mainframes. Unfortunately, the competition is never restrained by such thoughts. Other companies raced in where IBM hesitated to tread, leaving it a late starter and, so far, an also-ran in the burgeoning RISC field. A former senior executive of IBM, who was himself a leader of the lollygag team, said recently that RISC ''will sort of take over the world.'' In his opinion, IBM's not being there first with the most may turn out to be its sorriest mistake. Another contender for that distinction will surely be the company's failure to grab a full plate of PC profits, a flub that occurred soon after Cary passed the reins to John Opel in 1981. In its mainframe line, IBM makes the whole package -- the hardware, the software, and the semiconductors -- and therefore controls all the sources of profits. But when the company finally got ready to bring out a serious personal computer, IBM signed up Microsoft to supply the software and Intel to supply the chips. A FEASIBLE alternative for IBM might have been to buy all or part of both companies and to have captured the profits that are now -- you will pardon the expressions -- flooding their gates (first name: Bill) and greening their grove (Andy). In the early 1980s, IBM did buy 20% of Intel for $640 million, seeking to help the chipmaker through a rocky period. But a few years later, IBM sold its stake, cheerily pocketing a gain and apparently never thinking about what larger profits it might be relinquishing. The shares IBM held would today be worth $3.6 billion. To be sure, IBM's decision to bestow business on Microsoft and Intel was good for the PC market, in that software developers were encouraged by this rare display of openness on Big Blue's part to whip up all manner of applications for the machines. Had IBM tried instead to control all three sources of profits, the developers might have been scared off by their distrust and even hatred of IBM. The PC market might then not have bloomed for everybody as it did. But that is an uncertainty arrayed against several indisputable facts. Last year, after taking $11.6 billion in charges for employee departures and plant closings, IBM recorded a $5 billion loss. Microsoft made $700 million and Intel $1.1 billion. The combined market value of the pair was recently $48 billion vs. IBM's $30 billion. The third step that IBM did not take reaches to the core of its culture: In the face of the dismaying results that began in 1985, just after John Akers became CEO, and that persisted, IBM failed to accept the reality that its so- called full-employment practice, in which it forswore layoffs, was no longer workable. A retired IBM manager who worked closely for years with IBM's top executives recalls the mystique that grew up around this practice: ''It was a religion. Every personnel director who came in lived and died on defending that practice. I tell you, this was like virginity.'' Just recently, a day late and a dollar short, IBM at last gave it up. There should also be mentioned the possibility of a fourth IBM mistake, which can best be explained in terms of AT&T and its 1984 breakup. The world has since seen the energy unlocked by that event. As the table presented earlier shows, AT&T -- the surviving company -- had a market value at year-end 1992 of close to $70 billion. But if the other seven Bell companies were to be tacked on, the value of the package would have been $208 billion, a figure that, first, is more than seven times AT&T's value in 1972 and, second, would dwarf all other figures on the 1992 list. The point is that perhaps IBM's management and board should have acceded in the 1970s to the Department of Justice's call for a breakup. That outcome, for example, might have spurred the company -- or a splinter of it -- to make the most of such marketing opportunities as RISC instead of frittering chances away. In any case, that outcome almost happened. Discouraged at one point in the litigation, and seeking a settlement, IBM actually offered to split itself into two pieces -- big computers and small. But Justice wanted seven pieces, and that struck IBM as unbearable, so no deal was made. And thus is history written. For these three humongous, hurting companies, these past 12 months have been the year that was. Last August, Sears Chairman Edward A. Brennan finally gave in to pressure to step down as head of merchandising and brought in an outsider, Saks Fifth Avenue executive Arthur C. Martinez, to take the post. In October, GM Chairman Robert C. Stempel resigned under pressure, after which the board elected outside director John G. Smale of Procter & Gamble as chairman and put a young new management team, headed by President John F. ''Jack'' Smith Jr., 55, in charge of operations. In January, IBM's Akers announced he would resign, and the IBM board launched an urgent search for a new CEO, eventually selecting an outsider, RJR Chairman Louis V. Gerstner Jr. He may well continue Akers's work in splitting up the company, possibly into more than seven pieces. So the cultures of these companies are under attack in a way that they have never been before. But because the job of resuscitation is so difficult, no guarantees of success can be issued. New bosses Martinez, Smith, and Gerstner certainly aren't cinches to do well, as each surely knows. Even raising the question of whether their companies can ultimately ''make it'' requires a definition as to what that means. Does making it connote survival in any form? If so, the answer's easy: The names Sears, GM, and IBM are likely to be around, and have substance behind them, for a long time. Or does making it mean a return to the world of superior companies, which may in ; turn be defined as those demonstrating a consistent ability to produce upscale returns on equity -- as earned, say, by the top third of all companies. That is a tall order for this bunch, considering the ground they have lost and the turmoil they still surely must face. The most one can say for now is that a perceptible shift has occurred. These companies have finally stepped up and done a reality check, and that is at the very least a start. In the meantime, companies or even operating divisions still cherishing success and leadership might consider a few warnings that arise from this tale of the once invincible: Don't be overimpressed by your market position; your moats; your wealth; or yourselves.

CHART: NOT AVAILABLE CREDIT: FORTUNE CHART/SOURCE: R.L. POLK & CO. CAPTION: MARKET SHARE GM Non-U.S. automakers

CHART: NOT AVAILABLE CREDIT: FORTUNE CHART/SOURCE: WILSHIRE ASSOCIATES CAPTION: MARKET VALUE IBM Microsoft