(FORTUNE Magazine) – There's a reason your company doesn't give inkblot tests to job applicants--though 50 years ago a great many companies on the first FORTUNE 500 did so, and with enthusiasm. It's the same reason your company isn't trying to convince investors that millions of people merely looking at its website creates $1 billion of new shareholder value, though plenty of companies were attempting just that in the late '90s. The common reason: They were stupid ideas.

You might suppose, if you'd never been anywhere near an actual business, that managers aren't susceptible to idiotic fads. Surely businesspeople are too disciplined, too unwavering in the eternal pursuit of shareholder value. But obviously that's nonsense. Managers are mere humans, and what's most eternal in a big company is the incredible difficulty of achieving consistent profit and growth. When a plausible idea for achieving them shows up, you can't blame managers for swarming all over it. T-groups? Conglomeration? Quality circles? Hey, what if it works, and my competitors use it--and I don't? That's how fads happen.

The managers of the FORTUNE 500 have pursued all manner of fashions over the past 50 years, and they weren't all stupid. Some were smart. Some have come and gone and come back. And the tinkering will never stop, because managers will never, ever get this management thing completely figured out, though they'll always suspect that somebody they've just heard about might finally have the answer.

Those inkblot tests were a small emblem of the dominant management idea back when the first 500 appeared: that science, mathematics, and sheer brainpower could conquer the impossibility of running a big organization. Turning the dials, their spectacles glinting in the fluorescent light, was a fearsome new class of men--professional general managers--that had simply never existed before. The people who had run companies in the past were entrepreneurs or nephews of the founder or lifelong "coal men" or "steel men" or "beef men." But as members of a new professional class, the managers of the '50s seemed beyond any of that. Not beholden to any particular industry, they were, metaphorically, lab-coated technicians of profit.

Some of their tools were so powerful that they have endured to this day: linear programming, statistical theory, precise cost accounting. Others, like inkblot tests, psychodramas, and hypnosis-based consumer research, faded away because they didn't work. But no one knew that at the time. These managers were business Olympians with a fresh quiver of thunderbolts, or so they thought.

The era's mindset shows up clearly in one of the most popular business concepts of the time, "planning and control for profits," or P&C for short. There you had it: Managers planned and controlled pretty much everything. They believed they could make people buy with new tools like "motivation research," as peddled by the then-famous Dr. Ernest Dichter, an Austrian psychologist who said things like, "A sedan is a wife, but a convertible is a mistress." They could plan revenues with newly developed market projections--50 years out! They could control costs. They could make the profits happen.

That was not a system that celebrated individual managers. What it celebrated was the system itself, the organization, the company. Celebrity CEOs disappeared. Back in the '20s and '30s everyone knew--and even today many people still know--that Owen D. Young and Charles Coffin ran General Electric, George Washington Hill ran American Tobacco, Alfred P. Sloan ran General Motors. Who ran any of those outfits in the '50s? In a coincidence that symbolized the era's conformity, GE and GM were for a time run simultaneously by CEOs named Charles E. Wilson. When GE's Wilson was succeeded by Ralph Cordiner, he established a seven-man "office of the president," a new term. Message: The group, not the individual, surmounts all.

As always in human affairs, the sense of omnipotence was false. You can plan and control only so much. The Santa Fe Railroad bought hundreds of acres around Los Angeles based on its expected needs in 50 years, never imagining that railroads would be diminished by trucks and planes (though the land proved a good investment). Sears built up its presence in the South based on 50-year forecasts, never suspecting that within 35 years a Southern competitor, Wal-Mart, would seize Sears' seemingly God-given status as America's largest retailer.

Managers' faith in P&C and in the organization's superior wisdom faded not because those notions failed spectacularly--they petered out rather than blowing up--but because society changed and new fashions arose. As the '60s progressed, technology became cool. This was technology we still recognize as technology: computers, chips, space exploration, brain chemistry, satellite TV. Much of it derived from the epochal invention of the transistor, arguably the greatest spawner of management fads over the past 50 years. Anything electronic was cool. IBM was ultracool. It was a time when, as one man told FORTUNE, "a pig farmer could rename his farm Hogotronics Inc. and have a public offering." Tech coolness, rising and falling like skirt lengths, would become a constant of managerial style forever after.

Growing at the same time was a truly massive managerial concept: conglomeration. Few fads have matched this one for its ability to make managers drunk with visions of glory--and in the '70s, give them throbbing morning-after headaches. The idea was an extension of the '50s corporate masters-of-the-universe theme, this time executed not through cerebral group analysis but through bold cowboy takeovers. Our company will get into the very best businesses, no matter what they are, right now! A conglomerate called ATO made bomb fins, vacuum cleaners, and baseball gloves, among other things. ITT rented out Hertz cars, ran the Chilean phone system, and made Wonder Bread. Wall Street loved it. The CEO of a conglomerate called Bangor Punta predicted that one day America would have only 200 corporations, all of them conglomerates.

For much of the '60s, no one in American business was more of the moment than a conglomerateur named James J. Ling, a brilliant high school dropout from Hugo, Okla. His company, LTV (for Ling-Temco-Vought), was in not just electronics but also aerospace, another superfashionable business back when "Space Age" was still capitalized. Through relentless acquisitions he built built LTV into a FORTUNE 500 star (it peaked at No. 14 in 1970). How did he pull it off? Most people couldn't exactly say. He was always selling convertible debentures or merging one subsidiary into another or spinning off something to the shareholders or issuing a new class of stock--but however he did it, lots of managers wanted to do it too. LTV shares went from virtually nothing to $150.

The fact that you've never heard of James Ling if you're under age 50 is your tipoff that something went wrong. Ling's trajectory can be plotted by the headlines of FORTUNE's articles about him: "Jimmy Ling's Wonderful Growth Machine," then "Some Candid Answers From James J. Ling," and after that "LTV's Flight From Bankruptcy."

Ling's trouble was common to many of the conglomerateurs: He was addicted to doing deals, bored to tears by the tedium of managing a business. As long as acquisitions kept up the growth, Wall Street was happy. But if you're forced to keep doing deals, you'll eventually do some pretty scaly ones. Ling ended up buying awful businesses, such as Jones & Laughlin Steel and Braniff Airlines. Decent targets got harder to find, and the machine stopped spinning. Profits fell. Wall Street snapped out of it. The chart of LTV's stock price from 1967 to 1972 looks like that of any dot-com from 1998 to 2001.

The demise of the conglomerate fad wasn't just the result of bad acquisitions. Managers learned something important about investor behavior in those years. Much of the rationale for conglomerates was that a carefully chosen portfolio of disparate businesses would produce what investors love most: steady, predictable earnings. In a well-run conglomerate, when one business was down, another would be up. Total profits would rise majestically, and investors would pay a premium for the stock.

The reality was exactly the opposite. Rather than buy Jimmy Ling's portfolio of businesses (or Roy Ash's at Litton or Harold Geneen's at ITT or G. William Miller's at Textron), investors preferred to assemble their own portfolios by buying stocks of one-industry companies. Once they got over their conglomerate infatuation and thought it all through, they decided the only way they'd buy a preassembled portfolio--i.e., a conglomerate--was not at a premium but at a discount. This lesson was so painful that, unlike most, it has been largely remembered: In the greatest merger wave since then--in the late '90s--most takeovers were within industries.

The fall of the conglomerates in the early '70s was a suitable start to a decade of manifold miseries in which even the managerial fads were glum. Inflation got so bad that today, just as economists refer to the '30s as the Great Depression, they call the '70s the Great Inflation. The price bubble sparked a managerial fad for borrowing, on the theory that the dollars you paid back would be worth less than the ones you borrowed. But that worked only if your inflation forecast was better than your lender's, since of course he budgeted for inflation in setting his interest rate. So it was pretty much impossible to make that tactic pay. Then there was the Arab oil embargo, which got managers briefly jacked about alternative energy sources. But there really weren't any that made economic sense, and that was another bust.

The worst bummer of all for businesspeople was that the stock market went nowhere. For perspective on just how nowhere it went, consider this: We know in retrospect that the great bull market began in August 1982, with the Dow at 777. Yet it had been higher way back in 1964, when it closed the year at 875. In those intervening 18 years it broke 1000 a few times, but it also descended into the 500s after Nixon resigned. The ups never lasted, and the downs never stayed away for long. Executives felt cursed. The market, conventional wisdom concluded, was "broken." What other explanation could there be? We're brilliant executives, are we not? And yet for more than a decade the stock market has refused to acknowledge our achievements!

Thus began a managerial fashion that lives on: paying executives munificently despite lousy market performance. The reasoning: If the boneheaded market can't see how terrifically these superhumans are performing, that's not their fault. They have to be paid for their efforts. At some point in the late '70s--compensation experts disagree on the precise coordinates--an American CEO was paid more than $1 million. Every other CEO noticed. The dam was breached. The flood followed and has yet to subside.

More '70s grimness, depending on your industry, came from Japan, which was taking up increasing managerial brainspace. In industries like cars and TVs, Japan was evil. But in industries less immediately threatened, Japan actually started to become cool. Japanese companies began building U.S. factories, exposing Americans to the weird and much-parodied system of Japanese management, with uniforms, morning exercises, and hokey company songs.

But the laughter was nervous, for the cruel truth was that by 1980 American management hadn't had a good idea in years. With the Japanese and German economies fully recovered from World War II, some of their companies began drubbing U.S. competitors, already demoralized by the bad-news '70s. That was the setting for one of the great managerial fashion statements of all time: the famous NBC documentary called If Japan Can Do It, Why Can't We?

In retrospect the whole thing seems absurd. Full-hour documentaries are now almost extinct on network TV, let alone programs on business, and even then statistical process control was not a promising topic. Such a TV show was going to galvanize the world's largest economy? Yet that's what happened. At a stroke, U.S. business learned about quality, Japanese-style, and embraced it with a panting ardor rarely seen before or since.

Quality became the inescapable management topic of the '80s, and on the whole it was a good thing. Our cars were awful. Appliances were awful. Service was awful. Six Sigma statistical processes really could fix them, and did. Yes, some people got carried away. Managers started calling meetings for 9:51 and 2:19 so that everyone would get used to showing up precisely on time. At big quality conferences--there were thousands of them--someone would ring a bell when the speaker's time was up, even if he or she was in mid-sentence. But the big idea, which endures, is that quality isn't an expensive add-on. On the contrary, quality pays.

Idea-starved managers glommed onto plenty more concepts in the early '80s. They were so hungry they created an entirely new phenomenon in publishing, the business bestseller. In Search of Excellence popularized "management by walking around." The One Minute Manager is still on business bestseller lists after more than 20 years. Lee Iacocca's book was the first business autobiography to sell millions.

Among an exclusive subset of American managers, U.S. industry's general lameness provoked a sharply different response: Buy underperforming companies, throw out the jerks who'd been mismanaging them, and install executives who would do what needed doing. Hostile takeovers were old stuff, but in the '80s they returned in mammoth dimensions, nourished by prodigious debt after Fed chairman Paul Volcker got interest rates under control. As a management fad they were never all that widespread, but they didn't have to be. Just as executing one mutineer can sober up a whole boatload of sailors, a few hostile takeovers left vast swaths of the American CEO class in mortal dread of their secretary's saying, "Mr. Pickens on line one."

Conventional CEO wisdom thus held that Boone Pickens, Henry Kravis, Ted Forstmann, and other raiders were evil men. But they weren't. They shaped up a lot of companies that desperately needed it. They established the new idea of a continuous market for corporate control, an idea that rightly persists.

By convenient coincidence, the '80s as a concept ended almost exactly with the '80s by the calendar. Within mere months of decade's end the U.S. went into recession, Michael Milken went to prison, the Japanese stock market crashed, and Perrier, the decade's official drink, was found to contain benzene.

The decks were cleared for the '90s, and in all the annals of virulent managerial obsessions few decades can match it. Things started out quiet. The must-have concept of the early years was reengineering, a sensible idea that somehow got snakebit and became despised as a code word for firing loads of people. It lives on, as it should, only now it's so mainstream it doesn't need its own term anymore.

Even in those early years of the decade, Silicon Valley's mystique was building. It had been around awhile, of course, its appeal confined mainly to geekdom. Now the fascination was broadening. What gripped Joe and Jane Manager was not so much the wonder of the computer as the Valley way of doing business--the venture capitalists, the mutating business models. The whole scene was a mixture of youth, high IQs, confidence, and technical miracles, and somehow it was churning out a growing chunk of the U.S. economy. It was becoming way cool.

Or so everyone thought. It's now clear that until Aug. 9, 1995, no one actually had the remotest idea what cool could be. That was the day of Netscape's IPO, the event that ignited the historic frenzy. It possessed the three elements that would prove such a powerful combination in the many, larger IPOs that followed: the Internet angle, of course; the child genius founder, in this case Marc Andreessen (aided by an experienced manager, Jim Clark); and the fact that the company really shouldn't have been going public at all. It was too young. By all previous standards Netscape was still a late-stage venture capital situation, not an IPO. But Clark needed to make a payment on the yacht he was having built, so they took the company public anyway. Investors mobbed it, creating the precedent for similarly immature startups--that is, most of the dot-coms that went public in the late '90s--to be financed no longer by tough-minded Sand Hill Road financiers but by teachers, dentists, and pipefitters from Lubbock to Bangor.

It was Hogotronics all over again, only this time it was Hogs.com. And while the Net caused one of the great investment manias of all time, it also begat a managerial mania. After all, the era's conventional wisdom came down to a single sentence: "The Net changes everything." At boring old companies burdened with physical assets, every department budget and every presentation to Wall Street analysts suddenly included a PowerPoint slide headed "Internet Strategy." Zapata, a fish-meal and sausage-casings outfit that had started life in the '50s as George H.W. Bush's oil-drilling company, announced one morning in 1998 a vague plan to become an "Internet consolidator" under the name Zap.com. By 4 P.M. the stock had doubled. Across the land managers were insisting that their company would spin money and crush competitors in all kinds of futuristic, Net-based ways. Like what, for example? Like, we don't know, dude, but just talking about it adds a billion dollars to the market cap!

The unwinding of it all is too familiar to describe here, but the managerial effect is worth noting. At a certain point in 2001, as if someone had flipped a switch, the phrase "back to basics" seemed to replace all other management philosophies. In press releases, at board meetings, at industry conferences, managers were falling over themselves making crystal clear that they were now totally devoted to basics.

The most amazing part is that no one was sheepish about it. You might think a declaration of "back to basics" included an implicit confession of having gone astray, but no. It was as if floodlights had suddenly been trained on the dimly lit room where the Internet orgy was going on, and the participants, instead of stammeringly trying to explain themselves, simply put on dark suits and announced they were going to church for some old-time religion. That was then, this is now. Adapting is a virtue, right?

And there we see the power of any big managerial idea. It may be smart, like quality, or stupid, like conglomeration. Either way, if everybody's doing it, the pressure to do it too is immense. If it turns out to be smart, great. If it turns out to be stupid, well, you were in good company and most likely ended up no worse off than your competitors. Your company's board consists mostly of CEOs who were probably doing it at their companies. How mad can they get?

The true value of conventional management wisdom is not that it's wise or dumb, but that it's conventional. It makes one of the hardest jobs in the world, managing an organization, a little easier. By following it, managers everywhere see a way to drag their sorry behinds through another quarter without getting fired. And isn't that, really, what it's all about?

FEEDBACK gcolvin@fortunemail.com

FORTUNE.COM Go to fortune.com for related stories, including why CEO pay reform is taking so long, and the full FORTUNE 1,000 list.