(FORTUNE Magazine) – If you call the main number of Bethlehem Steel, and cool your mind while the polite, automated female voice on the other end tells you what to do if you know the four-digit extension of the person you want, and how to key in a name if you don't, and how to reach billing and accounts payable, you will finally hear this chilling statement: "We have no one to answer questions."

That is what Bethlehem Steel has come to--this builder of 1,127 ships during World War II and of the Golden Gate Bridge, this supplier of metal for every bridge and tunnel that takes you from New Jersey into Manhattan, and for much of the skyline you see when you get there, this munitions maker once called the Krupp of America, this mighty symbol of industrial power through great sweeps of the 20th century. This pillar, in fact, of the FORTUNE 500 when the list was begun: Bethlehem was then very near its top, peaking at No. 8 in 1955.

And then it waned, dropping inexorably in rank, as U.S. industry in general grew and Bethlehem stalled or only sputtered upward. Later, disaster arrived: 15 years of losses, the first wiping out 1977, the others papering long stretches in the 1980s and 1990s. Even then, the losses understated reality, because for years the company's books--in line with practices throughout America--lowballed the costs of pensions and almost totally ignored bountiful health-care promises that the company was making to retirees.

Bethlehem meanwhile kept falling on the 500 list, descending in 2002 to 440. But it was by then in bankruptcy, ignominious, once-unimaginable bankruptcy. Next the finale: Last year Bethlehem's assets--mainly six plants to be kept running--were bought by Wilbur Ross's International Steel Group. With that, the steel artist formerly known as Bethlehem disappeared into the land of no one to answer questions.

Had Bethlehem lived, it would have turned 100 this year. People die of old age; companies normally don't. So this famous corporate death calls for an autopsy--a close inspection of the afflictions that laid Bethlehem out on a steel slab. Was the demise of this once proud and arrogant company inevitable, the unhappy result of its being in the wrong industry at the wrong time, amid unremitting pressures on prices and unrelenting increases in costs? Or could expert and aggressive management, not wedded to history, have seen the handwriting on the wall and saved this company?

In this exercise of complete hindsight, one piece of autopsy evidence is a crowded mortuary: More than 30 U.S. steel companies have gone bankrupt in the past few years. But that limp lineup can't exactly explain away Bethlehem, because over the years it wasn't just any steel company. It was Little Steel, a close second to Big Steel--variously known as U.S. Steel (No. 3 on the first 500 list), USX, and to its competitors, always, "the Corporation."

Mid-century, the two big rivals operated in a world where steel was king and where U.S. Presidents tracked its every move. "As steel goes, so goes the nation," people said. On Wall Street, investors even consulted an index of blast-furnace activity to get a line on what the stock market might do.

Oh, well, that was another era. But if we are chronicling who died, Big Steel didn't. It had its own diseases, including eight years of losses (one of them in 2003). But it diversified big in the early 1980s by buying Marathon Oil, closed plants with relative judiciousness, ran its finances conservatively--and somehow survived. Today, separated from Marathon and back to its core business, U.S. Steel is much smaller but still No. 209 on the 500 list.

Furthermore, if proof is needed that a steel company could positively thrive in the 50 years of the 500, there's Nucor to examine. In 1954 it was a nonentity called Reo Motors, soon to be renamed Nuclear Corp. of America and then Nucor. The company, which was never unionized, swung into making steel in electric furnaces--"mini-mills" became the tag--and began to prosper. This year Nucor is No. 297 on the 500, and a moneymaker, as it has been all of its 25 years on the list.

Autopsies of corporations may be harder to do than autopsies of people, because typically there is a conspiracy of causes, rather than a single malady, that fells a company. Bethlehem fits that mold. Its past 50 years were harshly invaded by competition from foreign steel and the mini-mills; by substitute materials, such as aluminum and prestressed concrete; and by onerous environmental costs.

But those are the usual suspects, and they won't get much attention here. The truth is that Bethlehem did itself in. It was not fit enough to deal with the new competition nor perceptive enough to sense its need for radical treatment.

It could not have happened to a more distinctive company. In its history, Bethlehem had 11 CEOs: Charles Michael Schwab, who founded the company in 1904 (and who is not related to the stockbroker of the same first and last names); Eugene Gifford Grace; and nine who followed Grace. Charlie Schwab, a charming bon vivant and risk taker, was a legend in the steel business, and Grace became one as well. Tall, autocratic, yet hard working, Grace took over steel operations for Schwab in 1913 and, amazingly, was still running Bethlehem in the mid-1950s, when FORTUNE began the 500. It was Grace who orchestrated Bethlehem's wartime heroics. Entering 1943, he promised President Roosevelt a ship a day--and then actually beat that mark by 15 ships.

Grace's company was the heart, soul, and grime of the small city of Bethlehem, Pa. (population today: about 72,000), from which it never moved. In this epitome of a company town, whole families often worked for "the Steel," as it was called. Nepotism was common among Bethlehem's legions of white-collar folks as well.

Bethlehem was also famously into golf, habitually building courses near all its plants, even in Liberia when it undertook a joint venture there. Grace was a champion at golf, and so were many other Bethlehem executives. That may not have been accident: A 1941 FORTUNE story about Bethlehem made the point that in recruiting college graduates it wanted not only brains but also all-around excellence, including "a good physique."

The new hires were then sent through Bethlehem's celebrated "loop course" (named for a round, or "loop," of golf), in which they were exposed to each part of the business before taking on a permanent assignment. Alfred T. "Moose" Copeland, an ebullient 78-year-old New Yorker who sold fabricated steel for Bethlehem, remembers the thrill of getting into its loop course when he graduated from Princeton in 1949: "There wasn't any job I could have acquired," he says, "that would have had more prestige."

The company even polished its image by employing attractive young women to lead plant tours. These so-called escorts also did their part each workday morning by waiting for a lookout on the plant roof to signal that Grace's motorcade was approaching. An escort then made sure that the boss had an elevator to himself as he rode imperially to the executive floor.

It is easy to mock the business style of that day and easy to say that it was abysmally matched to the intense problems about to bear down on Bethlehem. But Grace's business skill raises the question of whether he, had his career spanned a slightly later period, could have bested the challenges. The probable answer is no: He ardently believed in expansion, and the time to get bigger in steel was passing fast. His coping ability in any case did not get tested, because he suffered a stroke in 1957 and largely turned operations over to Bethlehem's president, Arthur Homer.

Grace, then over 80, continued for a little while to run meetings of Bethlehem's big board, 21 men strong. These meetings were sometimes bizarre, according to Bethlehem Steelmaker, an autobiography written many years later by Edmund Martin, who succeeded Homer as CEO. Grace, reported Martin, would on occasion nod off to sleep. The world then stopped: "Sometimes an hour would go by," wrote Martin. "We didn't leave. We didn't continue with whatever business was being discussed. We just waited in silence until he woke up and resumed the meeting as if nothing had happened."

Why, you ask, would an outside director put up with this madness? There were no outside directors then, nor had there been for many years. It took Ed Martin to bring in a few outsiders, in 1965. But only in the early 1980s did a new director, "Reg" Jones of General Electric, lead a reformation in which outsiders gained the majority of the board.

Well before that revamping, Art Homer did something more drastic: He lowered executive pay. Schwab had believed heartily in incentive compensation, and Bethlehem's executives proceeded to become famous for raking it in. Grace was for a while the best-paid executive in the U.S., making, for example, $1,624,000 in 1929. That was a mind-boggling figure for the time and even in today's high-comp world would be pay--expressed in current dollars of $17.5 million--that most CEOs would consider respectable. By the 1950s, Grace had conferred with himself and decided to earn much less. But in 1956 he was again the highest-paid executive, with earnings just over $700,000. Other Bethlehem executives scored too: In Business Week's survey of pay for 1957, nine out of the top 12 earners were from Bethlehem.

A 1959 shareholder suit then rocked this luxury yacht, noting piercingly that this pay had been set by a board including only insiders. CEO Homer buckled, moving to a less-rewarding formula for determining bonuses, and Bethlehem's crew went south on the pay list.

That is also a pretty good description of what this whole prideful company, run by these richly compensated men, was doing at the time. A strong case can be made that in the very early years of FORTUNE's 500--say, the period from 1955 through 1965--Bethlehem's eventual fate was sealed. Some books about steel describe its glory days as extending into the 1960s, but the industry's financial results refute that. Instead, a bitter war between revenues and costs had developed by the '60s, and the biggest victim it left on the battlefield was that critical measure of profitability: return on equity.

Not that you could possibly have known its importance by what Bethlehem was telling its shareholders. For this article, this writer browsed through the last 50 years of Bethlehem's annual reports and for a long spell searched in vain for ROE statistics or even an acknowledgment that Bethlehem knew what ROE was. Finally, in the 1971 report, Stewart Cort, a sales and marketing type who succeeded Martin as CEO, spoke in his chairman's letter of a "difficult" year and said, "Our net income was far from satisfactory, representing a return of only 7.1% on stockholders' equity."

The fact is that for 16 years, beginning in 1958, Bethlehem never came close to the 500's ROE, which for those years averaged 11.2%. Bethlehem consistently did no better than a single-digit return and sank down to 4.6% one year. Overall, its ROE averaged a mere 7.5%, almost four percentage points less than that of the 500. That's a huge inferiority, signaling a company that was in deep trouble. Grown-up businesses generating ROEs like that are the kind that investors want to run from, as fast as possible.

That point is made in a story often told by Jack Welch, who in his General Electric career got a call from a Pan Am executive asking whether GE might be interested in buying Pan Am's hotel subsidiary, Intercontinental. Welch, who knew bad from good when he saw it, said, "You're selling the wrong business. You should be selling the airline."

"We can't," said the Pan Am executive. "That's our business."

Confronted with what was happening to its business those many years ago, Bethlehem should have been tirelessly trying to unload its steel plants on some buyer. But of course that wasn't going to happen, wasn't even going to be thought about, probably couldn't have been carried out if it was thought about. Today Walter Williams, 75, one of four former Bethlehem CEOs FORTUNE interviewed (he was No. 6 in the nine who followed Grace), says ruefully that the course of events at the company, and throughout most of the country's integrated steel industry, seems to have been almost inevitable: "We were all stuck with our basic steelmaking--just too much to write off and too much to shut down."

Unable to cope with that calamity in any way, Bethlehem's bosses just dug their hole deeper, all the while contending with a set of impossible economics. The single piece of good news for the company as it burrowed was a gain in productivity--that is, reductions in the man-hours required to make a ton of steel. But this progress was impeded by both anachronistic work rules that the United Steelworkers of America defended tenaciously and capital expenditure needs that exceeded what Bethlehem could afford. Meanwhile the company's per-head employment costs rose irrepressibly, not only because of the steelworkers' tough demands but also because Bethlehem regularly extended whatever benefits labor won to its white-collar battalions. The last huge burden for both company and industry was extreme softness in prices because of the new competition. (Right now, steel prices are very strong; had they moved up earlier, Bethlehem's life might have been prolonged, but not likely saved.)

Foreign steel, of course, enraged Bethlehem and the integrated industry, which believed it to be unfairly priced. The mini-mills initially bothered the industry much less, because it saw the new ventures as little more than annoying gnats, buzzing around making inconsequential niche products like reinforcing bars. But these producers then moved up the food chain, first beginning to make plate and then structural products. That hit Bethlehem where it literally lived--in Bethlehem, Pa., site of the company's big but antiquated structural steel mill. This plant, which was for years sentimentally kept on life support, was finally and mournfully closed in 1995.

A few statistics show the brutality of steel's overall economics. A Purchasing magazine index of hot-rolled steel sheet prices shows them rising in the 1954--2003 period by a dismal 220%. In contrast, the consumer price index increased by 540%. And steel wages? According to the American Iron & Steel Institute, the average wage was up by more than 900%.

Knowing these disparities might make Schwab and Grace turn over in their graves, though Grace did get an early taste of union power. He was forced in 1941 to let the steelworkers union into the company. Thereafter, Bethlehem, typically joining in industrywide bargaining, grimly settled into triennial negotiations in which the union worked unbendingly at keeping the maximum number of its members employed at the highest possible pay. That vault in wages shows how well the union succeeded. Worse--this is an industry in which that word is persistently relevant--the industry let itself be locked into work rules and narrow job descriptions that protected workers whom automation and efficiency should have made superfluous.

The point is driven home stunningly by the contract changes that International Steel Group wrung from the steelworkers for those six Bethlehem plants it began operating in 2003. ISG's CEO, Wilbur Ross, who characterizes Bethlehem's contract as "terrible," got the union to allow the cutting of job categories in his plants from 32 to five! As just one example, there is no longer a job category called electrician, which means that when a light bulb needs changing, a machine operator doesn't have to call such a specialist in (perhaps from the far reaches of a plant) but can screw in a new bulb himself.

The ultimate error was the steel industry's approach to pension and health benefits. The history here is revealing. From World War II on, wage and price controls intermittently slowed wage increases. Written into the contracts as offsets, though, were a long string of benefit improvements. These took up their role as company killers.

Management, however, originally viewed its benefit promises as benign compared to wage increases calling for immediate cash. Health costs were modest for many decades and, besides, were normally pay-as-you-go. That is, few companies during this era bothered to fund health benefits in advance by putting money into a trust for employees. Until the 1990s, when new accounting rules came in, pay-as-you-go also meant that companies were not accruing expenses for what they would eventually have to pay retirees. Corporations were thus left oblivious to true costs and real earnings.

Pension plans, in contrast to health and insurance plans, were normally funded and eventually had to be. But before that requirement took hold, companies had wide latitude as to how much, if anything, they would contribute in any given year. Consequently, a lot of the pain related to pensions could be pushed into the future if that's the way management wanted it.

David Roderick, 79, who ran U.S. Steel in the 1980s, says in fact that differences in pension strategy between his company and Bethlehem significantly contributed to the fact that U.S. Steel survived ("We were never even remotely threatened with bankruptcy") and Bethlehem didn't. U.S. Steel's strategy decades ago was to aggressively make contributions to its pension fund, book these as the deductible expenses they were, and exit with a reduced tax bill. Bethlehem took the approach that its cash could better be used for modernization of its plants. Bethlehem, Roderick says, even used the difference between its strategy and U.S. Steel's as a selling point, telling its customers that Bethlehem was investing to improve its products whereas Big Steel wasn't.

Bethlehem did pour major amounts of money--$7.2 billion in the 1960s and 1970s--into modernizing old plants and building a new integrated mill in Indiana, Burns Harbor. There even came a glorious time, in late 1973 and much of 1974, when demand for steel surged, ROE leaped upward, and the investments looked very wise. But that was one of steel's false springs, the greenest of several that came along and briefly, if fraudulently, lifted the hopes of all. By 1975 the industry was again flattened, and was in fact moving into a new and prolonged era of closing plants.

Lewis Foy, Bethlehem's CEO in the late 1970s, described the company's past capital expenditures, heavy though these were, as "far short of our requirements." He then put his mouth where his money wasn't by closing some outmoded capacity in 1977 and booking $791 million in nonrecurring charges that gave Bethlehem its first loss since 1933. Of the total charge, $483 million was for pension costs and other termination benefits due both workers in the affected plants and thousands of white-collar people shoved out as well.

This huge amount, $483 million, drove home the terrible fix Bethlehem was in: Every time it closed a plant, it accelerated the retirement of longtime workers, rang up a bill for shutdown benefits, and put new burdens on a pension plan that by the 1970s the company acknowledged was underfunded. Bethlehem couldn't just ignore this weakness either: It had to build up the pension plan and, leaving aside any outside capital it could bring in, finance this repair work with money it was currently generating. What Bethlehem had here was simply a demographic nightmare, in which an ever-shrinking number of active employees were charged with making profits sufficient to support the present and future of an ever-growing number of retirees and dependents.

Bethelehem's tumbling employment numbers tell the story: Head count peaked at 167,000 in 1957 and by the mid-1980s was down, gulpingly, to 35,000. And at that point (though this certainly wasn't the end of it) Bethlehem had 70,000 retirees and dependents! It also had a pension plan that it judged underfunded by more than $1 billion. The company was meanwhile feeling the great rise in medical costs that walloped the 1980s. And for its medical promises to retirees, Bethlehem had no funding at all. None.

What it had instead was this colossal burden called "legacy" costs. Legacies are usually good when they come from a kindly spinster aunt. They are colossally bad when they are handed down by corporate hierarchies that both overpromise and fail to plan for the day that payment comes due.

From 1980 on, Bethlehem had five CEOs who, on the one hand, agonized about the legacy costs they inherited and, on the other, created new ones by plant closings and layoffs and even by labor settlements that added to the benefit burden. It was a never-ending, expanding mess. Lewis Kaden, a Davis Polk & Wardwell partner who joined Bethlehem's board in 1994 (indeed, as a representative of the steelworkers) has the perfect mental picture: "Bethlehem," he says, "was like an old man carrying bags of rocks on his back, with the man getting ever frailer and the rocks getting ever heavier." Carry that thought a bit further: It was the old man himself who bent to pick up the rocks he was loading into his bag.

Once in a while Bethlehem's brass seemed to recognize that those rocks meant ruination. As early as the mid-1980s, the board considered putting Bethlehem into bankruptcy. But then one of those false springs came along, and pretty soon the company was successfully selling new issues of common stock. In fact, during the 1980s and 1990s, Bethlehem raised nearly $1 billion with equity offerings (that, among other things, provided money for its pension fund). Clearly speculators never die.

In the 1990s there were other signs of premonition, evident in the company's moving to annuitize a large portion of the supplemental retirement benefits scheduled to go to top company executives. That meant Bethlehem's general credit no longer stood behind these pension promises, but that one or more insurance companies--well paid for assuming the obligation, you can be sure--instead became liable for eventually paying up. The executives thus moved themselves largely out of the line of fire.

And yet Bethlehem's CEOs, though they wore out the word "difficult" when summarizing their years in annual reports, never seemed to give up. Sure, Walt Williams retired in 1992, feeling after six years in the job "totally disappointed, totally disillusioned, and worn out." But his successor, Curtis H. "Hank" Barnette, introduced a brave new slogan: "Our vision is clear--to be the premier steel company." And his successor, Duane Dunham, came in talking about new strategies that will "ensure our long-term success."

Even Robert S. "Steve" Miller, the restructuring expert whom the board hauled in as CEO on Sept. 20, 2001, took on his job optimistically: "I did not come here," he said upon arriving, "to put this company into bankruptcy." But within weeks, on Oct. 15, he did exactly that.

And the frail, old man, lugging his rocks? After Miller himself had made some head-count cuts, there were--though it's almost unimaginable--11,500 active employees and 120,000 retirees and dependents.

As 2002 ended, the company was also showing--we are now measuring its "legacy"--that the value of its pension fund was $3.7 billion and that trailing along was an unfunded obligation of $2.9 billion. In other words, by Bethlehem's actuarial figuring, the company was that much shy of having enough to ultimately pay all its pensioners and their covered dependents. When the Pension Benefit Guaranty Corp. (PBGC) took over the fund--the biggest it has ever assumed--and figured liabilities on a more conservative basis, it upped the unfunded obligation to $4.3 billion. But by its rules, the PBGC will cover only $3.6 billion. So that's the bill this agency (which is financed by assessments levied on all companies having defined-benefit plans) expects to shoulder.

As things work with the PBGC, it will pay a maximum of $44,386 this year to each Bethlehem pensioner. In addition, the executives who did the annuity hop will be getting a second check from some insurance company. Nobody, though, will be paying that other part of the legacy: the health and insurance benefits that Bethlehem promised and didn't fund. Created cavalierly in a succession of union contracts, this "obligation" amounted at year-end 2002 to a mountainous $3.1 billion. That's simply a promise that evaporated into thin air.

You may reasonably ask whether the steelworkers should not have, for their own long-term benefit, eased off on Bethlehem way back when. One answer is that unions don't care for "givebacks." Another point, says director Kaden, is that were the union to have done that, it would have been making sacrifices not shared by other interested parties, such as stockholders or creditors. As it was, bankruptcy creamed everyone. The shareholders got nothing, the creditors got very little, and the workers, at the least, got stripped of their health benefits.

It is also apparent that Leo Gerard, president of the steelworkers, was ready by bankruptcy time to be shed of Bethlehem. Trying to pull something from the ashes, Steve Miller proposed reorganization plans whereby Bethlehem's plants would be freed from the work rules that had stood in the way of efficiency. Gerard rejected Miller's ideas, preferring to do his retailoring of the work rules with ISG's Ross. Gerard thinks Bethlehem could not have overcome the bloat in its management nor tolerated his union members becoming working partners on the plant floor, as is the scheme at ISG. Ross himself doubts that Bethlehem's bureaucracy could ever have learned what lean management is about. At his shop, he says, there are three layers of management between CEO and factory floor--at Bethlehem, there were eight.

FORTUNE asked Ross and a number of other informed people whether they thought some really great businessman--say, a Jack Welch--could have come into Bethlehem maybe 40 or 50 years ago and saved this company. All thought the question intriguing, but unanswerable.

So we asked Jack Welch himself. As a buyer of steel at GE, he knew Bethlehem. So, how about it, Jack--could you have done it? He immediately responded that he wouldn't dream of being so "presumptuous" as to think so. Then he puzzled a while longer and finally said simply, "I don't think Christ could have done it." Amen.

FEEDBACK cloomis@fortunemail.com