(FORTUNE Magazine) – The U.S. economy in the mid-1950s was colossal in size and the envy of the world. Yet American businessmen were weary of decades of turbulence--of the "underconsumption" of the Great Depression and of wartime shortages and postwar booms--and yearningly ready for what was often termed "normality." Fortune, however, starting a new series in 1953 called "The Changing American Market," thought tranquillity not too likely. The barriers in its way, we said, were the well-known characteristics of the American people--"the restless, enthusiastic energy, the lack of traditional impedimenta, the almost dogmatic optimism, and the special delight in the brand new."

It was a paradox of those vibrant times that the bigness of U.S. business was not well measured. Starting a small attack on the confusion, Fortune began in the early 1950s to compile, for its own use in identifying stories we should be doing, an annual list of the largest U.S. companies. Then one day came the big, light-bulb idea, advanced by an editor named Edgar P. Smith: "I think," he said, "that our readers just might be interested in this list." And so we published--in 1955, 40 years ago--the first Fortune Industrial 500.

I was a rookie on the staff then and played a minor, data-grubbing part in putting the first list together. I would like to claim I knew what the 500 was to become, but it wouldn't be true. I remember no sense of excitement on our floor as we worked up the list, no glimmer at all that we were launching an institution. Next thing, though, and with a quickness that was amazing, the term Fortune 500 moved into the language, began to define might, and even started to induce some companies--we don't approve of this, but it happens--to budget their place on the list.

We had a box-office smash, no question. Then came the improvements. We added new lists that eventually became the Fortune Service 500; we reclassified some companies from industrial to service; and we added important new financial information, such as return on equity and total return, to the industrial and service lists alike. In 40 years there has in fact been unending change, capped off by the new list that begins on page F-1.

Even so, those changes in format pale beside the changes that whipped through the companies on the Industrial 500, shaking it up annually and over time enormously transforming its looks. In its entirety, the 500 has been a wide window on the U.S. economy and a reference point for all. It has been cited by Presidents and Congressmen, dissected by economists, and viewed by business leaders--still yearning for normality and never quite finding it--as the Bunyanesque yardstick of industrial progress.

In its components, it has also precisely delineated this business world that never stands still. During the 39 years before this one--that is, in the list's industrial--only period--no fewer than 1,318 different companies appeared on the list at least once. Some subsequently left it because they shrank in size and some because they ceased to be industrials by the definition we used, which was that over 50% of a company's sales had to come from manufacturing or mining. Another huge sweep of companies disappeared from the 500 because they were acquired. Merger waves in the late 1960s and in the 1980s, mania time for leveraged buyouts, particularly sped the acquisitions.

On the flip side, in a fact that powerfully demonstrates the dynamism of the U.S. economy, an astounding number of companies have grown to 500 size from recent, absolute nothingness. About 90 companies on the list in this issue did not exist when the first 500 was published 40 years ago. True, the newly born include some instant giants, such as the seven Bell operating companies carved out of AT&T when it was broken up in 1983. But many other companies grew from infancy, among them such bigtime players today as FedEx (founded in 1971), Home Depot (1978), McDonald's (1955), Nike (1964), Apple (1977), and Microsoft (1975).

The new kid ranking highest on today's 500 is mighty Wal-Mart (No. 4), whose first store went up in 1962. A jump down the line is Price/Costco (No. 47), the company created by the merger of two warehouse-club retailers that aren't even 20 years old: Price, of San Diego, founded in 1976, and Costco Wholesale, of Kirkland, Washington, born in 1983. Here's a cultural thought to ponder: Computing was on the map in 1954, though barely; warehouse clubs didn't exist then at all.

In the rarefied air of the 500's Big Ten, the long-term record shows both stability and remarkable change. In recent weeks, as I worked on this article, I have asked friends to guess which companies made up the top ten in 1954. Everybody promptly gets General Motors and Exxon, then called Standard Oil (N.J.). Most people also guess Ford, which was not on the list, though only because it was then still a private company. Some people come up with General Electric, U.S. Steel, and Du Pont, and vaguely add "other oil companies." But does anybody guess Swift and Armour? Are you kidding? Yet there they were--see the table on the previous page--at spots 5 and 7.

Their top-tier rank in 1954 suggests both the amount of red meat then being consumed by Americans and the extent to which they were giving their business to big-city, and particularly Chicago, packers. Soon after, upstart companies, among them newly formed Iowa Beef Packers (now IBP and No. 83 on the 1994 list), gained efficiencies by locating in stock-raising areas and began taking market share from the leaders. Chickens did their damage also, aggressively pecking away at the shares of beef and pork.

Gradually, Swift and Armour slid down the list and then disappeared entirely from it because they turned into red meat for acquirers. Both are now a part of ConAgra, No. 24 on the 1994 list, and a company that in 1954, when it was known as Nebraska Consolidated Mills, had sales of only $42 million. That didn't qualify the company for the 500, whose smallest member that year had sales of $49.7 million.

General Motors can brag in certain ways of stability: The 500's history shows it ranking first 31 times out of 40, and in this year--the 41st--it again leads. Exxon is the only other company to have topped the list, grabbing first from 1974 through 1976 and again from 1979 through 1984. The position shifts of these two companies are tightly linked to the workings of the market. Most of Exxon's years were periods of price spikes in oil and gasoline and of consequent decisions by millions of Americans that they'd just as soon not buy cars-or at least not the big ones rolled out by Detroit.

In fact, a deeper look at GM exposes huge cracks of instability. In 1954, when the 500 started, GM was the invincible giant of the automobile industry, with a U.S. market share that exceeded 50%. Even so, the company was widely suspected of deliberately holding down its market share so as not to raise antitrust problems.

Then came the onslaughts of foreign automakers, falling on a set of Detroit car guys generally out to lunch. Today GM's U.S. market share runs below 35%. Erosion in the company's dominance is also visible in one 500 statistic. In 1954, GM's sales of $9.8 billion accounted for 7.2% of the total sales of the 500. The company has pushed out many millions of cars since and put in its share of price increases. (For Cadillac's, see the price chart on the preceding page.) But had we published an all-industrial 500 this year--and we did indeed compile that 500 to learn precisely this fact and many others--GM's sales of $154.9 billion would have amounted to only 6% of the total. That's still a king-size wedge. The difference of 1.2 percentage points, though, is equivalent to about $31 billion in sales that GM didn't get and certainly could have used.

Obviously, GM is one large link between the first 500 and the list in this issue. But just how many companies in total were on that kickoff list and are still in the 500 today? This is the question we are most often asked about the 500, and it is apt to keep coming in even though we have broadened the list beyond industrials. So here are two answers today: Had we published an industrial-only list this year, it would have included 160 companies that were on the 1954 list. And 116 of those companies have the distinction of making the new, significantly altered list. You will find this select crew--the survivors, so to speak--arrayed along the sides of these two pages.

The sharp-eyed reader may spot an occasional outlander. It's unclear just why we classified Melville, the retailer, an industrial in 1954. Maybe I, as a lowly and clueless staff member in 1954, made that mistake. Or perhaps we designated Melville an industrial because it manufactured many of the goods, particularly shoes, that it sold in its stores--Thom McAn shops, mainly. In any case, next year we changed our mind, putting Melville on our new list of retailers. This year it's No. 93 on the new 500.

Alongside our list of the survivors you will find the statistic that tells the ultimate story about their long-term performance: total return in the stock market for the 40 years. These are not data you have seen before; they are unusually revealing about companies and industries, and they also deserve a bit of explanation. Most of the returns were calculated by the University of Chicago's Center for Research in Security Prices, and the rest by Fortune (sometimes working with data that, for the early years, were skimpy). When a company is the product of a merger--Unisys would be an example--the returns for the pre-merger years are those of the party deemed dominant, which in Unisys's case is Burroughs.

The percentages shown wrap in both price changes and dividends (reinvested when they are paid), and are expressed as annual rates, compounded. For five companies out of the 116, though, we could not come up with a total-return finding. That's because there were gaps in the price data for each of the five, among them RJR Nabisco, Owens-Illinois, and American Standard.

These returns happen to be calculated off a year, 1954, that in stock market history was a dramatic attention getter. The Dow Jones industrials opened that year at 276 and closed at 404, having gone hair-raisingly through the 1929--and previous all-time--high of 381. November and December in 1954 were months of wild trading, including days when volume on the New York Stock Exchange actually got above four million shares! (On the busiest day in 1994, NYSE volume was 482 million shares.) A great buzz went through financial circles, as various experts pondered whether speculation was about to ruin America. Politics being then as it is now, Senator J. William Fulbright (D-Arkansas) scheduled hearings on the stock market, posing the question of whether "a new stock market collapse on the 1929 model lay directly ahead." Even the famous investor Benjamin Graham, called as a witness, opined that leading industrial stocks "definitely are not cheap."

From this background of anxiety, we are happy to report that since 1954 the median total return for the 111 survivors for which we have figures is--roll the drums!--a highly respectable annual rate of 12.3%. For perspective, the rate of inflation during the 40 years was 4.4%, and Standard & Poor's 500-stock index had a total return of 11.5%. So capturing the median return of these 111 survivors would have given an investor a considerable advantage.

Where there are medians, there are also superlatives. The winner is . Philip Morris, with a truly extraordinary annual return of 20.2%. Let's put that in dollars. If you had bought $1,000 of Philip Morris stock at the end of 1954 and reinvested all dividends, your stake at the end of 1994 would have been $1,561,100.

And however you feel about smoking, you would have to give tobacco the credit. Philip Morris has been a prodigious acquirer, buying Miller Brewing, General Foods, and Kraft, among other companies. But it is tobacco that has spun out the earnings, provided the cash flow for both large acquisitions and steadily rising dividends, and almost solely accounted for Philip Morris's superior record. The tobacco industry's ability to raise prices has driven earnings: Since 1954, a pack of Marlboros--then a weak brand that had not yet gone cowboy--has risen in price (with taxes excluded) more than 900%. The upward trends, of course, took a brutal licking in 1993, when Philip Morris suddenly announced it would cut prices so as to combat market share losses. Total return in 1993 was an embarrassing minus 24%. Last year the company got the game restarted with a plus 8.8%.

At the opposite end of the spectrum, the 111 companies include a fizzling four that did not keep up with that 4.4% annual rate of inflation for the 40 years. They are Bethlehem Steel, at 3.4%; USG, a onetime bankruptcy victim, at 2.4%; and two outright losers, Navistar (once named International Harvester), at minus 2.8%, and Manville, another bankruptcy disaster, at minus 3.5%.

Good pickers of industries could have had a real run over the 40 years. Among the major groups (not counting tobacco), pharmaceuticals did the best, with a median of 15.8%. Even the worst of the pharmaceuticals, Pfizer, Warner Lambert, and Lilly, recorded 14.7% rates of return. Food and beverages came in second among the industries, with a median of 14.6%. On the other hand, woe to the investor who chose to be in metals: The steel, aluminum, and copper survivors had a median of only 6.6%. Over the years, pundits have deplored the reluctance of the major U.S. steel companies to pour money into modernizing their plants. The total returns they have been delivering investors suggest why no one is rushing to throw good money after bad.

Progress in the stock market has a kissing cousin: progress with customers, for which one kind of reading is sales growth. For the survivors, the median rate was just under 9%. The top rate of growth, 16.04%, belongs to PepsiCo (which had almost precisely the same total return rate). Pepsi's leaps in sales, garnered from both acquisitions and internal growth, took it from 364th on the 1954 list to 20th on the new list. Close behind in sales growth was Philip Morris, with a 15.98% rate. The cellar dweller was Bethlehem Steel, with a crawl rate of 2.7% (quite comparable to its dragging total return rate of 3.4%). Bethlehem's limps dropped it from 11th on the first list to 239th on the latest.

As these three examples suggest, sales growth and total return for the 111 were significantly associated, but by no means precisely. Sales increases didn't tell the story, for example, at Crown Cork & Seal, a manufacturer of cans and other containers. Crown Cork's rate of sales growth for the 40 years was only 9.7%, but its total return was a classy 16.8%, sixth highest among the 111. The company's strategy was huge, judicious repurchases of its own stock, an approach that did little for sales growth but that added value to every share left on the books.

In general, the survivors had to grow just to keep up with the Industrial 500, which in 40 years increased its sales at an average annual rate of 7.6% and had only four downdraft years--in 1958, 1982, 1986, and 1991. That 40-year rate makes the 500 look like a near twin of the U.S. economy, whose annual rate of growth, as measured by GDP, was 7.5%. But in fact the 500 and GDP were on far different tracks, which become visible when the 40 years is broken down to shorter periods.

For the 500, the tracks switched from fast to slow. During its first two decades, from 1954 through 1974, the 500 had growth in sales--the average annual rate, compounded, was 9.5%--that greatly outstripped the rise in GDP, which was 7.1%. But from 1974 on, the relationship dramatically reversed, with GDP leading 8% to 5.6%. The change almost certainly reflects the growing importance of the service economy, whose robustness showed up in the Industrial 500 only to the extent its companies had stakes in service businesses.

In another and critical change of tracks, the 500 went from significant growth in employees to rapid reduction. From 1954 through 1979, this universe, though it sometimes faltered in its hiring, added employees at an average annual rate of 2.9%, increasing its work force from 7.9 million to 16.2 million. But 1979 turned out to be the peak year, after which the 500 put on its productivity britches and got tough about hiring. In each year since, with the exception of two, the employee count fell. One exception was 1984, and the second would have been 1994, had we published an all-industrial list. Even so, the 1994 employee total of 11.6 million is only a shade higher than the total was in 1965, 30 years ago.

Not one of the 500's industries, not even computers, has escaped downsizing. But some of the employee reductions seem especially noteworthy. In 1980, a boom year for oil, the seven largest refining companies in the 500 employed 665,000 people. One of the seven, Chevron, later bought another, Gulf. A couple of companies have also shed nonoil divisions. But that doesn't really explain why this package of seven companies (counting Chevron and Gulf as two) gets along today with only 270,000 workers. Fact is, the managements realized they were drastically overstaffed and finally slimmed from fat to lean.

The weirdest tale in the 500 concerns its profits. Not surprisingly, these bobbed around over the years, dropping during recessions and picking up steam as the economy recovered. In the first 37 years of the 500, the worst drop for profits occurred in the recession year of 1982, when the decline was a horrifying 27%.

But then the 1990s produced a bookkeeping mega-event: Financial Accounting Statement 106, which provided that corporations must begin to recognize their liabilities for retiree health benefits. In reality, these liabilities had existed for many years; they simply weren't acknowledged. Had they been faced up to earlier, their effects would have been spread out over all the years of the 500, with resulting reductions in profits.

Instead, FAS 106 carpet-bombed the years of 1991 through 1993--and particularly 1992. In that year the 500 fell for the first time into the red, recording a loss of $196 million. Making a special calculation, Fortune reported that FAS 106 charges had knocked profits down by no less than $71 billion. Of that, $20.9 billion was the work of General Motors, which had spent years caving in to the health benefits demands of the United Auto Workers and in this year to be remembered, 1992, paid the piper. GM also had some other 1992 losses to own up to and for the year reported by far the biggest loss in the 500's history, $23.5 billion. GM was also, by way of a $4.5 billion loss, the red-ink champ in 1991. Then there was 1990, when the bad news was only $1.9 billion.

Other repeat "winners" of the annual-loss championship include two defunct automakers, Studebaker Packard (the leader in three years) and American Motors (two); LTV (four); and Chrysler (five). Chrysler's loss leadership--last exhibited in 1980--is obviously of no moment to Kirk Kerkorian, current bidder for the company.

No one could have forecast in 1954 that the auto companies were about to traverse many years of trouble, and few would have had the wit to pick Philip Morris as the 500's long-term leader in total return. The difficulties of choosing winners and goats for the next 40 years are equally severe.

But we expect to be here in 2035 to report what happened-well, not me personally, given that I regularly describe myself as "Jack Benny's age." But maybe some rookie now on the staff will get the job, and in that person's behalf I issue a challenge. Write us stating your opinion as to which company today on the 500--reminder: page F-1--will still be there 40 years from now and will have had the best total return. If your submissions are interesting enough, we'll summarize them soon in our Letters column, and, if this magazine can only hold on to its institutional memory, we'll give the best forecaster (assuming she's held on) a prize in 2035. Of course, if you really pick the total-return winner and back your opinion with money, you won't need a prize. You'll already be rich.