(FORTUNE Magazine) – TO FULLY FATHOM the depth and pervasiveness of the ailments that forced General Motors to order the greatest upheaval in its modern history, consider the 1992 Cadillac Seville STS. GM is billing the redesigned $38,000 luxury sedan as its flagship, the embodiment of its engineering superiority, its answer to Mercedes and Lexus. The car is vastly superior to the model it replaces and has won several awards from enthusiast publications. And yet its introduction was delayed for a year because of design problems. A new multivalve V-8 engine planned for the car won't be ready until 1993. Production ran behind schedule owing to difficulties with the paint system, among other things. The plant in Hamtramck, Michigan, where the Seville is built, has operated at little more than 50% of capacity since it opened in 1985, rendering it perpetually uneconomic. Such are the wonders and the woes of the world's biggest manufacturing company. GM is capable of achievements bordering on brilliance. Today it builds the best cars and trucks in its history. In the 1992 model year it is introducing 16 new vehicles -- the most ever. J.D. Power & Associates ranks Cadillac and Buick among the industry's ten top makes for customer satisfaction. GM is a world leader in several critical technologies, including antilock brakes and electric cars. But the company is foundering. Even with the massive cutbacks announced by Chairman Robert Stempel in mid-December, it lags behind its major competitors in almost every measure of efficiency. By some key standards -- how many worker-hours it takes to assemble a car, for instance -- GM is an astounding 40% less productive than Ford. In 1991 GM lost, on average, $1,500 on every one of the more than 3.5 million cars and trucks it made in North America. It ended the year with barely more than 35% of a U.S. market that bought fewer than 13 million new cars and light trucks; in 1979 the market exceeded 14 million and GM commanded 46% of it. Efforts to reform the company have been crippled by a stubborn middle-management bureaucracy, as well as the take-no- prisoners attitude of the United Auto Workers. Union locals still strike at the drop of a wrench to protest work rule changes designed to improve efficiency. Much is at stake, not only for GM but for the entire automotive industry and the U.S. economy. In North America, GM employs 429,000 hourly and salaried workers, enough to populate Fort Worth, the U.S.'s 30th-largest city. Some 21% are blacks and other minorities -- the largest minority work force of any private employer. GM's 30,000 suppliers and 10,000 dealers provide jobs for another big city's worth of people. The company is America's largest consumer of steel, rubber, glass, plastic, and carpeting. Thirty-nine years ago GM's then-president, ''Engine Charlie'' Wilson, remarked that what was good for the country was good for General Motors, and vice versa. He was lambasted for saying so, but it is hard to envision a prosperous U.S. without a prosperous auto industry -- and a prosperous GM. GM's long decline has taken a toll among its more perceptive managers, who see a company that has succeeded at a critical task it set for itself -- building better cars and trucks -- but has yet to confront the enormous structural problems that afflict it. Says one: ''There is a monumental challenge ahead of us. We can make great products. But can we do that and make money?'' Spirits were buoyed somewhat by news of December's changes. Says another insider: ''We did something we should have done in the mid-Eighties. I think Bob Stempel has taken charge, and what he is doing will create a lot of internal change.'' Salvaging GM requires not reform but further upheaval. In addition to eliminating the jobs of thousands of white-collar workers and closing half a dozen more of its 33 assembly plants, it needs to restructure radically to get more efficient at everything it does. That will mean sacrificing some longstanding GM shibboleths, such as maintaining a separate marketing operation for each of its six car divisions -- Chevrolet, Pontiac, Oldsmobile, Buick, Cadillac, and Saturn. Keeping the company going financially will not be easy: After spending an astounding $90 billion in the 1980s on new plants, equipment, and acquisitions, GM is nearly broke. The company's future is in the hands of CEO Stempel, 58. After 33 years with GM, including five on the board of directors, Stempel, an engineer, knows cars & -- and GM -- inside and out. But like everybody else there, he is the product of a highly centralized and insular culture that has always had huge resources and abundant time to bring to bear on any problem. A onetime football player several inches over six feet tall, Stempel has made thousands of friends at GM during his 18-month tenure as CEO because of his loyalty, his skill at promoting teamwork, and his willingness to delegate. He has been known to return memos to subordinates with the handwritten message, ''Why are you sending this to me? This is your decision.'' FOR ALL HIS TALENTS, some critics say, he sometimes gets lost in details, deliberates overlong on problems, and prefers evolutionary, not revolutionary solutions. He has refused to consider changes in the cumbersome six-division marketing system and has promised never to reorganize the company. Stempel resolutely avoids public exposure when he can, leaving to deputies the operating levers that move the company. Says a Chicago management consultant who has often worked at GM: ''I've heard guys in the organization ask, 'Where's Bob?' He has a lot going for him. But it isn't clear that he's using the chips he's got to make the organization move.'' Declaring they were too busy working out details of the cutbacks, Stempel and all six other members of GM's management committee declined to be interviewed by FORTUNE for this article. Understandably, few of the dozens of people inside or outside GM who commented for this story wanted to be quoted by name. A series of horrific events in the fall of 1991 forced Stempel to move. One top GM official allowed that even the direst of pessimists could not have foreseen the company's predicament. Despite a sunburst of new models and heavy buyer incentives, GM was selling 600,000 fewer cars and trucks than in 1990; bread-and-butter Chevrolet alone accounts for two-thirds of the drop. After losing $2 billion in 1990, GM now faced a companywide deficit of nearly $3 billion for 1991. Seeing GM running short of cash, and with economic recovery nowhere in sight, Standard & Poor's announced in November that it was putting GM on a credit watch. For a company that once routinely luxuriated in triple-A ratings, the prospect of a downgrading was not only an embarrassment but a financial menace. Borrowing costs for GMAC, which is GM's finance arm, could increase by $200 million or more annually. The company's deteriorating fiscal condition got the attention of GM's board of directors. At their regular December 9 meeting, in Washington, D.C., the directors pushed Stempel to move faster to cut costs by closing plants and getting rid of white-collar workers. Several days later, a sign of urgency and confusion in high places emerged: GM had to delay the closing of a $1 billion stock sale because it hadn't accounted for the multibillion-dollar cost of the write-offs required for the downsizing when it priced the stock. THE CUTBACKS finally announced on December 18 will help get GM's payroll and production capacity more in line with its sales. By 1995, GM will close six assembly plants and 15 other factories, reducing its capacity by one-fifth so that it can operate profitably with a 35% market share. In the process, it will cut 74,000 blue- and white-collar jobs, trimming its work force to half the size it was in 1985. But while productivity -- the number of cars produced per worker -- will improve, efficiency -- the number of worker-hours required per car -- will not. Likewise, the cuts surely got the attention of managers in GM's famous ''frozen middle'' but did nothing to refocus their energies, restore potency to GM's waning brands, or unshackle a company tied down by a contract that forces GM to pay its blue-collar workers whether they are on the job or not. GM has been closing plants for nearly a decade, but this could be the most expensive round ever. It now has to consider shuttering facilities that were either built or extensively modernized in the 1980s. Five of the plants that make midsize GM cars are scheduled for consolidation into four or fewer. Four of them were built new or completely renovated in the 1980s at a cost of $1 billion each. With characteristic loyalty, Stempel has refused to replace his longtime sidekick and onetime rival, President Lloyd Reuss, 55. Reuss has suffered a number of career setbacks. He put middle-American Buick into a long skid by introducing European-style performance models for younger buyers and created an unwieldy, top-heavy bureaucracy at GM's small-car group, Chevrolet-Pontiac- Canada. Yet in 1990, Stempel insisted over board opposition that Reuss be named president. (The board withheld the title of chief operating officer.) Several sources report that when news of Reuss's appointment was broadcast throughout GM in 1990, a number of white-collar workers booed and jeered. Says Ronald Glantz, longtime auto analyst at Dean Witter in San Francisco: ''If Lloyd Reuss were to quit tomorrow, it would add a couple of points to the stock.'' The man being talked about as Reuss's successor is John F. ''Jack'' Smith Jr., 53, vice chairman and head of international operations. Smith, a son of Massachusetts who retains his Boston accent, won acclaim for helping turn around GM's European operations in the mid-1980s. He is known as a down-to- earth, results-oriented manager. Surely not all of GM's problems are its alone (see charts). Ford and Chrysler have also suffered mightily in the recession, laying off tens of thousands of workers and scaling back their horizons. Their dwindling fortunes have devastated Detroit, the seventh-largest U.S. city. Auto parts suppliers all over the Midwest are failing, and unemployment in Michigan has flirted with double digits. But Ford and Chrysler have one advantage over GM: They have been through the wringer before, when both companies nearly went bankrupt in the early 1980s. Although Detroit auto men are reluctant to criticize each other, a high executive at one of GM's competitors described its problems this way: ''GM has been operating by the philosophy that, one, there is no problem that is so intractable that we cannot solve it by throwing money at it, and two, we've got all the money in the world, and if that isn't enough, we'll get some more.' It is not working, and the cash has been just hemorrhaging.'' By the end of 1991's third quarter, in fact, GM was down to $3.5 billion in cash -- half the amount it usually keeps on hand to meet payrolls and pay suppliers. Estimated fourth-quarter losses of $700 million even before the write-off, plus dividend payments of $285 million, meant that the company needed the $1 billion from the preferred stock offering just to stay even. Smaller Ford had $7.3 billion, and even Chrysler, often described as troubled, had $2.5 billion -- much more in proportion to its revenues than GM. In mid- December, GM's stock fell to a four-year low of $26.875 per share. Long- term GM investors have taken a terrible beating. While the S&P 500 has more than quadrupled since 1965, GM common stock, adjusted for splits, is worth little more than half what it was then -- even without adjusting for inflation. GM has been paying out more than it takes in for nearly two years, so it desperately needs fresh capital to replenish its bank account. Altogether in 1991 it issued $2.4 billion in new common and preferred stock and took the unusual step of selling and leasing back some $650 million worth of assembly plant equipment. Since 1985, GM's long-term debt has more than doubled and now amounts to a steep 35% of equity. Because the auto business is nearly stagnant, that level of debt will put a drag on earnings and make it tough for GM to borrow more. The board cut the dividend from $3 to $1.60 a share in February and is likely to slash it further. GM's cash squeeze is limiting its strategic choices and weakening its potential for future gains. It now plans to reduce capital spending by $1.1 billion, or 8%, in 1992 and 1993. While Stempel says the cuts won't mean eliminating major new cars or trucks, analysts say GM has delayed for a year or more a number of important programs, including reworking the high-profit Chevrolet S-10 pickup. The plant in Linden, New Jersey, that was being readied to produce the new truck will now stand idle until 1994. Fixing GM will be a mammoth undertaking. Like a man out in the cold wearing a thick down parka, GM was protected by its bulk. Says the same high-placed Detroit competitor: ''There is a sense of invulnerable permanence at GM. There is not a single guy there who's had to deal with a tough situation like this before.'' At the same time, says a New York City consultant who knows GM intimately, ''Stempel and his managers are clearly in a situation where if they don't take action in a continuous way, they can get weaker and weaker every day.'' Sympathetic consultants and industry experts suggest some drastic remedies to improve its sub-par performance:

-- Reorganize. Unlike any other auto company in the world, at GM neither the chief of design nor the head of research reports to the automaking side of the business. Instead, both report to the head of R&D, who in turn reports to another executive who runs the aerospace and computer divisions. Besides being cumbersome, the arrangement penalizes GM in two ways. Designers don't work closely with vehicle engineers in creating a new car, so development takes longer and costs more. Scientists don't either, so GM is slow to apply new technologies even though it is quick to develop them. Where Stempel has moved organizations around, GM hasn't yet gained efficiency because it hasn't eliminated redundant operations or people. Last fall he finished combining four separate divisions that cast engine blocks and assembled engines and transmissions; he melded them into a single behemoth Powertrain Division that employs 67,500 (only a little less than Chrysler's entire U.S. payroll). But few have lost their jobs so far, and GM is only now getting around to reducing the number of basic engines it makes from nine to five. The company produces three different four-cylinder engines and imports a fourth, even though all are comparable in size and output.

-- Instill accountability. According to a thoughtful 1988 analysis by one high-ranking officer circulated among the top-most executives, fewer than 100 salaried workers out of well over 100,000 were dismissed annually for poor performance between 1977 and 1983. Even bad mistakes are sometimes rewarded with promotion. (The most vivid example in recent years involved Saturn's head of manufacturing, who left the job shortly after production snarled during the launch -- and was promoted to a vice president's post.) Says a business school professor and management consultant who has studied GM closely: ''The challenge for Stempel is to look people in the eye who may be his friends and tell them it's time to go home.''

-- Promote solutions, not programs. GM is a hothouse of creative problem- solving ideas, but little gets accomplished. Says an automotive industry consultant familiar with GM: ''They have programs that are going to solve every problem, but they never get solved. At the end of the day, the biggest problem Stempel has is to get the organization to actually do something rather than put out page after page describing new programs.''

-- Encourage candor. The 1988 memo observes: ''Our culture discourages frank and open debate. The rank and file of GM personnel perceive that management does not receive bad news well.'' Communications with suppliers and customers also suffer. After proclaiming a new ''partnership'' with outside parts makers in 1991, GM declared they should cut prices 7% over three years. ''They don't walk their talk,'' the CEO of one big supplier complained.

-- Rationalize the product line. The divisional marketing system created 70 years by Alfred P. Sloan has long since outlived its usefulness. Three GM divisions -- Pontiac, Oldsmobile, and Buick -- market cars that are similar in size and function, even if they no longer look so much alike, and generally sell for $14,000 to $28,000 with popular options. Oldsmobile sales slumped from one million in 1983 to slightly more than 400,000 in 1991, and its car market share has dropped from 10.9% to 5.2%. Since Olds buyers are almost identical to Buick's in age and income, it would make sense to retain the Olds brand but combine its operations with, say, Cadillac. To ease the transition for dealers, GM might have to offer financial incentives and make the changeover gradually.

-- Get tough with the union. Stempel agreed to a three-year UAW contract in 1990 that obligates GM to pay its 304,000 U.S. blue-collar workers even when they're on layoff. During slow periods when many plants are closed, GM pays 80,000 workers at least 95% of their regular wages and benefits in effect to stay home and watch TV. The contract ought to be renegotiated immediately to give GM more flexibility. In the early 1980s the union agreed to reopen the contract, but Stempel says they haven't heeded his requests this time. Penalizing the union for GM's mistakes is unfair, but so is life. If the company keeps on shrinking, there will be fewer jobs for union workers.

-- Improve efficiency. According to an internal GM analysis, 20 of its car assembly plants require an average of 35.7 worker-hours per vehicle, vs. 21.6 hours for Ford. Worse, GM needs 32.2 hours to 36.3 hours to put together such midsize models as the Chevy Lumina and Pontiac Grand Prix, while Ford can turn out the similarly sized Taurus and Mercury Sable with under 17.2 hours. Figuring labor at $31.50 an hour for wages and fringes, overall Ford has a $441 cost advantage on the factory floor alone.

-- Increase flexibility. In Arlington, Texas, the hot-selling Buick Roadmaster is built on the same line as the less popular Chevrolet Caprice Classic. The plant is now listed for possible closing. When it shut down temporarily in 1991 to keep Caprice inventories in check, supplies of the Roadmaster dried up. GM has to learn how to build more models in each plant and adjust production among the models more deftly to better balance output with demand.

-- Cut production engineering costs. GM lags behind Ford in the design and engineering of the tools and stamping dies needed to make new models. One Michigan-based manufacturing consultant says GM spends 20% to 25% more than Ford to get a factory ready for production -- a potential $100 million difference. GM, for instance, traditionally designs its dies to produce several times more volume than they actually do.

-- Speed new model development. Although GM is making progress at getting cars ! to market quicker, it remains slower than its major competitors. Harvard business school professor Kim Clark, who has studied product development worldwide, believes that it takes GM 42 to 48 months to complete a redesign. That is six to nine months better than in the mid-1980s, but still nine to 12 months slower than the Japanese. Says Clark: ''The average Japanese firm has almost double the development productivity and can get to market a year faster than the average U.S. firm.'' To see an example of a successful automotive turnaround, Stempel need look only as far as his own European operation. A big money loser as recently as 1986, it was revived after judicious investment in new factories and cars. GM's German and British makes, Opel and Vauxhall, sold 1.6 million cars in 1991, returned nearly $2 billion in profits, and challenged Volkswagen and Fiat for European sales leadership. Furman Selz analyst Maryann N. Keller says Opel's plant in Russelsheim may be the most productive on the Continent. Opel boss Louis Hughes, who took the trouble to learn to speak German on the job, has smoothed relations with Germany's contentious labor unions.

GM Europe is not a perfect paradigm, however. Being 4,000 miles from headquarters, it is relatively free from bureaucratic interference. It is only a third as large as North American operations, markets only two car brands, and lacks any feeling of invulnerability because it has already endured a life-threatening crisis. Perhaps most important, it competes in a market that traditionally yields higher profit margins and where Toyota, Nissan, and Honda have not been major forces until recently. Besides helping itself, GM could also do with some wisdom from Washington to help redirect the regulatory and legislative crosswinds that buffet the industry. Automakers and consumers alike need relief from drastic fuel-economy and clean-air legislation that will drive up costs by several thousand dollars a vehicle while producing little for the environment. Tax collectors ought to ensure that foreign automakers operating in the U.S. aren't using accounting tricks to avoid paying their fair share by, say, inflating on their books the price of parts imported from Japan. Loopholes that allow four-door trucks like sport utility vehicles and minivans to be labeled as cars when they land in the U.S., thus avoiding a 25% tariff, should be closed. Laws regulating local content of Japanese cars built in the U.S. ought to be redrawn so that it is % no longer possible, say, for a Toyota Camry engine assembled in Kentucky to count as 100% American even if three-quarters of the parts were imported from Japan (FORTUNE June 17). While the recession exposed and aggravated GM's problems, it didn't cause them. Poor assembly quality in the early 1980s, combined with indifferent design and inadequate product differentiation, eroded the value of GM's brands. GM fares particularly poorly with customers under 45, who grew up when things were going badly for GM and bought Japanese cars instead. As buyers grow older, General Motors worries that they may decide to move up from Hondas and Toyotas into Acuras and Lexuses, bypassing the Buicks and Cadillacs their parents drove. In trend-setting California, GM is picking up some lost ground, but it still trails the imports badly. It gets only 24.6% of sales, to 42% for the Japanese. No list of GM's self-inflicted wounds could overlook the series of misguided decisions, acquisitions, and investments the company made during the 1980s. Ford and Chrysler committed similar errors, but nobody spent as much as GM and got so little for it. For example, GM put $77 billion into new plants and equipment to reduce labor costs. Investing that much money sensibly is almost impossible, and GM paid more than it should and bought more than it needed. Some robots it acquired in the mid-1980s stand unused today. The highly automated equipment never delivered the promised savings because GM did not train workers properly to use it, and -- unlike Honda, say -- failed to design new models for easy robot assembly. In a costly effort to diversify, GM spent $2.5 billion for Ross Perot's EDS and $5.2 billion for Hughes Aircraft. EDS is paying off as a stand-alone business, though it hasn't created the synergies predicted when it wired GM into a single computer system. Hughes hasn't produced the off-the-shelf high technology GM wanted for its cars. A complicated scheme to reimburse the Hughes Medical Institute for its stock will eventually boost the total acquisition cost by $2.6 billion. Another 1980s legacy: By combining North American car operations in two groups -- Chevrolet-Pontiac-Canada and Buick-Oldsmobile-Cadillac -- in 1984, GM added another level of management to an already unwieldy bureaucracy without producing new efficiencies. Staff functions such as marketing, engineering, and strategic planning are now performed at three different levels in the company: corporate, group, and individual car division. The | reorganization stripped the car divisions of what remained of their autonomy, compounding brand-identity problems and customer confusion. It also created 18 months of chaos, delaying new-model programs and contributing to an incredible six-point loss of market share in a single year. And there is Saturn. Convinced that it couldn't build a profitable small car within its existing structure, GM created a new division, gave it $3.5 billion, and told it to do something revolutionary. Saturn has tried some new wrinkles in manufacturing techniques and labor agreements, advertising and distribution. While the car has been received enthusiastically by some buyers, it is priced so low (base model sticker: $8,195) that other Detroit auto executives say GM will never recover its investment. Some 18 months after startup, the plant in Spring Hill, Tennessee, is still debugging production. Saturn cost GM an estimated $800 million in 1991 and could run almost as deeply in the red in 1992. So GM now finds itself in the position of former Washington Redskins football coach George Allen, who, given an unlimited budget, proceeded to spend it all. Hugely profitable international operations as well as solid computer, aerospace, and finance subsidiaries have not prevented GM from piling up the biggest loss in corporate history over the past two years. The cupboard is now bare. GM will finally have to learn how to make do with less or try to raise new equity from what must by now be thoroughly disillusioned investors. Great men demand great times. Bob Stempel has an opportunity that was denied all of his predecessors: the crisis required to bring about revolutionary change. Stempel said in December that he is making fundamental alterations in the way GM does business. If he fails, it would be a tragedy for Detroit, for the Midwest, for the U.S. economy, and for a million or more GM workers and retirees -- not to mention all those potential customers for high-quality, made-in-America automobiles. But if he and the rest of GM management are successful, even his competitors will have reason to cheer the renewed vitality of an American corporate institution.

CHART: NOT AVAILABLE CREDIT: SOURCE: WARD'S AUTOMOTIVE REPORTS CAPTION: FALLING SALES. . .

CHART: NOT AVAILABLE CREDIT: SOURCE: JACOBS AUTOMOTIVE CAPTION: FALLING PRODUCTIVITY. . .

CHART: NOT AVAILABLE CREDIT: NO CREDIT CAPTION: FALLING PROFITS. . . | CHART: NOT AVAILABLE CREDIT: NO CREDIT CAPTION: AND FALLING SHARES

CHART: NOT AVAILABLE CREDIT: NO CREDIT CAPTION: GENERAL MOTORS