COLUMN ONE : Detroit’s Struggle to Shift Gears : U.S. auto makers are trying to trade in their moribund past for an innovative future. The result has been a mixture of sparkling successes and dismal failures.

On the assembly line for the Saturn automobile, the instrument panel slips smoothly into place in a matter of seconds. But on the Chevrolet Lumina, workers have to wrestle it into position like some contraption they have never seen before.

Chrysler Corp. talks loftily of modeling its car dealership culture on Disneyland and Neiman Marcus. But in some showrooms, salespeople still browbeat customers, treat women like dopes and manipulate dollar figures in time-honored fashion.

The 1992 Cadillac Seville captivates car enthusiasts as a classic, world-beating new Detroit design with top-rate quality. But an army of trouble-shooters swarms over the assembly line where it is built, trying to figure out why half of the cars need to be repaired before they leave the factory.

Ford Motor Co. revamps its U.S. assembly plants into the nation’s most efficient, outranking even the new ones built by the Japanese. But when General Motors Corp. builds nine new assembly plants, most are worse than the ones they replaced.

Given such two-headedness, it is no wonder that the outside directors of GM--the struggling giant that sits astride the best and worst of Detroit--ordered a historic management bloodletting this spring. And no wonder the U.S. auto industry is still struggling up off the floor after its second encounter with mortality in a decade.

A return to slim profitability in the just-finished first quarter and recent market successes against the Japanese point to a few years of recovery for U.S. auto makers. But the last time Detroit got some breathing room, in the mid-1980s, the opportunity was squandered by all but Ford.

One might ask, as the auto industry pronounces its importance to the nation’s well-being and asks for a reordering of national priorities, is Detroit cleaning up its act or not?

It depends on which Detroit has the upper hand at the moment.

After a decade of turmoil that promises to continue through the 1990s, the industry’s critics and supporters alike view today’s Detroit as at once smart and dumb, innovative and fossilized, venturesome and wimpy.

The “industry of industries,” as management guru Peter Drucker once called the U.S. automobile makers, is struggling to remake itself. And it has made believers of some who have seen the changes being wrought on the factory floor, in the design studios and in the engineering laboratories.

In addition to closing factories and blaming various governments for many of its woes, Detroit has spent the last decade reorganizing, sitting in quality circles, unplugging time clocks, sweeping away tiers of managers, revamping plants, spending $70 billion, redesigning its cars and engines, closing some of its executive dining rooms and inviting outside ideas--especially Japanese innovations--into the musty tent.

But for all the self-proclamations of a rebirth, the gains have been spotty and fragile, limited largely to one-time projects. Capable of pulling off impressive success stories in everything from getting along with each other to developing new transmissions, the industry has been hard-pressed to repeat those triumphs.

“It’s like a good cop-bad cop personality,” said Alan M. Webber, editorial director of the Harvard Business Review and co-author of “Changing Alliances,” a 1987 book on the auto industry. “You look at all this great stuff going on, but then you look at the nature of the companies and you get the feeling it’s fundamentally the same beast.”

Despite the well-documented quality gains, there are peaks and valleys from car to car, and significant gaps remain between the average U.S. and Japanese vehicle. For every factory with hand-holding workers and managers, there’s another warped by hostility and distrust. Some dealers shoot straight with customers, others hustle them.

Kim B. Clark, a Harvard Business School professor who studied the world auto industry’s systems for developing new cars for a 1990 book, “Product Development Performance,” said: “Certainly there’s an old Detroit and an energized Detroit. In the 1980s, a lot of energy and leadership was expended on improving productivity and quality. Now the auto executives have realized that it is much harder than they ever imagined. They need a whole new round of renewed energy and leadership. If that leadership is forthcoming, there is every reason to be optimistic.”

It is dangerous to generalize too much about the three remaining U.S. auto makers. Ford was first to impose fundamental change, and now Chrysler is credited with a new entrepreneurial bent. General Motors, the last and biggest ship to get in trouble, has just noticed that water is crashing over the deck.

But each of the Big Three has the same prerequisite to righting itself: All must escape the beast of Detroit’s past. Nearly all of the U.S. auto industry’s gains have occurred in free-standing operations or in costly “skunk works,” projects set up at remote sites to shield them from management tinkerers and entrenched constituencies.

As a result, the auto industry’s triumphs have sometimes been idiosyncratic--even resented by others in the same company.

An early example of such dedicated projects was Ford’s so-called “Team Taurus,” which brought out the milestone family sedan that saved the company’s U.S. operations in the mid-1980s. And there is the ultimate skunk works--GM’s Saturn project in the Tennessee hills, which produced the first true domestic small car to win the blessing of Consumer Reports magazine.

But both projects were budget-busters: The Taurus took six years and $3 billion; the Saturn, seven years and $3.5 billion. And GM and Ford have stumbled in efforts such as the current Ford Thunderbird, an overweight, over-budget vehicle said to have derailed some careers, and the GM-10 line of family sedans, said to be too late, costly and difficult to manufacture to generate anything but trouble for the company.

The way David E. Davis Jr. sees it, Detroit can walk or chew gum, but it cannot seem to do both at once.

Davis, editor of Automobile magazine, says the U.S. auto industry was working so hard on improving its woeful quality and productivity problems in the 1980s that it neglected to bring out memorable cars, the exceptions being the Taurus and Chrysler’s minivans. Meanwhile, the shifting relationship between the dollar and the Japanese yen made efficiency less critical--and Japan bolstered its lead with several expensive, cutting-edge vehicles like the Lexus.

“This is the Detroit tradition,” Davis said. “They take a single focus, take problems one at a time. So they were solving quality problems in cars that were a generation old.”

Still, the intentions seem right. GM spent hundreds of millions of dollars to salvage the 1981-vintage Chevrolet Celebrity, Oldsmobile Ciera, Buick Century and Pontiac 6000, whose quality problems by 1985 were a major embarrassment. By 1991, J. D. Power & Associates ranked them as the highest-quality mid-sized cars on the market, even ahead of Japanese competitors.

“In the past, they would have said: ‘Well, we’ll just have to live with it’,” said Sandy Monroe, a Troy, Mich., expert in a fast-growing engineering discipline that designs--or redesigns--products for ease of assembly. “Now they’re into a cycle of continual improvement.”

The two faces of Detroit permeate the business, sending mixed signals throughout the employee ranks. For example, in 1988, GM dreamed up a merit system requiring that 10% of all white-collar workers be declared unsatisfactory and 10% superior. Simultaneously, its low-profile powertrain division was pioneering a radically different approach that assumes that most problems stem from top management. The “curve” system was dropped after a near mutiny, and now the powertrain plan is under study companywide.

For all Detroit’s experimentation, the toll of the 1991 recession, a staggering loss of $7.6 billion for the Big Three last year--nearly double the industry loss of 1980, its previous low-water mark--stunned auto executives and made brutally clear that none of the domestic companies had accomplished as much in the 1980s as claimed.

In fact, not until 1990 was the epochal nature of its task--a burden actually shared across the U.S. and European industrial economies--suggested in writing.

That occurred when researchers from the Massachusetts Institute of Technology declared that Japan had done nothing less than usher in a new era of manufacturing, effectively leapfrogging the mass-production system, created by Henry Ford, that has dominated Western manufacturing throughout the 20th Century.

Their five-year study, published under the title “The Machine That Changed the World,” spelled out in persuasive detail how the Japanese have done more with less. The work has become a reference point from which change in the auto world is being measured.

The MIT experts dubbed Japan’s breakthrough a system of “lean production” developed chiefly by Toyota Motor Corp. in the 1950s and 1960s out of the rubble of World War II. The approach explains much of Japan’s remarkable economic success, the study says. But Detroit, scrambling to adapt, has found that it is quite another thing to impose such radical change on an industrial system that has been in place, mostly unchanged, for 80 years.

The flexibility, teamwork and egalitarianism that are hallmarks of “lean production” run counter to the specialization, rigidity and top-down traditions that run so deeply through many U.S. industrial corporations.

By this theory, Detroit operates like a bowling team: Each player is responsible for his own score, and at the end the teammates’ scores are added up. Japan, unfortunately, is playing basketball.

The heart of Detroit’s problem is a routine--traceable to the early days of Henry Ford and legendary GM Chairman Alfred P. Sloan, but now being painstakingly dismantled--in which designers design a car and then “throw it over the wall” to the engineers.

They would do their best to engineer what the designers had tossed them, then throw it over the wall to the manufacturing experts, who would decide how best to build the car and then tell the assembly-line workers, “It’s all yours.” And then the marketers would see if they could sell it.

In the same way, the auto makers would dictate their needs to the hundreds or thousands of companies making the 10,000 parts that go into each car, then award short-term contracts to the lowest bidders. If everything got done on time and the instrument panel matched up with the front-door trim, it was a miracle.

“My predecessor used to pride himself on never having entered the office of the vice president of manufacturing,” said Thomas Stallkamp, Chrysler’s vice president for purchasing and manager of its big-car programs.

With the idea of creating sister divisions that would thrive by competing against each other as well as other companies, Detroit built redundant worlds over 80 years, with overlapping engines, overlapping engineers, overlapping factories and overlapping press agents.

Richard Donnelly, head of GM’s powertrain division, said: “We had excess capacity long before we lost market share.”

This compartmentalization engulfed the companies and shaped rules and attitudes, forging constituencies that were threatened by outside ideas and spawning “a whole generation of executives who need to save face,” in the words of an ex-GM manager.

“Resistance?” laughed Monroe, whose job often requires telling auto companies how they screwed up. “I’ve had people take swings at me. I’ve been in a small room with 35 people yelling and screaming at me and kicking doors and wanting to take me apart. These are managers who couldn’t handle what was happening.”

The old ways led to such extreme specialization that an engineer could base a career, for example, on designing door locks. Lo and behold, Ford eventually found itself with 47 types of door locks for its cars. As the new day has dawned at Ford, the number of door locks has been slashed to 12.

Such nitty-gritty change comes hard, but the payoffs are mind-boggling, says John E. Decker, manager of program planning for Ford’s body and assembly operations.

Ford thought it would need a new body shop at a Michigan plant slated to produce a 1996 vehicle. But by making its manufacturing engineers privy to the clay models of the car, they have already figured out how to arrange tooling to fit the existing body shop. If the problem had been dealt with later, as usual, panicky managers would have thrown up a new $35-million shop, Decker said.

This embodies Ford’s latest initiative to get $1 worth of investment for every 80 cents--an example of “lean” thinking that resulted from knocking down the walls that have separated people by their roles.

GM, by contrast, is just now getting around to its “creative crisis,” as the MIT researchers call the necessary reawakening, and has the furthest to go. Earlier attempts to revamp GM, lacking the financial urgency of today’s predicament, repeatedly fell short or proved to be wrong-headed, notably a 1984 restructuring that created more bureaucracy instead of less.

GM has announced that it will eliminate 74,000 jobs by 1995. But even at that, “they would have to take out another 50,000 to be where we are now” in terms of productivity, said Lloyd Hansen, assistant controller at Ford’s domestic car and truck unit.

Ford’s great achievement has been in making better cars and trucks and doing it with fewer workers. Ford has now matched the average Japanese car in quality, various studies show. And according to rankings by industry consultant James Harbour, Ford boasts the four most productive U.S. assembly plants, ahead of the Japanese transplant facilities. (GM struggles with the 11 worst, including its Van Nuys, Calif., factory, which will be shut down this summer.)

Yet some complain that Ford has not yet pushed the product-development lessons of Taurus into the company’s mainstream. The company says it has four skunk works under way at leased sites now, notably the 1995 Mustang project. But Ford’s bid to match Toyota’s phenomenally successful Lexus, the Mark VIII due out this fall, was done the conventional way, to some internal chagrin.

“Ford hasn’t been able to institutionalize some of these things,” said David E. Cole, head of a University of Michigan auto industry think tank.

Still, Clark estimates that instead of taking an average of 14 months longer than Japanese firms to develop a new vehicle, Detroit now only takes nine to 12 months longer. With big improvements in the works and financial woes in Japan triggering auto spending cutbacks, the gap could vanish.

Chrysler has imposed an Americanized version of the “lean” approach across its entire system. After a long study of Honda Motor Co., the smallest U.S. auto firm is using so-called platform teams, which assign engineers and others full-time to the development of one vehicle, to bring out five new lines of cars, trucks and vans over the next five years.

At any given time, designers; body, chassis, powertrain and manufacturing engineers; accountants and plant foremen work together, and their tasks proceed simultaneously instead of sequentially--key elements of the “lean” approach.

The first Chrysler vehicles developed fully by a team, the LH mid-sized sedans due out this fall, took 39 months versus Japan’s average of 36 months. It was blinding speed for Detroit and a relative bargain at $1.6 billion.

The big question is whether the U.S. auto industry’s probable recovery later this year undermines its tenuous reforms. Detroit doesn’t seem to act unless peering over the abyss, and even then it’s been difficult enough to change attitudes, industry analysts agree.

A vice president at a GM subsidiary said: “There are still people in the corporation who think that all we need is a little uptick in the market so that we’ll be making money again and we’ll be OK.”