This article reviews various changes and developments that the General Motors’ (GM) management has brought after going bankrupt. According to a study, GM began to make plants more standardized and flexible years before Ford. Yet, Ford progressed faster because its centralized management was better able to make and implement decisions. GM needs to improve the speed of its decision-making. It plans to invest billions in United States and European manufacturing. It continues to expand in China, where it makes more vehicles than in the United States. GM’s plants focused on differentiating those things that gave its brands their character: sheet metal, power train, and interiors. Most importantly, GM began assembling every vehicle in every plant in the same sequence. The company has placed a large bet on the Volt, a plug-in electric car with an internal combustion engine. The combination means that the car is not limited to its 40-mile battery range. GM is making profits and succeeding now these days because it is building vehicles that customers want to buy, not vehicles they have to buy.
Little more than one year after bankruptcy, General Motors has changed its ways. It has been consistently posting profits after years of bleeding money. Earnings in the first quarter of the current fiscal year were $1.7 billion ($3.2 billion including onetime gains and charges). It was GM's most profitable quarter in 11 years.
The quarter followed a good 2010, GM's first full year after its reorganization. The company last year not only earned its first annual profit since 2004, but came within 28,000 vehicles of passing Toyota to reclaim its crown as the world's largest automaker. In 2009, GM made 800,000 fewer vehicles than Toyota.
Quite an improvement for a company that lost $31 billion in 2008 alone. When GM asked the government for a bailout in 2009, its prospects looked dismal. Many pundits argued that it was better to let GM go under. Clearly, they argued, GM could not compete against the powerful Japanese and European manufacturers.
It was hard to argue with the numbers. But analysts who looked under the hood saw a different picture. Manufacturing was GM's strength, not its weakness.
According to Ron Harbour, who has probably toured more automotive plants than almost any human being on the planet, “GM did not go bankrupt because of anything related to manufacturing. If there was one thing they did extremely well, it was that.”
Since 1982, his Harbour Report has measured the manufacturing productivity of global automakers. GM, he said, was as good as any of its rivals, even while losing $31 billion in 2008. It had made vast improvements in productivity and quality. Harbour starts plant inspections in the back, where vehicles with quality problems await fixes. “At GM, there used to be 1,000 vehicles waiting to be fixed. Now if you see 10 or 20, that's a lot,” he said.
GM's problem in 2008 was not manufacturing. It was not labor, either, since it had resolved its worst union problems by 2007. Its problem was legacy costs. During its reorganization, it shed those burdens—and Pontiac, Saturn, Hummer, and other unprofitable operations—so it could take full advantage of its manufacturing muscle.
Today, Harbour ranks GM among the top third of global vehicle manufacturers. In fact, GM has become so productive, it has begun in-sourcing, making previously outsourced components in its own plants. “When I walk through an airport, I see all these ads from consultants who want to help you outsource production,” Harbour said. “I laugh, because that's the opposite of what's going on in the auto industry today.”
And that matters. Once GM was the standard other manufacturers measured themselves against. Somewhere along the line, things went horribly wrong. Now, after decades of wrestling with manufacturing issues, GM has righted itself. What, if anything, does this say about the future of American manufacturing?
Cash, Culture, and Competition
Like a classic tragedy, GM's fall involved success and a fatal flaw, arrogance.
GM dominated the American market in the 1950s and 1960s. “In 1965, GM had half the U.S. market. Its net margin was an astonishing 50 percent. There was talk about breaking up the company. Its biggest problem was what to do with all the cash,” said Joseph Phillippi, a long-time industry analyst and founder of Auto Trends Consulting in Andover, N.J.
It was easy to give unions higher wages and more benefits. “The attitude was, ‘Do whatever it takes, but don’t stop the money machine,’ ” Phillippi said. Only later would the bill come due for pensions and medical benefits.
Meanwhile, the world was changing. The 1973-74 Arab oil embargo quadrupled the price of oil. Detroit's Big Three scrambled to boost fuel economy from their 13.3 mpg average. That led to poor designs, problematic engines, and transmissions that were anything but smooth.
Detroit's struggles gave Japanese exporters their opportunity. They built solid, zippy little cars that got great mileage. And their quality was outstanding.
Since the 1950s, Japanese manufacturers had applied a technique—statistical process control—that had fallen out of favor in the United States. As they reduced part and process variability, they operated their production lines faster yet produced higher-quality vehicles.
Toyota was the leader. It used workers to identify and help fix quality issues. Japanese automakers and their suppliers became so good at reducing variability they did not need to keep extensive inventories. The result was lean manufacturing, decades before anyone named it that.
Detroit responded with bluster. “I’ll never forget [former Chrysler CEO] Lee Iacocca saying, ‘Let them try to make cars here,’ ” Phillippi said. “He meant, let them try it with fat, lazy American workers. The Japanese came here and made him eat his own words.”
In 1986, Toyota opened its first American factory. Honda, Nissan, BMW, Mercedes-Benz, Hyundai, and others followed. Relentlessly, they ate into GM's market share. It fell from 48 percent during the 1960s to 45 percent in the 1970s, 41 percent in the 1980s, and 35 percent in the 1990s.
GM tried to change. In the 1980s, its chairman, Roger Smith, invested in robots and automation. He launched Saturn as “a different kind of car company,” but never got the buy-in it needed to thrive. He tried to reorganize GM's management structure, but the changes brought decision-making nearly to a stop.
GM's culture was just too embedded and Balkanized, according to William Holstein, author of Why GM Matters. “It was not managed as one company. Each division had its own plants and manufacturing processes,” he said. “There was huge resistance to creating one corporate style. The Chevy guys didn’t want to do it the Buick way, and the finance guys didn’t want to listen to either of them.”
According to Phillippi, “Nobody was ever held accountable, and nobody ever got fired. They used to call it ‘Generous Motors.’ If you screwed up, they’d put you into cold storage until you finished your career.”
In product development, people would rotate out of programs before the vehicle launched. “You’d have these huge cost overruns or issues over features,” Phillippi said. “The old manager would say, ‘I’m not there anymore.’ The new guy would say, ‘Hey, this is what I inherited.’ Guys were just punching their ticket.”
And then there was the GM arrogance. Art Schwartz, who rose to become GM's general director of labor relations, recalls one incident from 1986. Productivity data is closely held in Detroit, but Schwartz had found a way to estimate Ford's numbers from an obscure data source that showed factory employment by Congressional district.
“I compared that with GM's internal numbers, and it was clear we were not doing very well,” Schwartz said. “I started to give the presentation to our president, Jim McDonald.
He stopped me before I got very far and said, ‘The data must be wrong. GM could not be less productive than Ford.’ That was as far as it went.”
Climbing the Learning Curve
GM's entrenched culture ultimately did in Smith, but not before he launched two joint ventures, NUMMI with Toyota in California and CAMI with Suzuki in Canada. Those plants taught GM lean manufacturing.
GM was not a quick student. It assigned senior engineers and product developers to NUMMI. When they returned to Michigan, they tried to share their insights. “The response was always, ‘We don’t do that here, so go sit in the corner,’ ” Phillippi said.
Many just gave up. One group did not. Flying underneath GM's radar, they opened a lean Opel plant in Eisenach, Germany, in 1990. The plant was an instant success, boosting productivity and driving down costs. Then they did the same thing in Brazil, Holstein said.
It took years for the lessons to penetrate because after losing billions during the early 1990s, GM was again undergoing a corporate restructuring. This one lasted 10 years, eliminated 90 percent of core managers, and simplified decision making.
By the late 1990s, GM's new CEO and later chairman, Rick Wagoner, was ready to tackle manufacturing. Phillippi describes Wagoner as a numbers guy “who didn’t know jack about cars.” His office was dominated by basketball pictures from his alma mater, Duke University.
“But he knew what he didn’t know,” Phillippi said, “so he reached out to Bob Lutz and made him vice chairman in charge of product development and design.”
Lutz was everything Wagoner was not, a car guy with an MBA and a degree in production management. He had held top positions at several competitors, most notably Chrysler. His “Who says?” campaign hit the company like a whirlwind.
It was an extension of Lutz's hands-on management style. Lutz questioned design book dictates that added cost without a customer benefit. He asked why stamping dies were built for 1 million hits, and why alternator grease had to withstand -70 °F temperatures. He sided with designers against accountants. “It's not that budgets didn’t count, but he found ways of doing great design and really quality interiors,” Phillippi said.
GM also began redesigning its factories to make them more flexible. Its goal was to have fewer plants, but make them flexible enough to make several different vehicles and run them 24/7, Harbour said.
This was a radical break. Until then, GM started almost from scratch every time it launched a new vehicle. It bought or rebuilt stamping presses, redesigned internal parts, and reconfigured assembly lines. This brute force attack optimized production costs of a single vehicle, but married each plant to the vehicle alone. Switching between vehicles could close plants for six months.
Honda, on the other hand, could go from the end of one model to full volume production of a new vehicle within six to eight weeks, and only shut the plant down for a couple of weeks, said Brett Smith, co-director of the Center for Automotive Research's manufacturing group. This flexibility made it easy for Honda to switch production quickly to adjust to market conditions.
To match its Japanese competitors, GM had to standardize its processes, equipment, and as many internal components as possible. GM's sheet metal stamping presses, for example, all had different specifications. It replaced them with units whose gates all had the same dimensions and locations. This made it possible to share dies between plants, and shift production between factories.
GM sought to standardize its platforms and standardize components that consumers never noticed, such as door-locks. According to GM's vice president of manufacturing engineering, Eric Stevens, this let GM perfect those parts, drive down costs, and improve their interchangeability.
GM's plants focused on differentiating those things that gave its brands their character: sheet metal, power train, and interiors. “The Opel Astra in Europe, the Cruze in Korea, and the Cruze in North America are all different in performance and look, but they are all built off the same architecture and parts,” Stevens said.
Most importantly, GM began assembling every vehicle in every plant in the same sequence. Harbour explained why: “As the car moves along the conveyor, I put in the wiring, seats, and console. If I’m on station 104, I put in the steering wheel on all six models made in the plant. Each car's steering wheel may have a different design, but they all fit the same way and I use the same torque gun and fasteners. The only thing that changes is me selecting the right wheel.
“You can bring in any vehicle—one could be a Cadillac, the other a Chevy—and I don’t care because I build it the same way and it takes the same length of time. I get really good at putting on steering wheels, and the quality goes up. I know exactly what quality checks to do. I’m working at the same pace, I can be productive, high quality, and low cost.
“You can build six different cars each year, but give me a common design and common process, and it's not a new vehicle for me. You’re building a high level of product differentiation as a manufacturer, but I’m not. For me, I’m building one car over and over again,” Harbour said.
Having a single repeatable process did more than improve the efficiency of a single plant. It optimized the system, according to Michael Robinet, IHS Automotive's director of global production forecasting.
“If you put in a wire harness and you have a crimp, you can solve the problem on one line and you’ve fixed it for everyone,” Robinet said. “If every plant does it differently, they’ll all have different problems and can’t share solutions.” A standard process helps GM's plants develop best practices that do not change from vehicle to vehicle.
Standardization gives GM more flexibility. When demand for Chevy's Equinox crossover exceeded capacity at Ingersoll, Ontario, the facility expanded body welding capacity. It then shipped the body shells 120 miles east to its Oshawa plant to take advantage of an unused assembly line. Without standardized equipment and processes, such sharing would have been impossible.
“The GM of old would have built a second plant,” Harbour said. “It would have taken two or three years. And as soon as it was done, sales would have declined and they would have had to lay people off. They did this fast, with low capital investment. They’re going to make 350,000 or more Equinoxes this year, and every one of them is going to a dealer who is selling it at sticker price with no discounts.”
Before it could become a truly world-class manufacturer, GM had to deal with labor issues. Nearly everyone who studied GM has a labor story to tell. Most involve rigid job classifications and work rules. It was one of the reasons that Japanese producers could change a die in 10 or 20 minutes while at GM it might take 8 to 24 hours.
Others might point to “cost generators,” agreements that built ballooning costs into labor contracts. These included cost-of-living agreements that ratcheted up wages with inflation, job banks that paid workers for not working, and rules that required full-time tradesmen at each plant.
And then there were retiree benefits, especially insurance. Costs were so high that wags called GM an insurance company that made cars as a sideline. To critics, those deals were evidence that GM was profligate and management cared about nothing but its annual bonuses.
There was another side to this picture. Many agreements had been built into contracts when GM was still a money machine. Management gave unions what they wanted to keep the lines moving, and health care was not the expense it would later become. Once granted, benefits were hard to rescind.
When the United Auto Workers did grant concessions in 1992, management promptly toasted its success with huge bonuses. That left a legacy of distrust.
GM's attempts to reduce labor costs paradoxically made the problem worse. Its U.S. employment peaked at 618,000 workers in 1979. By 2008, the year before GM declared bankruptcy, it was down to 92,000. That meant every GM worker had to be productive enough to support nearly seven retirees.
Those cost drivers added thousands of dollars to the cost of each GM car. While Japanese plants in the U.S. now pay U.S. workers wages similar to those at GM, the workers are not unionized and have more flexible work rules. Because their plants are no more than 25 years old, they have very few retirees.
To close the cost gap with its competitors, GM shaved costs wherever it could. It put enormous pressure on suppliers to cut costs, but paid a price in quality. It gave accountants more control over development, and this often produced unimaginative designs and chintzy plastic interiors that turned off consumers.
Things came to a head after a crippling 1998 strike that cost GM $2 billion, Schwartz said. Now retired, he was involved in seven sets of national negotiations since 1987. “Management changed the whole labor staff over, and we tried to develop a new relationship with the UAW,” he said. “GM was going down the tubes, there were lots of quarters with losses. The UAW saw this and said, ‘If we are going into the sewer together, let's do something about it together.’ ”
GM and the UAW bargained continuously. In 2005, the UAW agreed to take over GM's health care obligations in exchange for a share of the company. In 2006, GM undertook a massive attrition program. In 2007, the UAW approved a two-tier wage that allowed GM to pay up to 20 percent of workers half the union wage.
“We thought the 2007 contract would get us where we wanted to go,” Schwartz said. “One could argue that had there been an active capital market in late 2008, GM would never have gone to the government.”
GM also began to clean up its supply chain. In the early 1990s, it had torn up purchasing contracts and began pressuring vendors for better deals. “If you beat the crap out of your suppliers, you don’t always get the best technology, and maybe they ship parts that don’t work,” Harbour said. “You could have the best manufacturing process, really excel, but if you use parts that go through the line but they’re not perfect, you may not have sustainable quality.” He said GM has revised its purchasing practices, but still wrestles with vendor quality.
GM needed to preserve capital. Despite quality issues, like many other American companies, it turned to outsourcing. Harbour likened the rush to outsourcing and then offshoring to “lemmings going over the cliff.”
When it comes to outsourcing, GM's Stevens picks his words more carefully. “In the past,” he said, “we probably outsourced for what we believed were good business case reasons. I think maybe 90 percent of those decisions were right, but things changed over time.”
Manufacturing flexibility, reduced workforce, and lower wage structure let GM rethink outsourcing. As a general rule, GM wants to make parts that impact quality and performance— things that matter to customers—in-house.
“Take sheet metal,” Stevens said. “A customer sees it, and good sheet metal looks better than bad. We can leverage engine performance to give us an advantage in fuel economy, mass, or output. These are things we would make ourselves. A wire harness doesn’t fit that conceptual framework. Quite frankly, customers can’t see, feel, or hear it.”
Now GM wants to put its considerable manufacturing muscle to work. It plans to invest billions in U.S. and European manufacturing. It continues to expand in China, where it makes more vehicles than in the United States.
Yet concerns remain. Small cars are one of them. Phillippi said that GM insiders told him the company makes only a few hundred dollars on each Cruze. Japanese competitors do significantly better. Given the trend towards more fuel-efficient cars, GM must show it can profit from smaller models.
The company has placed a large bet on the Volt, a plug-in electric car with an internal combustion engine. The combination means the car is not limited to its 40-mile battery range. The technology looks like a winner—users report getting hundreds of miles per gallon—but GM must get prices down from its $42,000 price tag if it wants to sell to the mass market.
The company has made giant strides, but has not yet caught up with its toughest competitors. Yet the jury is still out on how far GM has to go.
According to Smith, GM began to make plants more standardized and flexible years before Ford. Yet Ford progressed faster because its centralized management was better able to make and implement decisions. GM needs to improve the speed of its decision making. GM's best plants are now where their Japanese competitors were 15 years ago, he noted.
Harbour, on the other hand, finds GM's plants very consistent. He says GM is as good as or better than any Detroit automaker, and among the top third of automakers in the world. To close the gap, GM needs to move from lean manufacturing to a lean system that includes design engineering, purchasing, and other functions.
What does that look like? “In a good company, plant performance is not a metric for the manufacturing side of the house, but for the entire organization. In an effective organization, the guy in charge of design engineering is measured by the performance of the plant making that vehicle.
“Good companies like Toyota, Honda, and now Hyundai know how to design cars that require fewer people to assemble and are impossible to build wrong. They use fasteners that snap together in eight seconds versus eight fasteners that take one minute to screw in. It's impossible not to build their cars without high quality,” Harbour said.
Still, GM has come far. So what does its resurrection say about American manufacturing?
“It shows that an old line American manufacturing company can transform itself,” Holstein said. “The prevailing myth is that manufacturing has no future in the United States, that it's all service and consumer spending. Clearly, the percentage of the workforce in manufacturing has declined. But what still exists is world-class competitive. The assembly lines of the old days are fading. Taking its place are workers with multiple skills, who work as teams to help solve production problems,” he said.
Holstein's latest book, The Next American Economy, profiles U.S. companies like Caterpillar, John Deere, and Timken, that have done good jobs transforming their cost structure.
As always, the level of competition keeps rising. GM now competes in a market where everyone makes very good cars. Companies can only succeed if they are flexible enough to follow customer demand. Customers will pay more to get exactly the vehicle—color, engine, options—they want, Robinet explained. If automakers have the short, fast supply chains and flexible plants to deliver the goods, they won’t have to bribe customers with discounts to settle for less.
GM is doing that, Robinet said: “Its transaction values are going up because GM is building vehicles customers want to buy, not vehicles they have to buy. It's a mindset, not just in one executive, but throughout the whole company. They are changing the way they do business.”