Clean machine
Sep 2nd 2004
From The Economist print edition

The car of the future will be bristling with electronics, and may be run by electricity

THE two big changes in cars themselves over the next ten years will be in electronics and engines. Electronics are taking over more and more of a car's controls, and are now making it possible to connect cars as well as drivers with the outside world. It is only a matter of time before the industry switches from 12-volt batteries to 42-volt ones to meet the extra power demands of all this equipment.

At the same time demands for yet more reductions in tailpipe emissions of carbon dioxide and noxious gases such as carbon monoxide and nitrogen oxides are causing changes under the bonnet. With Hollywood stars such as Harrison Ford and Cameron Diaz thinking it cool to arrive at Oscar ceremonies in their little hybrid electric Toyotas, alternative fuels are firmly on the consumer as well as the regulatory agenda. Toyota has handed out free Priuses to the glitterati to get attention.

But for the moment the action is electronic rather than electrical. Telematics, which connects computers and other electronic devices by radio, has been under discussion for a long time, but has been slower than predicted to take off. At the Detroit motor show in 2000, with the dotcom bubble still fresh in people's minds, pride of place on Ford's stand went to a very plain, cube-shaped car called the 24:7. It looked like a small child's drawing of a car, a box with a wheel at each corner. The dashboard had a computer screen, and the driver was meant to log on to “his” or “her” car. The vehicle could be programmed to perform differently for different users, and its software would adapt the communications devices to each user's needs. It was connected to the internet, so if you were stuck in a traffic jam you could use the time to check your e-mails or browse the web.

It was wired, weird and typical of the times. Now carmakers have gone back to talking about design, speed, handling, ride and comfort. Yet electronics in cars are becoming more important in all kinds of ways. For example, electronic circuits determine the optimum fuel/air mixture, ignition sequence and valve timing on a modern engine 6,000 times a minute. Without that kind of technology, there would have been none of the dramatic improvements in cars' fuel economy over the past 20 years.

Electronic devices also decide within microseconds of a crash how to inflate the airbags. Modern cars have about 80 such electronic control units, in effect tiny computers, with software running to 1m lines of code; electronics are replacing mechanical parts as the nervous system of the modern car. Its ride and handling characteristics can be changed simply by altering the software settings for the steering, suspension and brakes.

The use of telematics takes all this a giant step forward. Telematics connects cars with the outside world and with each other. Early experiments such as the automated highway in California on which electronically linked cars were whizzing along while the drivers dozed off or read the paper will probably never lead anywhere. But the growing use of on-board electronic devices monitoring the car's mechanical systems, and the arrival of digital radio (which is better at transmitting electronic signals), look set to produce real benefits. Unbeknown to drivers, systems built into cars already record, for instance, when and where airbags inflate; this information is used after a crash to find out exactly what happened.

Some analysts say that within three years there will be telematics devices in two out of every three new vehicles in America, compared with only one in 20 in 2000. Over the same period the number of users of satellite navigation systems is expected to rise from 1m to 30m.

The age of telematics

Currently the industry is debating three main telematics applications for cars. The so-called front-seat market covers safety and security devices, including traffic reports and route-mapping via satellite navigation displays, and automatic connection to emergency services in the event of an accident. That means the paramedics will know how to find a crashed vehicle even if the occupants are in no state to tell them where to look.

The “back-seat” market promises on-demand digital movies and interactive games. The third application is the transmission of technical information from computers monitoring the car's mechanical parts direct to the manufacturer or dealer. This makes it possible to diagnose faults remotely and even predict the impending failure of parts of the car. Some enthusiasts also forecast voice-activated internet connection without a mouse or keyboard, but tightening rules against driver distractions may stop this before it gets going. Another obstacle to the spread of on-board devices may be the availability of hand-held ones, in the same way that carphones have largely given way to mobile phones.

One sector where telematics is already making a huge impact is commercial traffic. For all sorts of security, safety and operational reasons, freight companies find it invaluable to know where their lorries are and what they are doing at any one time. On-board GPS systems make this possible. In Britain, such a system is being introduced as part of a road-pricing scheme for lorries on some motorways.

Telematics could also help to make cars safer. The American government's road-safety body has calculated that lane drift on motorways is responsible for 43% of all fatal accidents on highways there. Telematics devices now being installed in luxury models tell the driver when he strays sideways or gets too close to another vehicle. If these were widely adopted, they could help to make driving much safer. Some cars now have a form of cruise control that stops the vehicle colliding with another. Fiat has even demonstrated a “driverless” car that can spot and steer round obstacles on a test track.

The next phase will involve electronics taking over from mechanical controls on vehicles. A combination of electronics and pneumatic power could be used to open and shut the valves on each cylinder of an engine, eliminating the complex mechanical valve drives of today. Steering and braking could also be controlled electronically and operated by tiny electric motors, in theory at least, so cutting the need for heavy metal brakes and pedals. Transmissions could become entirely electronic. After all, aircraft have been “flying by wire” for 20 years, so it should not prove impossible to convert mere motor cars.

But for the moment, such gee-whizz electronics applications could be marred by a problem familiar to every PC user. Luxury-car owners can find that their fancy new Mercedes or BMW suddenly goes awry because of a software glitch. And there are now so many electronic devices built into cars that manufacturers are having to install sophisticated electronics networks to integrate them. An executive at one parts company says that electronics in a car are seven times more likely to fail than mechanical parts. No wonder the manufacturers are now working together to draw up common software standards. According to an apocryphal industry story, Microsoft's Bill Gates once quipped that if GM had kept up with technology as the computer industry had, people would be driving around in $25 cars. To which one GM wag replied, “Would you want your car to crash twice a day?”

Hybrid solution

After 40 years when environmentalists' main grouse about cars was the damage their emissions of nitrogen oxides caused to local air quality, the emphasis has now switched to carbon dioxide, which is thought to contribute heavily to global warming. A few years ago the European Commission persuaded carmakers in Europe to agree to a voluntary deal to cut overall emissions across their car fleet by 25% by 2008, or face the imposition of strict emission rules for specific models. This summer California drafted regulations requiring car manufacturers to reduce emissions of carbon dioxide by 30%, starting in 2009.

The manufacturers are likely to challenge this in court. They point out that carbon dioxide itself is not toxic, and a state has no right to apply what are in effect fuel-economy rules, the prerogative of federal government. But California's governor, Arnold Schwarzenegger, has already expressed his support for the legislation behind the carbon-dioxide rules.

California already has regulations that will require at least 10% of a car company's fleet to be “zero emissions” vehicles (ZEVs) from the start of next year if the company wants to continue to sell in the state. California accounts for a tenth of America's car market, so its state government's environmental policies are taken seriously. After some concessions on the timing of the ZEV ruling, Ford and Chrysler dropped court actions to block the rules, and now every manufacturer is scurrying to get a ZEV ready in time.

The only conceivable ZEV is a car powered by an electric motor that runs on a fuel cell. A fuel cell combines hydrogen with oxygen from the air to produce water. The process generates an electrical current strong enough to power a car. Honda was the first to get a ZEV ready for the market, but nearly all the other car companies are working on their own versions. GM recently drove a fuel-cell minivan from the Arctic Circle to Portugal to test the robustness of the system. But fuel cells are still about ten times more expensive to make than internal-combustion engines, and cost, technical and safety problems remain to be resolved in setting up a distribution system to get hydrogen into the cars.

Meanwhile most car companies are introducing hybrid-electric cars, emulating Toyota's Prius, of which 200,000 have already been sold. Hybrids hook a small petrol or diesel engine to an electric motor and a storage battery. The electric motor runs the car in slow and stop-start motoring and the conventional engine cuts in on the open road. The battery also provides extra power for acceleration. It is kept charged by the conventional engine and by the energy generated during braking.

Quantum leap

In theory, once all the bugs have been sorted out fuel cells should deliver better total fuel economy than any existing engines. Allowing for the resources needed to extract hydrogen from hydrocarbon, oil, coal or gas, the fuel cell has an efficiency of 30%. That is twice as good as the internal-combustion engine, but only five percentage points better than a diesel hybrid.

But once hydrogen is being produced from biomass or extracted from underground coal or made from water, using nuclear or renewable electricity, the way will be open for a huge reduction in carbon emissions through the whole system. Experts such as Larry Burns, head of research at GM, reckon that only such a full-hearted leap will allow the world to cope with the mass motorisation that will one day come to China and India.

In the meantime, given the questions still to be resolved on fuel cells, hybrids are widely seen as the best way of cutting carbon dioxide and other emissions for the next 20 years. Another interim solution, according to Gerhard Schmidt, Ford's head of research, could be burning hydrogen in internal-combustion engines; this is less efficient than using it in fuel cells, but would ease the transition to hydrogen fuel cells in due course, allowing time to build a hydrogen distribution system. One way or another, electricity is changing cars.

Driving change
Sep 2nd 2004
From The Economist print edition

Technological change could help reform the car industry

IT IS easy to see why the motor car and the motor industry became symbols of 20th-century consumer capitalism. Cars are an expression of personal freedom, to go where you want, when you want, without having to follow the herd or abide by someone else's timetable. Congestion may have narrowed that freedom, but it remains part of the lure of the automobile. The French philosopher Roland Barthes, writing about the 1950s Citroën DS, compared cars to great Gothic cathedrals. “I mean the supreme creation of an era, conceived with passion by unknown artists, and consumed in image if not in usage by a whole population which appropriates them as a purely magical object.”

The magic faded a little when the oil-price rises and safety regulations of the 1970s turned most cars into drab front-wheel-drive boxes that all looked much the same. But for the past ten years the design flair of older generations has been creeping back. GM's new Cadillacs, Nissan's new 350Z car (a retro nod to a 1960s model), and Renault's bustle-backed Mégane, Scenic and Vel Satis are recent examples. Even BMW, long famous for severe, classic straight lines, has recently gone curvy (too curvy for some of its traditional customers). Cars are becoming sexy again.

The car may be the ultimate consumer good in a consumer age, but it is also increasingly a product of fashion as well as engineering, with design once again becoming crucial to brands. And brands have become far more important in recent years, in cars as in other consumer businesses. Appropriately enough for a nation famous for fashion, France, with its Renault and Peugeot brands, was among the first to put the designers back in the driving seat in the mid-1990s. Now all manufacturers are seeking a family look across their whole range so that people can identify their cars at 50 paces. Successful car designers appear at motor shows to introduce their latest models, like couturiers showing off their new collection.

But why is this pillar of capitalism not more successful at consistently delivering the ultimate reward of capitalist enterprises, fat profits? A recent analysis of the financial performance of all the leading car companies by Goldman Sachs, an investment bank, came to a sobering conclusion: out of the world's top 17 car companies, only half were earning more than the cost of their capital (see chart 8). The value creators in Europe were Porsche, the Mercedes bit of DaimlerChrysler, BMW and Peugeot. In Asia, Toyota, Nissan, Honda, Hyundai and Kia made the cut. But America's Big Three, GM, Ford and Chrysler, were all in the value-destruction group, along with Renault, Fiat, Mazda, Mitsubishi and VW.

The overriding reason for the lack of profits is excess capacity in mature markets. Under normal market conditions such a surplus would be competed away, but that rarely happens with cars. Manufacturers hold on to their capacity, grimly hoping that the next model will win business away from the competition and fill their coffers. They provide jobs, pensions and health care for their workers, who in turn buy a quarter of the cars produced. No wonder GM has been called a workers' collective, and Detroit a mini-Sweden.

Imperfect competition

One plausible explanation of this aberrant behaviour lies in the industry's ownership structure. Families play a big part in Ford, Fiat, BMW and Peugeot. Where there is no family, it may be corporate giants propping up the industry. Without the approval of Deutsche Bank, Jürgen Schrempp would not have survived the decline in DaimlerChrysler's value since the merger. More importantly, there is often government involvement of one kind or another. Renault and Volkswagen, for instance, remain in partial public ownership, whereas the British and French governments throw money at their Japanese-owned car factories at the mere suggestion that a new model might be made elsewhere. And America's states have been offering incentives to lure foreign car companies ever since Honda opened the first American transplant factory nearly 22 years ago.

GM's boss, Rick Wagoner, regularly complains that the Japanese government is providing indirect aid to the country's car companies by holding down the yen, thereby lowering their costs of producing export models and parts for their American assembly plants. GM economists reckon that this currency-market intervention amounts to a subsidy of $2,500 for every Toyota sold in America, and $12,500 for an upmarket Lexus. American manufacturers, and not just those in the car industry, maintain that the Japanese do this solely to boost their exporters, but they are overstating their case: the main purpose of the policy has been to head off deflation and revive the domestic economy.

Allowing for all these powerful outside influences on the industry, most car firms will try to muddle through. There is little enthusiasm for further consolidation: the example of Daimler's troubles with Chrysler is enough to put anyone off. Fiat is widely seen as the company closest to the brink, but the Italian government is unlikely to let it die outright. As McKinsey's Mr Mercer puts it, “No politician wants to stand in front of a closed car factory.” Over a longer period, though, once-mighty companies can wither away. In the mid-1970s British Leyland was the world's third-biggest car firm, but now it has all but gone, its brighter brands long sold off.

The only sort of consolidation that seems likely is the defensive sort that might come from distressed companies selling off assets. In their book, Messrs Maxton and Wormald suggest that a troubled Ford might be forced to sell Land Rover, Jaguar or Volvo. But these fine brands make hardly any money: among Ford's premium brands, only Volvo has been trading profitably in recent years, and Jaguar is reckoned to have swallowed $6 billion of Ford's money in the 15 years since it was acquired by the American giant. Even after much management attention, its current business plan is still not working, with sales around 120,000 a year instead of the hoped-for 200,000.

If GM felt financially desperate, it could generate some cash by selling its stakes in Suzuki or Subaru. But that would be a complete reversal of the firm's strategy, of building a network of global alliances to enter distant markets and share the development costs of small cars specially designed for them.

The car industry's long-standing obsession with scale arose from the needs of its manufacturing processes. Garel Rhys, director of the Centre for Automotive Industry Research at Cardiff University, has calculated that economies of scale reach their peak at 250,000 cars a year in an assembly plant, although for the body panels the figure could be as high as 2m. This goes back to the invention of Buddism in the 1920s. Budd's pressed-steel monocoque body shells could be made only by huge, expensive press tools that needed vast production runs to repay the investment. If cars were still built with a chassis and a separate body, like Henry Ford's Model T, the industry might look very different.

Some of the alternative business models currently being touted would, in effect, attempt another revolution, with companies outsourcing more and more of the car. Some even envisage the return of the steel frame with pre-painted body panels hung on it, as used for the runabouts produced by Mercedes's Smart division. Fiat and Audi, too, have turned to alternative manufacturing methods, using what is known as space-frame construction rather than pressed panels for giving the vehicle strength. The body panels are riveted on to the frames in a way that Henry Ford would recognise immediately. This method is particularly attractive for production runs of less than 100,000 a year, obviating the need for large numbers of huge presses to stamp out expensive floorpans. Space-frame technology can also be easily adapted to make cars wider, longer or taller, which brings other benefits.

If consumers are demanding an ever wider choice of vehicles, it follows in a mature market that production runs have to get smaller. Car companies are already reconciling themselves to this trend. That increases the appeal of more flexible manufacturing methods. Many car companies are quietly developing expertise in this new way of making cars but are reluctant to talk about it. GM caused anxiety among its unionised workforce a few years ago when it talked publicly about its Project Yellowstone, to adapt production by assembling cars from pre-prepared modules made by outside parts suppliers.

Changes to manufacturing methods could also arise from new technology being incorporated in cars, particularly as less polluting alternatives to petrol and diesel engines come along. Over the next 20 years, the market share of petrol and diesel hybrids and fuel-cell electric vehicles will probably rise to about 10%. Even an arch-petrol-head like Mr Lutz has been convinced by the progress made with these technologies in the past two years. “But I don't say they'll be fun,” he quips, “Better keep a motorbike on the side.” Much of the heavy cast metal in cars will also go as drive-by-wire becomes standard and heavy metal modules such as steering and brakes are jettisoned.

As the industry's products begin to change, so will the way they are made. In time, there will be less need for huge, capital-intensive factories, so the barriers to entry will come down. Start-up companies could take business away from established traditional manufacturers. Low-cost carmakers could swoop in, rather as low-cost airlines have done in aviation.

The conventional wisdom in the car industry is that it will continue in much the same shape as at present, with today's six big companies dominating global markets, even though half of them fail to earn a wealth-creating return on capital. Car companies, says McKinsey's Mr Mercer, are like steel mills and airlines, often seeming to teeter on the edge of bankruptcy but somehow managing to keep going. But such long-established industries are already being shaken up by newcomers.

In carmaking, those newcomers could be parts manufacturers moving into assembly, or outside companies taking advantage of the shift to new technology. As cars are increasingly powered by fuel-cell electric engines and fitted with drive-by-wire electronics, new entrants with skills in these areas might start to offer their own models. And even before that happens in a decade or two, new competition of a more conventional kind is bound to come from Chinese manufacturers selling low-price cars in Asia and America.

The old car firms must reinvent themselves to seek profit, not just market share. Otherwise new, nimbler competitors will take advantage of technological change to do the job for them.

Sources
Sep 2nd 2004
From The Economist print edition

"The Second Century: Reconnecting Customers and Value Chain Through Build-to-Order", by Matthias Holweg and Frits K. Pil, MIT Press, 2004.

"Time for a Model Change: Re-engineering the Global Automotive Industry", by Graeme Maxton and John Wormald, Cambridge University Press, October 2004.

"The End of Detroit: How the Big Three Lost Their Grip on the American Car Market", by Micheline Maynard, Doubleday, 2003.

"Comeback: The Fall and Rise of the American Automobile Industry", by Paul Ingrassia and Joseph B. White, Simon & Shuster, 1994.

"The Automotive Industry: A Guide", by P.E. Wells and P Nieuwenhuis, Cardiff University Centre for Automotive Industry Research and British Telecommunications, 2001.

---

Detroit's new quality gap

It isn’t enough to avoid defective products. Consumers now want cars to be ’fun.’

Niladri Ganguli, T. V. Kumaresh, and Aurobind Satpathy

The McKinsey Quarterly, 2003 Number 1

For two decades, US automakers have struggled to close a vast quality gap with their rivals. To be sure, Detroit has improved considerably: US auto defect rates fell by nearly 80 percent from 1980 to 2001, according to the 2001 Consumer Reports auto survey. But the Big Three still lag behind some foreign competitors, primarily the Japanese, by most quality measures.

This lingering gap prefigures a continuing slide in market share, but now another dimension may be of greater concern to the Big Three: over the past few years, consumers have changed their ideas about what defines quality—a shift that is making the uphill climb for US automakers even more steep. Though buying decisions once emphasized quality defined as minimizing the number of defects in a car, consumers are now focusing more on maximizing the appeal of a handful of core product attributes—for example, whether the car is fun to drive, well designed, or stylish.1 This shift is due in part to Detroit’s very success in bringing the product defect rates of new cars to levels so low that a defect-free vehicle is now largely taken for granted. As a result, quality today is not only about getting fewer things wrong, as in the past, but also about getting more things right: in particular, better performance, greater comfort, tangible amenities, and a sense of aesthetics.

The stakes in the new quality battle are enormous, and Detroit is already at a disadvantage. Since 1996, when most US and some Asian and European automakers started offering sales incentives in the form of discounts and rebates, US companies in 2002 have increased their incentives to a staggering 14 percent of sale prices—twice as high as those offered by foreign competitors, on average. Even so, the US market share of the Big Three has slipped over this period, by a combined 1.5 percent a year

Chart: The big slide

How does a lack of product appeal translate into such losses? For one thing, the need to slather on increasingly large incentives to offset the lower resale value of US cars reflects perceived problems with their long-term quality and durability. Consider the story of the Toyota Corolla and the Geo Prizm, which, save for their nameplates, were virtually identical, made on the same assembly line in the California plant of a GM-Toyota joint venture. Because the Prizm sells in the used-car market for 15 to 20 percent less than the Corolla,2 the Geo Division of General Motors routinely offered purchase incentives of $1,000 a car, four times more than Toyota’s givebacks.3 Our own research shows that appealing design can generate brand premiums of 5 percent or more—sometimes much more. Assuming, conservatively, that Detroit’s vehicles (at an average price of, say, $22,000) sell at a 5 percent discount as a result of their lower appeal, the annual cost to the Big Three (and their dealers) comes to more than $10 billion, admittedly a sum to be shared between car companies and their dealers.

US carmakers recognize this evolving definition of quality and are changing their product-development processes to meet the new needs of their customers. But Detroit’s track record isn’t reassuring. In the case of one popular US passenger car, for instance, the manufacturer chose to cut costs by removing some sound insulation—saving about $50 a car. While the change had no impact on its safety or durability, interior noise levels rose, so it lost an attribute that is increasingly important to many customers: quietness. Customers noticed and sales fell, only to rise again when the manufacturer restored the insulation.

Measuring a bundle of these qualitative "appeal" attributes is certainly more difficult than measuring defects per thousand during the manufacturing process. What makes customers happy about such intangibles as styling or noise levels is inherently subjective. Nonetheless, the highest priority of US automakers should be making the transition from managing the quality of their products to managing the quality of attributes, because customers are clearly demanding that they do so. The transition requires them to take three steps.

The first is to define the two or three attributes that matter most to consumers and then to guide the product-development process in that direction by reverse-engineering customer satisfaction ratings. Since J. D. Power knows that car buyers want engines that perform well, for instance, it ranks them by looking at a handful of subattributes related to engine performance, such as total power levels, ease of starting, and the level of noise. A car’s performance in any of these subattributes can then be broken down into more specific engineering metrics, such as the number of seconds that pass from the moment the ignition key is turned until the engine starts—and even the angle of the key when it starts the car (some angles are easier for drivers than others). These metrics can be used to guide product design. Typically, US companies don’t follow this approach; they try, perhaps understandably, to make cars that perform reasonably well in 10 to 15 categories, not cars that perform distinctively in those few that make the most difference to customers.

US automakers could design cars for quality from the outset by starting with a clear sense of their performance as defined by a small set of important attri-butes. Nokia, which has used this approach to design mobile telephones, provides a model. Nokia handsets embody two attributes—ergonomics and distinctive styling—that are primary drivers of customer satisfaction. Even if a competing product should be smaller, lighter, and better engineered, Nokia’s focus on attributes is largely responsible for making it the most popular brand of handsets.

The second step is to make the same transition from products to attributes in testing. The Big Three, again understandably, continue to concentrate their testing efforts on the defect rates of product components. Unfortunately, this focus doesn’t increase the likelihood of producing a more appealing car: fewer wrongs don’t make a right. Tests should rate not only components but also the performance of the entire car as measured by the desired attributes, such as quietness. Japanese automakers are particularly effective at testing for the attributes that excite their target customers. The Honda Civic, for example, is tested extensively for three key attributes: fuel efficiency, initial product quality, and durability. The resulting vehicle is more than a collection of defect-free subsystems; it is a car that performs well in the areas that customers have come to expect from a Honda.

Finally, US automakers must develop cross-functional support systems for this new approach to quality, both within the company and beyond. To do so, they will have to increase coordination among the marketing team (which defines the important attributes), product development (which must design a car that satisfies the customer’s requirements), purchasing (which can provide realistic cost-benefit trade-offs when cars are developed), manufacturing (which actually makes the cars), and external suppliers (which are increasingly responsible for delivering preassembled subsystems, not merely raw materials or parts).

But the organizational structure of every US automaker continues to be remarkably compartmentalized; functions struggle with each other, primarily because of misaligned incentives. These companies tend, for example, to judge the success of marketing by total unit sales, of purchasing by costs per vehicle, and of manufacturing by production hours per vehicle. If marketing identifies a new technology that might increase sales but will also cost more and slow down production, purchasing and manufacturing could well fight its adoption. Close internal coordination for attribute management, particularly between marketing and purchasing, has been a major reason for the steady progress of Japanese and European auto manufacturers in the United States.

The management of appealing attributes is no panacea for the broader product-development problems of the Big Three. A great car can never be designed through market research alone, but a focus on attributes can help Detroit see quality through the eyes of its customers and focus more clearly on what is important to them. Combining traditional design innovation with attention to desired attributes is the path to automotive excellence; a focus solely on avoiding defective products is not. For US auto manufacturers, the dynamic of quality has involved a Sisyphean struggle. After making great strides to reduce the number of defects in their cars, they must redouble their efforts to hold their ground in the new auto quality landscape.
About the Authors

Neal Ganguli and T. V. Kumaresh are consultants in McKinsey’s Detroit office, where Aurobind Satpathy is a principal.