Share on FacebookShare on Twitter
Share on LinkedInShare on RedditShare on Google+E-mail






For the gearheads at West Virginia University, it was a minor commission. An environmental group, the International Council on Clean Transportation, had asked the school’s Center for Alternative Fuels, Engines and Emissions to test the tailpipes of diesel cars in the U.S., such as those sold by Volkswagen and BMW. Studies suggested that automakers’ diesel cars polluted more on the road than in the lab, and curiously, more in Europe than in the U.S. ICCT wanted to figure out what the automakers had done to meet America’s tougher emissions standards and how to repatriate these improvements for European car buyers.
The center, founded in 1989, is based on the rolling WVU campus in the foothills of Appalachia and occupies a warehouselike space full of young men in T-shirts and jeans fiddling with jury-rigged equipment, and it smells like a gas station. Run by Daniel Carder, a West Virginia native, it mostly tests heavy-duty engines for trucks and locomotives. Taking sleek little passenger sedans on a road trip was a novelty. One student at the lab, Marc Besch, thought it sounded interesting and asked to work on the project. Besch grew up in Switzerland, and his family ran a dealership that sold mostly Opels, a German competitor to Volkswagen.
There was one problem: Carder, Besch, and their team couldn’t find any diesel cars in West Virginia. So they plotted test routes in California, where lots of consumers have bought into the marketing promise of “clean diesel,” which touts remarkable fuel efficiency without sacrificing muscular acceleration. Out west, they could also use the California Air Resources Board’s dynamometers, the hulking apparatuses used to measure the exhaust coming out of a stationary car.
From February through April 2013, they tested three diesel cars: a Volkswagen Jetta, a Volkswagen Passat, and a BMW sedan. Besch took charge of the initial assessment, using the dynos, and he was impressed. The cars emitted almost nothing, prompting some crowing from Besch about European engineering. When Besch’s fellow investigator, Arvind Thiruvengadam, joined him to take the cars on the road, however, the results were different. The road tests captured a variety of conditions: high elevations up Mt. Baldy; stop-and-go urban errand-running in San Diego; freeway driving around Los Angeles. The two Volkswagens’ emissions exceeded standards by 5 to 35 times. The BMW’s didn’t.
First, they assumed the results were wrong, so the team recalibrated the instruments and kept on driving. Eventually they realized that, yes, they were seeing something noteworthy, if not exactly shocking. What comes out of a tailpipe on the road is always going to differ somewhat from regulatory targets met in tightly controlled environments. Speed, elevation, and temperature all affect the result. To isolate the cause, Besch pored over a two-part paper, published by Volkswagen’s top engineers in 2008, that described purportedly groundbreaking emissions control achieved by the new TDI 2.0 liter engine. And he pursued a theory related to the exhaust filtering system. He still couldn’t explain the full extent of the emissions.
On March 31, 2014, Besch, Carder, and the rest of the WVU researchers presented their findings in San Diego, at the most important conference in the car-exhaust field, the Real World Emissions Workshop. They didn’t name which cars they’d tested, but the technical specs and use of the Lean NOx capture system meant it could only be a Volkswagen.
Sitting in the audience was Jens Borken-Kleefeld, who studies emissions at the International Institute for Applied Systems Analysis in Austria. He’d made the trip for the same reason ICCT had commissioned its study: to understand why the diesel technologies seemed to be cleaner in the U.S. than in Europe. Borken-Kleefeld says the U.S. audience seemed to miss the significance of the findings. “For me, it was a big deal,” he says. “I was expecting to see ‘clean diesel’—and I realized that on the other side of the pond, it’s the same technology and the same underperformance.” U.S. regulators were also in the room, and Borken-Kleefeld recalls approaching some of them. “They shrugged their shoulders,” he says. “I don’t know what they knew at the time—seeing and realizing is different.”
Volkswagen, at least, took note. Someone from VW got in touch with the team to get more details. Carder tried to get Volkswagen interested in a follow-up study with more cars, but it never happened. Early this year, engineers from VW’s Oxnard test facility called to discuss the routes WVU used. (Besch declined to provide names of the engineers.) And that was pretty much it. Which was normal—carmakers regularly learn of emissions problems from researchers and quietly make fixes. “We didn’t hear anything more from them, so we fully assumed that they had made some recalls,” says Carder. “We didn’t even talk to our friends at [California] ARB to follow up on it. That’s really how confident we were that they’d have fixed this, that it was going to be a nonissue.”
What the West Virginia team didn’t realize was that they’d exposed one of the biggest deceptions in the history of the auto industry. On Sept. 3, 2015, VW admitted to U.S. and California regulators that it had not only been selling high-polluting cars but that it had deliberately outfitted them with “defeat devices” that sensed when an official emissions test was under way and temporarily trapped the worst of the exhaust—chemicals that contribute to smog and acid rain.
“Device,” actually, is something of a misnomer. The cheat is merely several lines of software code in the computer that controls a Volkswagen’s engine and exhaust systems. According to the U.S. Environmental Protection Agency, when the car detects a test—certain steering patterns; speed; barometric pressure; only two wheels spinning instead of four—it switches into a cleaner mode called “dyno calibration,” after the testing machines. The cars can run cleaner, but they can’t run cleaner without sacrificing fuel efficiency or some of the engine’s power.
The FBI has opened a criminal probe, and an investigation by the EPA could result in as much as $18 billion in penalties, compounding the immediate loss of tens of billions of dollars in Volkswagen’s market value. On Oct. 8, more than 50 German police and prosecutors raided Volkswagen facilities and employees’ homes at dawn. VW plans to recall 11 million cars and is bracing for class actions from consumers who paid a premium for what they were led to believe was a more efficient and environmentally friendly ride. To raise funds the company will need to pay all the costs associated with the scandal, several industry analysts foresee VW selling off one or more of its 12 brands, such as Bentley, the British racing-turned-luxury sedan manufacturer, which has struggled recently, or Ducati, the Italian motorcycle maker.
Volkswagen’s chief executive, Martin Winterkorn, has resigned. Five managers have been suspended or have gone on leave, including Wolfgang Hatz, the former head of engines and transmissions development at VW. (A VW spokesman declined to comment on any personnel matters.) Winterkorn’s replacement, Matthias Müller, has said a company investigation so far points to a small circle of employees as being involved in the scheme. Its U.S. chief, Michael Horn, told a congressional committee the same thing: “This was a couple of software engineers who put this in for whatever reasons,” he said.

When Horn testified before the House, Representative Michael Burgess, a Texas Republican, told him he couldn’t believe the fraud involved only a few renegades—not at a rule-bound, Teutonic technocracy. “It wasn’t written by one person in their basement in the dark of night,” the congressman said. “You’re going to integrate that into the supply chain of a multinational corporation, and nobody knows a darn thing about it?”
“We’re working full speed on a fast and unsparing clarification of the issue. As soon as we can make reliable statements on this we will do this without undue delay,” wrote VW in a statement to Bloomberg Businessweek. “According to current knowledge we continue to assume that only a few employees participated in the development of this software.” Note that careful phrasing about participation “in the development.” It is possible very few engineers were involved in writing the code. But the diesel engine has been around for more than 100 years. It was invented in Germany. Is it really possible that a German company run by engineers believed the diesel engine had suddenly become clean?


In the mid-2000s the auto industry faced its biggest transformation since the advent of seat belts. High oil prices put fuel economy at a premium, propelling the market for hybrid vehicles and putting a new gloss on diesel cars, which got better mileage than gasoline cars. Diesels famously generate smog, but spew a lot less greenhouse gas. That was another selling point, as fears of climate change helped drive increasingly stringent emissions rules, especially in the U.S. Federal and state regulators, notably in California, began to lay down timelines for new requirements that would demand ingenious feats of engineering.

The pressure sparked innovation. In Japan, Toyota made its half-electric Prius. In the U.S., Elon Musk got famous combining luxury and lithium ion batteries in his Tesla. But at Volkswagen headquarters in Wolfsburg, a bit more than an hour by commuter train west of Berlin, executives were dismissive of newer technologies. “The big run on hybrids remains a limited phenomenon,” then-CEO Winterkorn said in 2007.
VW liked diesel. A niche market in the U.S., diesels accounted for more than half the new-car registrations in the European Union that year. They were cheaper than hybrids and packed more muscle under the hood yet still often got more than 40 miles to the gallon. VW, which aspired to surpass Toyota as the largest carmaker in the world, saw the U.S. as a growth market for a new diesel engine, one that could be branded as green. The trick would be to engineer a way to strip sooty exhaust of its pollutants to meet the new U.S. rules.

In 2005 the company had hired Wolfgang Bernhard to head the VW brand and solve the puzzle of a next-generation, four-cylinder diesel engine that would go in consumer cars like the Jetta. Bernhard broke with existing VW management with two decisions. First, for the construction of the engine itself, he rejected a pump injection model that Winterkorn and others had championed and chose instead a “common rail” injection system that sent precise bursts of fuel into the motor.
Second, to clean up the exhaust and meet emissions standards, Bernhard adopted a competitor’s technology. He chose a Daimler invention called BlueTec that sprayed urea into the exhaust stream to neutralize harmful nitrogen oxides, which can cause asthma and other ailments. To make it work, cars need to be fitted with an extra pump and a tank of what is essentially cat pee. Volkswagen went so far as announcing a deal with Daimler. Another German rival, BMW, also joined in. Then, in November 2006, Volkswagen’s CEO was ousted in a boardroom battle. The new CEO, Winterkorn, got rid of Bernhard, too.
Winterkorn had come up through the Audi business line, and in February 2007 he gave Hatz, the division’s engine development head, the same job for the whole Volkswagen group. A cigar-smoking sports car obsessive, Hatz decorated his offices with race memorabilia, preferring to rhapsodize about the Porsche 911 than talk about business plans or sales. Hatz scrapped the BlueTec deal in a classic case of not-invented-here syndrome, according to an executive involved in the project. That decision, at the start of 2007, boxed Volkswagen engineers in as they tried to meet emissions targets and protect the driving experience and fuel efficiency.
The new team wanted to avoid urea tanks, which were expensive and took up precious space in tiny cars. Instead, they created a system with traps in the tailpipe to catch some of the harmful substances. They also devised a method to scrub away nitrogen oxide by pumping extra fuel into the engine’s cylinders and onward to the exhaust system, where it reacted with other pollutants to form harmless waste. They called it a Lean NOx Trap.


In 2008, VW brought its “clean diesel” to the U.S. with a publicity worthy of Barnum. It landed a Guinness World Record by providing a Jetta to a husband-and-wife team who drove the car through the contiguous 48 states and averaged 58.8 miles to the gallon—unprecedented for any car on that route. At the Los Angeles Auto Show, VW entered its 2009 Jetta TDI Clean Diesel in a Green Car of the Year contest whose jurors included Jay Leno and the executive director of the Sierra Club. The Jetta won, beating out hybrids. At a conference the same year on diesel emissions, in Dearborn, Mich., a VW executive boasted that the new engines had all the environmental benefits of a hybrid car, without any drag on performance. “You don’t have to sacrifice power to be environmentally conscious,” declared one of the slides.
While the idea of power without pollution took off among U.S. consumers, engineers familiar with the technology knew that such claims required magical thinking. “That is the story of Santa Claus,” says Ferdinand Dudenhöffer, director of the Center for Automotive Research at the University of Duisburg-Essen. Without the urea system, “it’s not possible to meet the standard.”
The Lean NOx Trap is a system of concessions. To get cleaner exhaust, you’d need to use a squirt of fuel every few seconds to burn up nitrogen oxide. Or, in the other direction, to get better fuel efficiency, you’d need to spew out dirtier exhaust. Managing this trade-off requires a complex calibration of the onboard computer, the engine control unit, so it can adjust constantly to variables like temperature and speed and optimize both emissions and fuel usage. The trouble was, Volkswagen hadn’t been able to get its new engines to comply with the stringent U.S. standards. In 2007 the company had even delayed for six months the release of the new diesel Jetta that was to be at the forefront of the U.S. push. What could be done to make it pass the tests?

Cheating was hardly a novel option. Defeat devices for emissions testing are as old as the tests themselves. The early ones, in the 1970s, were primitive—basically on/off devices, says Clarence Ditlow, executive director of the Center for Auto Safety, an advocacy group. VW installed sensors on some 1973 models that turned off emissions controls outside a certain temperature range.
In 1978, recognizing the increasing sophistication of emissions controls, the EPA issued a notice warning automakers not to use computer software or other new technology as defeat devices. That didn’t put a stop to anything. In 1995, General Motors agreed to pay $45 million after being accused of installing a chip that turned off pollution controls on 470,000 Cadillacs when the air conditioning came on.
Who exactly is responsible for devising and allowing the VW cheat remains a mystery. One of the clearest explanations so far has come from VW board member Stephan Weil, head of Germany’s Lower Saxony state, which is VW’s second-biggest shareholder. He explained in an Oct. 13 speech to his state parliament how the new diesel engines couldn’t meet the U.S. requirements, and rather than addressing the issue, VW covered up with the software cheat. “In subsequent years there followed a use of this software in other models and other countries,” said Weil. Perhaps the engineers told themselves that the cheat was a stopgap, and they’d address it later. If so, they didn’t.
It’s not credible that top managers were unaware corners had been cut, says Dudenhöffer, who worked at Porsche and other carmakers before entering academia. In contrast to GM, where finance people have run the show for years, and Ford Motor, whose former CEO is a turnaround specialist from another industry, VW is a company where the engineers are in charge. It’s always claimed that an engineer-filled executive suite was a precondition of building top-quality cars. Winterkorn ran around at auto shows with a tape measure and a magnet to examine vehicles from rival carmakers, while back in his own shop he got involved in technical details. When VW managers called for clean diesel without the urea system, “they must have known that it’s impossible, or else it’s not possible they have degrees as engineers,” Dudenhöffer says.
VW’s German headquarters certainly had a grip on pollution numbers. It oversaw the technical aspects of U.S. emissions, according to Horn’s testimony to Congress, with the data passed to an office in Auburn Hills, Mich., for submission to U.S. regulators. If any vehicle failed to meet emissions targets, a team of engineers from Wolfsburg or luxury brand Audi’s base in Ingolstadt was flown in, according to a person familiar with the operations. After the group had tinkered with the vehicle for about a week, the car would then pass the test, according to that person.
The pressure on engineers only grew. As VW boasted about its clean-diesel revolution, the U.S. continued to toughen pollution standards. In 2011, EPA officials announced the second phase of an ambitious environmental plan under which carmakers would have to increase the fuel efficiency of cars and trucks from an average of 35.5 mpg to 54.5 mpg by 2025, while also reducing carbon dioxide emissions further. Companies could win special credits for electric cars, hybrid pickups, and other technology considered groundbreaking—credits that would make the overall fleet standard easier to meet. VW and Mercedes-Benz were frustrated that they weren’t also given credit for their diesel technology.
Margo Oge, a native of Athens who’d watched air pollution erode the Parthenon, started at the EPA in 1980 and was, in 2011, director of its Office of Transportation and Air Quality. In that role, she negotiated directly with automakers over the new rules. Oge found Hatz, then VW’s chief of engine development, outspoken to the point of arrogance in his belief in diesel’s superiority. “They didn’t want to invest in hybrids,” Oge says. “They were 100 percent committed to diesel, and they wanted us to give them credit for bringing diesels into the market.” At the agency, that view was dead on arrival. The EPA wanted the credits to encourage new technology, not reward older methods, no matter how worthy. When Oge said as much in a meeting at the EPA’s Washington headquarters in the summer of 2011, Hatz flushed with anger. Hatz’s VW colleagues later called Oge to apologize.


In retrospect, it seems so obvious. In Park Slope, Brooklyn, Marcella Eckels noticed something was up in 2011 when she replaced her Mini Cooper with a diesel Jetta SportWagen to accommodate the arrival of twins. Eckels is an environmental lawyer, the kind of green consumer VW appealed to. She thought “clean diesel” sounded anachronistic; on the other hand, she wanted the fuel efficiency and liked the way the Jetta felt to drive. “It was winter, and there was all this black soot on the back of the car,” Eckels says. “I thought, ‘That’s weird, but it passed the emissions test, so whatever.’ ”
All U.S. regulators had to do was look across the pond to Europe to see evidence that “clean diesel” did not live up to its name. A 2011 study sponsored by the Joint Research Centre, the European Commission’s in-house scientific research body, found that on-road nitrogen oxide emissions from diesel cars were 90 percent higher than emissions limits. A second study by the JRC, published the next year, found nitrogen oxide emissions 130 percent higher. The authors of both studies called for new testing procedures to supplement the existing oversight, and noted the deterioration of air quality in many cities.
The data from Europe had an unexpected result. Europeans weren’t about to give up their cheap-running diesels for hybrids. But maybe they could make some adjustments based on the presumably cleaner versions pioneered for the U.S. market. Looking to rehabilitate diesel as a more climate-friendly option, ICCT decided to gather evidence in the U.S., where diesel cars met even stricter standards than in Europe. Surely the American models really were running cleaner.
“We thought the vehicles would be clean,” says John German, U.S. co-director of ICCT. “We had no cause for suspicion.” That’s when the organization made its proposal for road tests in West Virginia.
With the West Virginia study as a launching pad, in May 2014 the California watchdog and the EPA opened an investigation into Volkswagen. Talks between the parties went on for months. The company said it had identified the reasons for the higher emissions and proposed a fix. That resulted in a recall of nearly 500,000 U.S. vehicles in December 2014 to implement a software patch.
Most of those cars ran on the first-generation clean-diesel engines introduced in 2008. But the recall also included two subsequent versions that belatedly incorporated the urea system, starting with a Passat in 2012. In theory, the limited introduction of the cleaner urea technology could have solved the problem behind the cheat. Engineers could have quietly backed away from the ruse and hoped nobody would notice, at least for those models. But they didn’t. Instead, they kept the defeat device; there’s money and hassle to be saved by shutting off the urea system, because running dirty means not having to refill the pee tank as often.
More clues piled up. At this year’s Real World Emissions Workshop in Long Beach, Calif., a group of researchers, who’d worked with Borken-Kleefeld, the Austrian scientist, presented data gathered from a far larger sample—millions of cars on the road in Colorado. Emissions from Audis and Volkswagens stood out as much higher than they should have been. VW was paying attention to every turn of the screw. After the presentation, a regulatory specialist from Volkswagen’s Auburn Hills office asked for a copy of the Colorado analysis. The regulators at CARB, meanwhile, continued to test VW cars to see if the recall had fixed the problem. It was concerned that real-world road tests couldn’t confirm that the software patch was working. Sure enough, nitrogen oxide emissions were still in violation of California and U.S. laws. The agency shared those findings with Volkswagen and the EPA on July 8.
At VW’s Oxnard facility, the level of paranoia was rising in July and August, according to a person who worked there. In one instance, a manager accused a low-level employee of being an EPA plant.
At WVU, all was quiet until the Friday morning in September when the scandal broke. An EPA official e-mailed Carder to pass on the news and to say that the WVU lab would get the credit for leading regulators to Volkswagen’s cheat. The fallout so far has been predictable, from the departure of CEO Winterkorn to executives making the rounds of legislative hearings from Rome to Washington to offer their apologies and assurances. If there’s been any surprise, it’s that VW disclosed the defeat device so readily. Or maybe not: For VW’s vaunted engineer-executives, it’s better to admit guilt than stupidity. And in a little-noticed announcement on Oct. 13, Volkswagen said that from now on, “Diesel vehicles will only be equipped with exhaust emissions systems that use the best environmental technology.” Which one? BlueTec, the system whose rejection started it all.

—With Andrew Martin, Alex Webb, and Christoph Rauwald