The Pattern Repeats
Bad people do bad things. I don't think anyone could argue with that. But if that's the reason bad things happen, we're helplessly waiting for the next bad person to do the next bad thing.
What if there was a better question?
Idling
This Edition: when finding the problem is too expensive.
You love your job.
One day, upper management gives you a directive. But the more you research and the more you try to find a way to achieve it, the more you realize it’s absolutely impossible.
The issue is, it’s an open secret that telling management something is impossible is the fastest way to find yourself unemployed. So you keep trying, hoping maybe you just missed something.
Suddenly you realize something. If something looks like it performs, that’s almost as good as it actually performing.
You already know how it’s going to be tested. It would be easy to change a few settings so that it passes.
You keep your job today, and if someone finds the discrepancy in a few years, that’s their headache to solve.
Do you tell leadership you managed to get it done?
The Setup
In the mid-2000s, Volkswagen decided to win America over with diesel.
The pitch made sense.
Diesel engines were more efficient than gasoline.
This meant that they had better fuel economy and lower CO2 emissions. European regulators had spent a decade pushing for the shift as a way to combat climate change.
In the US, diesel was still associated with large trucks belching black smoke into the air.
Volkswagen wanted to change that image. Modern diesel was clean, quiet, and efficient.
It was a car for people who still wanted to feel responsible without giving up any performance.
There was a major issue though.
The US regulated nitrogen oxide (NOx) emissions much stricter than Europe did. And, due to how diesel engines work, they create a lot more of it.
To meet the American standards, they would have to add a system into the car that would spray a chemical into the exhaust to neutralize the NOx before it left the tail pipe. It worked but it was also expensive, added weight, took up space, and would require drivers to periodically refill a tank with a chemical they’d never heard of before.
Not exactly the quick, clean, and affordable they’d been promising.
In 2006, they found a solution.
It wasn’t a way to create better emissions, it was software.
Specifically, software that would recognize when the vehicle was being tested and certified for emission standards. Volkswagen's engine control units were built to recognize the precise conditions of a laboratory emissions test: a fixed sequence of speeds, a stationary chassis dynamometer, specific steering wheel behavior, a duration that never varied.
When the software detected those conditions, it activated full emissions controls. The car ran clean, exactly as certified. The moment the car left the lab and drove on an actual road, the software switched to a different calibration, one tuned for performance and fuel economy instead of compliance.
Emissions climbed as high as forty times the legal limit.
For nearly a decade, nobody caught it.
Volkswagen sold roughly 11 million vehicles worldwide fitted with the software, with about 600,000 of them in the United States. They were marketed the entire time as some of the cleanest diesels on the road.
What finally broke the story wasn't a regulator's routine inspection. It was a small nonprofit with a modest grant, a university research team, and a road trip that was never supposed to find anything wrong at all.
The Ban Nobody Enforced
In June of 2007, the European Union adopted a law specifically banning this exact kind of software. They referred to it as a “defeat device”.
While the regulatory agencies took pride in closing the loophole, they never built systems to actually enforce the ban.
The legislation shows they knew the risk. Passing it also felt like they had closed the gap. Meanwhile, Volkswagen spent the year and a half before it went into effect and the 6 years after building vehicles with the very components legislators knew they had banned.
Many EU member states would later admit that they had never inspected a single vehicle for a defeat device since the ban went into effect.
A Story Everyone Needed to Be True
The obvious question becomes “How did regulators know the risk, legislate against it, and still miss its blatant abuse?”.
The answer is simple. The EU had spent a decade claiming diesel was a key component in the fight against global warming. It was low CO2 and more efficient.
When data came back that reinforced that claim, there was no incentive to look deeper.
The bet had already been placed and the winning hand was diesel.
Politicians, regulators, and manufacturers all had incentive to make sure diesel remained the answer.
The convenient answer became the clear answer. If the data looks good, there’s no reason to actively work to try to disprove it or ask what you may be missing.
It didn’t require anyone to be corrupt. They just had to realize that looking for the problem would make them the problem.
When “Can We” Became “How”
Inside Volkswagen, the environment had even less room for questioning the story. When leadership gave a directive, there was no room for failure. Everyone knew it.
Telling leadership that a target couldn’t be achieved wasn’t just information, it was often a career ending admission. They would remove you and find someone who could get it done.
Leadership had set targets for cost and regulators had requirements for emissions.
To hit emissions legitimately, the cost would be too high.
To make the cost, emissions would be too high.
When the question became defy leadership and risk your job or find a backdoor around a regulatory position, it’s likely only one option felt survivable.
The Law: The Proximity Cost
Every failure in this story has the same shape.
A regulator wrote a law and never built a way to enforce it. A government believed data that confirmed a bet it had already made. An engineer heard "impossible" get redefined as career-ending before it could ever reach a decision-maker's desk.
They didn’t need to lie to themselves. They were answering a question nobody had asked out loud: what does it cost me, personally, right now, to find the problem?
For the regulator, it cost credibility in a policy the government had already staked its climate strategy on.
For the politician, it cost the diesel bet the whole continent had made.
For the engineer, it cost a job, a mortgage, a career built inside one company.
In every case, the price of looking closer was immediate, specific, and personal. The price of not looking was distant, diffuse, and paid by someone else, later.
That's the law.
Systems don't need corrupt or blind people to fail. They fail because someone, somewhere, is always the one who has to pay for finding the truth, and the bill for not finding it never seems to have their name on it.
Until it does.
The Collapse
Volkswagen’s strategy worked. They became a poster child of the clean diesel movement. But that status was also their downfall.
In 2013, the International Council on Clean Transportation (ICCT) commissioned a study. In Europe, air quality and diesel pollution was beginning to cause real health problems. ICCT believed it was because the regulatory testing in Europe was much weaker than that done in the US.
The goal was simple.
Prove that, with the stricter US testing standards, manufacturers could produce cars that were actually clean on the road.
They decided to use a VW diesel Jetta and a Passat as the positive examples. They also used a BMW X5 as the control.
In the lab, the cars performed exactly as expected.
The next step was to take the cars on a road test, driving from Los Angeles, California to Seattle, Washington. They attached a Portable Emissions Measurement System (PEMS) to capture actual emissions on the trip.
Once they reviewed the data, it was clear something was wrong. In the road test, the Jetta had hit 15-35 times the legal NOx limit and the Passat showed 5-20 times the limit. Meanwhile, the BMW came back clean.
In spring of 2014, the California Air Resources Board (CARB) began looking into the discrepancy. Then the EPA joined the inquiry.
VW spent over a year offering technical explanations for the discrepancy and even issued a recall before finally admitting to the defeat device in September of 2015, after regulators threatened to withhold certification of the 2016 model year.
That 2006 decision, the one meant to save some cost and a few years of engineering, ultimately cost VW over $30 billion, multiple criminal convictions, and the forced resignation of its CEO.
The math was never actually close
Volkswagen's engineers weren't choosing between cheap and expensive. They were choosing between a cost they could see and a cost they couldn't.
The exhaust system had a price tag attached to it before a single car was built. A few hundred dollars per vehicle, a redesigned tank, a maintenance habit customers would resent. That cost was immediate, specific, and easy to put in a spreadsheet.
The other cost didn't have a number yet. It was a probability, not a price. Maybe nobody notices. Maybe testing never catches it. Maybe by the time it matters, someone else is running the company. A cost with no number attached doesn't compete with a cost that does.
It just gets discounted until it disappears from the decision entirely.
The founder who skips the legal review to hit a launch date.
The leader who lets a culture problem slide because addressing it would cost this quarter's numbers.
The decision made under a deadline that quietly assumes the bill, if one ever comes, will be smaller and further away than the savings in front of you today.
Volkswagen wasn't run by people who thought $30 billion in fines was a good trade for a few hundred dollars a car. Nobody did that math and signed off on it. The device only got built because that math was never done at all.
The upfront cost was the only cost anyone was actually pricing.
