WASHINGTON -- The history of General Motors' ignition-switch fiasco is still fuzzy, but we know this: GM had the chance to replace a fatally flawed switch in the Chevy Cobalt with a part that cost less then $1 and did not do it.
CEO Mary Barra, testifying on Capitol Hill this month, called that "very disturbing." She diagnosed it as the flawed reasoning of Old GM, insisting that never again would such a paltry sum stand in the way of safety.
So Rep. Tim Murphy, R-Pa., the chairman of a U.S. House oversight panel, asked the natural question: Where does the New GM strike the balance between cost and safety?
"We don't," Barra replied. Later, when asked a similar question, she elaborated: "It's not acceptable to have a cost put on a safety issue."
It sounded good. It was just what GM's worried customers would want to hear. The only problem: It was clearly untrue -- and not just of GM but of any major company.
Cost-benefit analysis is a deep-rooted and necessary practice in every industry in which health and safety are at stake.
If automakers truly spared no cost for safety, cars in American showrooms would have nothing but five-star crash test ratings. All cars would have the most advanced safety features available, so brands such as Mercedes-Benz and Volvo would no longer be able to brag about their cutting-edge technology.
And they would all cost a lot more.
Our society has chosen not to go down that path. Every dollar spent on safety makes a car harder to attain for Americans with little money. So we accept a certain amount of danger behind the wheel. That's life.
"There is risk, but that's inherent in every activity that we undertake," says David Strickland, who until January was administrator of the National Highway Traffic Safety Administration. "There's risk in getting in and out of a bathtub every day."
Barra was trying to make a larger point about structural reforms at GM. There are certainly ways to remove incentives to cover up safety problems -- for instance, by having quality gurus make recommendations on recalls without considering costs. But the insistence that GM no longer puts any price on safety smacked more of public relations than reality.
It's hard to blame Barra for that. Cost-benefit analysis is necessary but ugly. It raises deep ethical questions. And in past recall crises, it has often been a cardinal sin in the public imagination.
Barra surely does not want GM to be treated like Ford Motor Co. in the 1970s, when regulators and safety advocates concluded that the Pinto was spectacularly prone to catching fire when struck from behind.
Ford had done itself no favors by lobbying regulators with a memo saying that the lives saved by adopting a stricter standard for fuel tanks did not justify the cost. The cost was estimated at $11 per car and truck, for a total of $137 million a year.
That memo, the magazine Mother Jones wrote in an incendiary 1977 article, "in effect says it is acceptable to kill 180 people and burn another 180 every year, even though we have the technology that could save their lives for $11 a car."
Cost-benefit analysis also came back to bite the Old GM in the 1980s, when safety advocates attacked the "side-saddle" fuel tanks mounted on GM's pickup trucks.
In a memo that remained secret for decades, GM engineer Edward Ivey told regulators that GM should not have to spend more than $2.20 per vehicle to prevent fire deaths. Any greater cost, he wrote, would exceed the benefits of saving 55 lives for every model year of GM vehicles, at a "price" of $200,000 apiece.
"It is really impossible to put a value on human life," he wrote. "This analysis tried to do so in an objective manner but a human fatality is really beyond value, subjectively."
That is the philosophical problem automakers face.
The cost of making cars safer is easy to measure. The value of lives saved is priceless. If a company makes a bad decision and people die, the company is said to be condemning customers to death as statistics on a ledger.
The new CEO had the chance to speak to Congress in shades of gray on behalf of the industry, saying that cost-benefit analysis is appropriate in some cases, and this was not one of them. But that was not the politically correct thing to say. Cost-benefit is wrong in all cases, Barra essentially said. It's a thing of the past.
Well, at least until the next time the auto industry gets caught -- either ignoring safety as with the Cobalt or cluelessly quantifying the value of human life. Both get automakers in trouble.