Monday, June 09, 2014
"The Switch From Hell": GM's Barra and the Ignition-Switch Debacle
Back in April, GM CEO Mary Barra testified to Congress about a massive recall involving GM's Cobalts, many of which carried a defective ignition switch that could be accidentally turned off by a driver's heavy keyring. This defect has led to numerous fatal crashes and lawsuits. In a public address to GM employees worldwide on June 5, Barra released a 315-page report detailing the results of an internal investigation into the switch problem, and used some of the harshest language in memory by a CEO of a major corporation about the performance of his or her own employees. Thanks to the release of this report, we now know more about exactly how such a serious defect was allowed in thousands of cars produced in the mid-2000s, and also why it was so dangerous.
The report gives fascinating insight into how a large, bureaucratic operation such as GM goes about developing such a complicated thing as a modern automobile. One way is specialization. Engineer Ray DeGiorgio, for example, joined the firm around 1991 and became an ignition-switch specialist. When a new design was called for around 2000, he eventually became the Design Release Engineer on the project that would make a new design of switch for product lines that eventually became known as Saturn's Ion and Cobalt.
The switch's early development did not go smoothly. Electrical problems plagued both the prototypes that were lab tested and units that made it into test vehicles. As the deadline for releasing the switch to production approached, DeGiorgio came under pressure to make decisions about whether to address a low-torque issue in a design that had already manifested numerous other problems. Because the thing was finally working, mostly, when the switch's supplier, Delphi, asked if they should try fixing the torque issue, DeGiorgio replied "maintain present course." He feared that increasing the torque to meet specifications in all cases would cause further electrical problems, and manifested his frustration by signing the email "Ray (tired of the switch from hell) DeGiorgio."
People tend to be able to deal with only one thing at a time effectively. While DeGiorgio knew that a low-torque switch could cause problems such as stalling while in motion, having a car stall while running once in a while didn't seem as serious as not being able to start the car at all, which was the result of some of the electrical problems. It is not clear that DeGiorgio realized what few other engineers at GM realized either: that when the switch accidentally slipped from Run to Accessory while the car was in motion due to the low-torque problem, this movement would also disable the airbag and the ABS (anti-lock braking) systems. Therefore, not only would a driver suddenly find that he or she had lost power and most of the steering (power steering dies when the engine dies)—the protection from collisions provided by ABS and airbags also disappeared. As subsequent accidents proved, the failure of these safety features was a contributing cause to several of the fatalities that have been ascribed to the ignition-switch problem.
DeGiorgio was not acting maliciously. He was simply trying to do a difficult job in a timely manner. As far as we know, he didn't falsify any records. He was even responsible for fixing the switch-torque problem in 2006, but he did it in a way that violated established GM policy: he made a silent supplier-specification change that did not change the part number or notify the rest of the organization that the new switch was significantly different. As a result, GM and outside investigators spent several years tracking down the mysterious ignition-switch failures, and trying to figure out why they disappeared in models built after the hidden 2006 change.
The theologian and philosopher St. Augustine was among the first to point out that evil is not so much a positive substantial thing in itself, as it is the absence or privation of something good. DeGiorgio's actions and inactions in the GM ignition-switch saga are an example of this. Faced with a laundry list of problems, both electrical and mechanical, he decided in 2002 to let the apparently minor torque issue slide, rather than delay the process further by trying to fix it at the risk of upsetting the delicately balanced design further. And he simply didn't know, or didn't think about, the fact that in the particular vehicles that used the switch, a "moving stall" (as engineers referred to it) could be very dangerous because it disabled the ABS and airbag systems, both of which might be especially needed in negotiating a safe escape from a sudden and surprising loss of power. And when he did a good thing—issuing the 2006 change that fixed the switch—he also, out of either fear or reluctance, failed to issue notifications that would point to the existence of the problem in the first place. Again, it was not so much a positive action on his part that was faulty, but inaction, and inaction possibly motivated by fear.
We have not been told whether DeGiorgio was one of the fifteen people fired as a result of the investigation. Perhaps he made the corporate version of a plea bargain, agreeing to talk in exchange for his job. Whatever his personal fate, his actions have entered the ranks of engineering ethics cases, and will probably stay there for years to come. CEO Barra has done a great service to the discipline of engineering by being as forthright and direct as she has been during the course of this issue. I cannot recall any male CEO who faced a similar problem without blustering, obfuscating, or generally following the advice of lawyers to admit as little as possible. Listening to Barra, one might almost think that she has told her lawyers to go where the switch came from, and not bother her while she says what she wants to say.
CEOs can say anything they want, but making significant changes in an organization of 220,000 people is a hard task. Nevertheless, in airing so much dirty laundry, Barra has made a good start toward fixing the systemic managerial and organizational problems that allowed the defective ignition switches to go unnoticed for so long. She has also presented a good example for other CEOs of major corporations to follow in 'fessing up to ethical lapses.
Sources: Mary Barra's "global town hall" meeting remarks were reported in USA Today on June 5 at http://www.usatoday.com/story/money/cars/2014/06/05/gm-ceo-mary-barra-speech-switch-recall-report/10012715/. I referred to the (redacted for privacy of private individuals) "Report to Board of Directors of General Motors Company Regarding Ignition Switch Recalls" dated May 29, 2014, which is currently available from a news item on the main page of the National Highway and Transportation Safety Board website (www.nhtsa.gov). I also referred to the Wikipedia article on the Ford Pinto. I blogged on Barra's testimony to Congress in this space on April 7, 2014.