GM Folds, Where’s the Fine?

On the First Day, attorney Lance Cooper called GM out on its recall of the 2005-2007 Chevy Cobalt and Pontiac G5 for an ignition switch problem that, the automaker announced, was linked to six deaths.

Cooper had represented the family of 29-year-old Brooke Melton, who died in 2010 when the ignition module of her 2005 Cobalt slipped into the accessory position as she drove along Highway 92 in Paulding County, Ga. Melton’s Cobalt skidded into another vehicle, and Melton died of her injuries in the crash. Cooper, of the Marietta, Ga.-based Cooper Law Firm, revealed that GM knew about the wandering ignition switch problem since before it sold the vehicle – during 2004 production testing. Based on the documents GM turned over during discovery, Cooper petitioned the National Highway Traffic Safety Administration to open a Timeliness Query, charging that GM had delayed the recall for an unconscionable length of time, and neglected to include all affected vehicles in the recall.

“GM continues to they continue to try and spin this story,” says Cooper. “Unfortunately, given the ugly facts it’s not going to work.”

And on the Second Day, the Press said: Let There Be Light Upon this Defect. Jim Healey and Fred Meier of USAToday broke the story that GM’s knowledge actually went back 10 years. Others followed, the New York Times, the Detroit News, the LA Times, etcetera, etcetera, etcetera – and The Safety Record Blog – natch.  

And on the Third, Fourth, Fifth and Sixth Days, the Facts Created a Storm of Sh-t.

It turned out that GM did a recall end-run by issuing Technical Service Bulletins describing this problem in 2005 and 2006.

It turned out that NHTSA had noted the problem the following year in a Special Crash Investigation. In 2007, the Indiana Transportation Research Center probed an October 2006 fatal crash involving an airbag non-deployment in a 2005 Chevrolet Cobalt. The investigation was prompted by questions surrounding the occupant sensing system and the airbag non-deployment, but the investigators quickly zeroed in on the ignition switch problem. The EDR readout showed that the ignition was in the accessory position at the time of the crash and the SCI investigators directly linked this to the 2006 version of the ignition switch TSB, which covered seven affected GM models. In fact, the only real root-cause analysis in the report surrounds the failure of the ignition switch.

And it turned out the GM tried to hide this history by filing a Part 573 Defect and Noncompliance Notice in which the chronology of significant events leading up to the decision to recall was essentially absent – despite acknowledging six deaths in its recall press release.

And on the Seventh Day, GM was “deeply sorry.” Less than a week after the acid drip, drip, drip of negative stories, GM found seven more deaths tied to the defect, and expanded the recall to include all of the affected models – 1.6 million vehicles. And Lo! A miracle – GM filed an amended Part 573, with an actual detailed chronology.

Here are some highlights:

In 2004:  Discovered the problem, but decided not to do anything.

“Engineers believed that low key cylinder torque effort was an issue and considered a number of potential solutions. After consideration of the lead time required, cost, and effectiveness of each of these solutions, the PRTS [Problem Resolution Tracking System inquiry] was closed with no action.”

In 2005: Got complaints, identified a solution, still didn’t do anything.

“GM employees received new field reports of Cobalts losing engine power, including instances in which the key moved out of the “run” position when a driver inadvertently contacted the key or steering column. Further PRTS’ s were opened to re-assess this issue. During the course of a PRTS opened in May 2005, an engineer proposed that GM redesign the key head from a ‘slotted’ to a ‘hole’ configuration. That proposal was initially approved, but later cancelled.”

GM eventually issued the TSBs, and decided that everything was okay, because “the car’s steering and braking systems remained operational even after a loss of engine power, and the car’s engine could be restarted by shifting the car into either neutral or park.”

In 2006: Changed the key hole design. Still getting complaints

In 2007: NHTSA tells GM about the Special Crash Investigation and the two deaths. GM starts tracking crashes of frontal impacts with no airbag deployment – learns of four in which the ignition was in the accessory position at the time of the crash. 

In 2009: GM changes the top of the key from a slot design to a hole design, and decides that the real problem is heavy key chains dragging the ignition out of position. Now GM is up to seven crashes in which the ignition was in the accessory position at the time of the crash. 

In 2010: The Cobalt is no more.

In 2011: GM’s legal department teams up with the Field Performance Assessment  and Product Investigations guys to finally investigate “a group of crashes in which airbags in 2005-2007 model year Chevrolet Cobalts and a 2007 Pontiac G5 had not deployed during frontal impacts.”

In 2012: More investigations – inconclusive results.

In 2013: GM discovered that “ignition switches in early-model Cobalts did not meet GM’s torque specification; changes had been made to the ignition switch’s detent plunger and spring several years after the start of production; and those changes most likely explained the variation from GM’s specifications for torque performance observed in the original switches installed in 2007 and earlier model year vehicles.”

GM parses supplier records and finds that changes had been made to the detent plunger and spring late in the 2006 calendar year, increasing “the switch’s torque performance:”

 Testing and analysis further determined that whether a key moves from the “run” to “accessory” position and how that key movement affects airbag deployment depends on a number of factors, including: vehicle steering inputs and path of travel immediately before key movement; the weight and load on the key ring immediately before key movement; whether the “installed ignition switch meets the torque specifications that GM provided to its supplier; and the timing of the movement of the key out of the “run” position relative to the activation of the airbag’s sensing algorithm of the crash event.

GM decides to recall.

Cooper doubts that this new chronology can put the matter to rest, and he renewed his call for a Timeliness Query. It’s doubtful, for example, that a supplier unilaterally changed a switch design to fix a defect, without someone at GM knowing about it and approving it.

“There’s no doubt that they knew about the 13 fatalities when they announced the first recall,” Cooper adds. “I believe that because of what we found in discovery, and because we filed the TQ, GM was pressured to announce this recall — although I still do not believe they have disclosed the whole story. There will be more [fatal crashes] that come out and there will be those cases that never come to light, because they were never investigated.”

Nonetheless, the company president said in yesterday’s USAToday:

“The chronology shows that the process employed to examine this phenomenon was not as robust as it should have been,” said GM North America President Alan Batey. “Today’s GM is committed to doing business differently and better. We will take an unflinching look at what happened and apply lessons learned here to improve going forward.”

Batey said, “We are deeply sorry and we are working to address this issue as quickly as we can.”

So, will NHTSA flinch in its response?  The agency let GM slide in 2007. The agency let them slide again in 2014, when they allowed GM to issue a partial recall and permitted the automaker to submit a Part 573 with no chronology to speak of. Now it’s time to either pull the trigger on Cooper’s TQ request – this chronology still has holes – like: when did they learn of these 13 deaths? Regulators, pull out your civil penalty calculator, and get adding.

Ten years. Thirteen deaths. “Deeply sorry” doesn’t cut it.