General Motors' own engineers, along with newspaper auto writers, were talking about the ignition switch defect in several GM (GM) models almost a decade before the carmaker announced plans last month to recall 1.6 million vehicles.
On June 19, 2005, the New York Times (NYT) reported that Chevrolet dealers were telling Cobalt owners to shed items from heavy key rings so they wouldn't bump the ignition into the off position. The reporter wrote that his own wife had knocked a Cobalt's steering column with her knee and found the engine "just went dead."
GM, in a statement at the time, called that scenario rare. The Cleveland Plain Dealer newspaper scorned GM's response as a "knee-slapper." Meanwhile, as early as 2004, two GM engineers involved in the Cobalt said there had been discussions about how the model's engines could cut out when the keys were bumped. In 2005, a Chevy dealer said a customer brought one of the cars back to the dealership, too frightened to drive it.
The almost decade-old discussions of the defects -- those voiced publicly as well as within GM -- are contained within more than 32,000 pages of documents and depositions gathered by Lance Cooper, a Georgia lawyer who argued that a Cobalt engine outage years later, in 2010, resulted in a crash that killed a Georgia pediatric nurse.
Documents from the suit, reviewed by Bloomberg News, are coming to light in the weeks after GM recalled the Cobalt and some other small Chevrolet, Pontiac and Saturn vehicles from the 2003 to 2007 model years,
The documents are fueling curiosity over what GM executives could have known about the defect, and are likely to provide insight into GM's answers as it responds to the National Highway Traffic Safety Administration's request to answer, by April 3, 107 questions about its executives knowledge of the defective parts.
GM quickly provided NHTSA a timeline of what its executives knew about the faulty ignition switches for the Cobalt and Pontiac G5. Wednesday, it provided a supplemental timeline for complaints it investigated on the Saturn Ion and the three other recalled U.S. models. GM's initial NHTSA timeline was consistent with dates in the Georgia lawsuit, including GM's acknowledgment that employees were aware of the situation in 2004. GM settled the suit with Cooper's clients for an undisclosed amount in September 2013.
"It appears that the company did know about this problem. The question is why didn't they do something about it," said Carl Tobias, a law professor at the University of Richmond who studied the recall of Toyotas (TM) tied to sudden acceleration. It will be increasingly difficult for GM to claim its high-level executives were unaware of an issue that appeared in the Times in 2005 and drew a formal media statement, he added.
The Justice Department has also opened a probe into whether GM executives violated criminal or civil laws by failing to notify regulators, said people familiar with the action. U.S. House and Senate committees have said they will investigate.
GM has said it is cooperating with the probes and has also opened its own internal investigation.
GM has said it's sorry about the issue and that it is working to address it quickly. "The chronology shows that the process employed to examine this phenomenon was not as robust as it should have been," GM North America President Alan Batey said in a Feb. 25 statement.
The documents GM filed Wednesday indicate the company got an early indication of the ignition-switch problems as it developed the Saturn Ion in 2001. It thought the problem was fixed. Then, in 2003, an engineer investigating a consumer complaint was able to replicate engine stalls while driving. GM ended up using the same switch in the Cobalt, the G5 and three other U.S. models.
The depositions in Cooper's lawsuit included several passages that were redacted as part of what Cooper's firm said was a protective order entered during the GM proceedings.
Mark Hood, an engineering expert who was hired by Cooper to look at GM documents and reports, said two GM engineers who were deposed had long known about the ignition-switch issue. Gary Altman, who in his own deposition identified himself as the engineering manager for the 2005 Cobalt, and Ray DeGiorgio, who was the design release engineer for the Cobalt ignition switches, were aware of it as early as 2004, Hood said in the deposition.
He testified that Altman experienced the failure when he bumped the ignition with his knee, and said there were meetings and discussion about the flaw prior to the sale of the vehicle in August of 2004. DeGiorgio said in his April 29 deposition that he shut off the ignition of a Cobalt while going about 25 mph in his neighborhood and was able to keep it under control.
Greg Martin, a GM spokesman, declined to comment on the depositions or make any of the executives named in them available for comment.
While a dozen deaths over a decade is noteworthy in any context, about 30,000 people die as a result of auto accidents each year in the U.S., according to NHTSA. Automakers in the U.S. have recalled more than 38 million vehicles through 1,217 recalls in the past two years, according to NHTSA records.
Brian Stouffer, a GM engineer who took over the investigation of the ignition issue in 2011, said in a deposition for Cooper last year that he looked through GM's database and for the 500,000 or so Cobalt models. He said he was able to find about 100 complaints that seemed to meet the criteria. "I have 100 complaints for that, which is a very, very low complaint rate," he said in the deposition.
The first press account cited in Cooper's depositions was a May 26, 2005, review of the 2005 Cobalt appearing in the Daily Item of Sunbury, Pa.
"Unplanned engine shutdowns happened four times during a hard-driving test last week," reviewer Gary Heller wrote. "I never encountered anything like this in 37 years of driving and I hope I never do again."
The next month in the New York Times, auto reviewer Jeff Sabatini described his wife's misadventure.
"During my time with the Cobalt, I encountered the problem once, or rather, my wife did," he wrote in a follow-up article to his review of a Cobalt. "She was driving on a freeway when the car "just went dead,'' in her words. She recalled bumping her knee against the steering column just before the car shut off. She was able to coast to the shoulder of the road, where, once parked, the car started and behaved normally."
Sabatini included a comment from a GM spokesman, Alan Adler.
"In rare cases when a combination of factors is present, a Chevrolet Cobalt driver can cut power to the engine by inadvertently bumping the ignition key to the accessory or off position while the car is running," Adler said in the article. "Service advisers are telling customers they can virtually eliminate this possibility by taking several steps, including removing nonessential material from their key rings."
Writing a week later in the Cleveland Plain Dealer, Chris Jenson quoted from the full GM statement issued following Sabatini's story. "The Cobalt is still controllable" when engine power is cut, the statement read. "The engine can be restarted after shifting to neutral."
Wrote Jensen: "So, if you're whisking along at 65 mph or trying to pull across an intersection and the engine stops, that's what you do. Only a gutless ninny would worry about such a problem. Real men are not afraid of temporary reductions in forward momentum."
Adler declined a request to make additional comments beyond those GM has offered on the recall.
Cooper also deposed Victor Hakim, identified as working in a GM department that studies how vehicles perform in the field. He testified about 90 incidents in which customers from Alabama to New York brought cars that had been stalling into dealerships.
As part of his deposition, Hakim reviewed those statements, including one from a New Jersey service manager who said an owner brought her car to the dealership in April 2005.
"This woman is scared to death of this vehicle," the service manager said in a copy of the report read into the deposition. "She takes care of her grandchildren and she is afraid one day she is going to be riding around with them and kill them."
Hakim said in the deposition that several GM employees had noted issues with the ignition placement in Saturn Ion models, which are also part of the recall.
"On several occasions I inadvertently turned the ignition key off with my knee while driving down the road," a GM employee said in a Feb. 19, 2004, report read as part of the deposition. "For a tall person, the location of the ignition key should be moved to a place that will not inadvertently be switched to the off position."
GM described the issue in a so-called technical service bulletin -- a notice to dealers about a potential vehicle issue and a fix that stops short of a recall -- in December 2005. The automaker said the bulletin, which is described in several depositions and in the timeline GM provided to NHTSA, resulted in repairs for 474 customers. In some of the cases Hakim discussed, customers weren't informed of the potential fix, according to the deposition.
The interviews with GM engineers show the company knew it had a problem and a solution, Cooper said. "They know the system is defective. People under ordinary driving conditions are having keys turn off. But they decide they're going to say it's safe by blaming the problem on the driver," he said.
In 2013, Cooper won a two-year fight to get access to company documents and learned, he said, that the switch had been changed. Engineers he hired to examine ignition switches discovered that starting in 2006, some cars had incorporated a replacement switch. The switch, while bearing the same part number as the old part, had a bigger and stronger spring than the one pulled from the car of the Georgia pediatric nurse, Brooke Melton.
When shown a 2008 ignition switch for a Cobalt that had a design change in a part called a "detent plunger" from what was in the 2005 switch, DeGirogio, the engineer, said no one at GM had approved the change, as far as he knew.
"I was not aware of the detent plunger switch change," he said in his April 29, 2013, deposition. "We certainly did not approve a detent plunger design change."
GM, in its timeline filed to NHTSA, said it had learned in late April 2013 that the design of the part had been changed.